00

Independent review of the impacts of the bottled water

industry on groundwater resources in the Northern
Rivers region of NSW
Final Report
NSW Chief Scientist & Engineer
31 October 2019
www.chiefscientist.nsw.gov.au/reports/independent-review-of-impacts-of-the-bottled-water-industry-
on-groundwater-resources-in-the-northern-rivers-region-of-nsw
The Hon. Melinda Pavey MP
Minister for Water, Property and Housing
52 Martin Place
SYDNEY NSW 2000

31 October 2019

Dear Minister

Final report: Independent review of the impacts of the bottled water industry on
groundwater resources in the Northern Rivers region of NSW

In November 2018, the Hon Niall Blair MLC, the (then) Minister for Regional Water,
requested that I undertake an independent review of the impacts of the bottled water
industry on groundwater resources in the Northern Rivers region of NSW. I submitted an
initial report from the Review on 1 February 2019. I am pleased to submit this Final Report.
The Initial Report outlined the preliminary analysis of available information and reports on
the bottled water industry, local groundwater systems, the regulatory framework, as well as
the range of stakeholder views gathered through consultations and two site visits to the
region in December 2018 and January 2019.
The Final Report draws on further work undertaken to understand better the potential growth
of the bottled water industry in the region, the sustainability of water extraction limits in the
relevant Water Sharing Plan and the assessment of associated impacts.
I would like to thank all stakeholders who provided contributions to the Review including
local residents, farmers, community groups, the bottled water industry, researchers and
academics, local government councillors and staff, state government agencies and local
water utilities. Their input has greatly informed the analysis of all relevant issues.

Yours sincerely

Professor Hugh Durrant-Whyte

Chief Scientist & Engineer

Chief Scientist & Engineer

GPO Box 5477, Sydney NSW 2001, Australia | Tel +61 2 9338 6786
www.chiefscientist.nsw.gov.au
EXECUTIVE SUMMARY

In November 2018 the (then) Minister for Regional Water, the Hon Niall Blair MLC,
requested that the NSW Chief Scientist & Engineer undertake an independent review of the
impacts of the bottled water industry on groundwater resources in the Northern Rivers region
of NSW.
The Initial Report submitted in February 2019 followed two site visits to the Tweed and
Ballina areas to meet with stakeholders, reported on issues raised in consultations and
described relevant groundwater and surface water systems including available data,
management frameworks and industry allocations.
This Final Report provides an update to the industry allocations and current and proposed
bottled water licences and considers potential growth of the industry, the sustainability of the
extraction limits for the groundwater systems in scope and the impacts that extraction of
groundwater for bottling may have on groundwater resources, surface water and other local
environmental consequences. In doing so, the Review examined available hydrogeological
assessments for the industry, how they are developed and assessed, and how localised
impacts are accounted for and managed. Technical approaches to address issues relating to
truck movements, road integrity and plastics are included.
The Northern Rivers region has alluvial, fractured rock, coastal sands and porous rock
aquifers. Four groundwater sources are relevant to the Review, being the New England Fold
Belt Coast, the Alstonville Basalt Plateau, the North Coast Volcanics and the Clarence
Morton Basin. All are fractured or porous rock aquifer systems and are part of the North
Coast Fractured and Porous Rock Water Sharing Plan.
Overall, fractured rock systems in particular are highly variable and complex, making them
difficult to characterise on a regional scale. In September, an expert workshop was held on
water sharing plans, extraction impacts and current knowledge to inform the technical review
findings.
The Review has occurred during a period of widespread concern and public debate about
extended drought and long-term water futures.
The issues that prompted this Review also go beyond technical and scientific matters. They
encompass fundamental views of community and how resources are valued and allocated.
For some respondents, a major concern is potential lack of water resources for agricultural
purposes. For others, any extractive activity for water bottling purposes is not supported for a
broader range of reasons articulated in the initial Review report. Although beyond the scope
of this report, legislative and policy frameworks include requirements for community
engagement and consideration of social, economic and environmental factors in planning
decisions as well as assessment of risk and scientific knowledge.
Questions of risk and uncertainty and how these are managed are central to many of the
issues raised. Importantly, it became apparent over the course of the Review that there were
different understandings of these concepts. For this reason, this report includes an
explanation about how these concepts were approached by the Review and informed
consideration of sustainability factors and impacts under the Terms of Reference.
For the purposes of this report, an impact refers to the physical change that occurs from an
action, such as depressurisation of groundwater due to its extraction. Consequences are a
result of the impact, for example, temporary or permanent loss of water access or loss of
environment for groundwater dependent ecosystems and associated flora and fauna.
High-level findings and recommendations are presented in this Executive Summary. A
complete reading of the full report provides further detail for the basis of conclusions
reached.

iv
FINDINGS
The bottled water industry
 Available industry data indicates that across Australia, over three-quarters of bottled
water is sourced from underground wells, and the remainder from standard
reticulated water supplies. Approximately 8% of Australian bottled water production is
exported.
 The Review identified seven operators in the Northern Rivers region with allocations
of 240.5 ML/y who are actively extracting for water bottling purposes, representing
0.55% of water licences and basic landholder rights (together defined in the WSP as
‘total water requirements’) and 0.008% of estimated total annual aquifer recharge in
the four groundwater sources.
 Four further proposals, if approved, would amount to an additional 168 ML/y, being
an additional 0.38% of estimated total water requirements and 0.006% of total annual
aquifer recharge.
 Changing consumer preferences, trade imbalances, the availability of tap water and
private (‘no name’) brands and population growth are expected to impact future
bottled water production and consumption volumes.
 Scenario analyses conducted by the Review suggest the Australian bottled water
industry is most likely to grow at a rate of less than 2% per annum to 2024 and that
growth in the Northern Rivers region is likely to be consistent with this trend. Under
most scenarios to 2024 considered, the 168 ML/y of additional proposed bottled
water operations would be sufficient to meet fully projected growth in demand.
 The Review also considered ‘highly unlikely’ and ‘extremely unlikely’ scenarios to
2034, being growth continuing at the current rate of 10% per annum and
establishment of a major premium bottled water exporter in the Northern Rivers,
respectively.
o If the ‘highly unlikely’ scenario occurred, the bottled water industry would
represent less than 2.3% of ‘total water requirements’ and 0.034% of
estimated total annual aquifer recharge.
o If the ‘extremely unlikely’ scenario occurred, the bottled water industry would
represent less than 4.6% of ‘total water requirements’ and 0.069% of
estimated total annual aquifer recharge.
 As the scenario analyses considered an unchanged regulatory and policy
environment, these forecasts may be affected by regulatory intervention which
directly or indirectly impacts the bottled water industry in this region.
 For the purposes of water extraction licensing, the bottled water industry is treated
the same as other prospective commercial users. However, development consent
under the Environmental Planning and Assessment Act 1979 is required for water
bottling activities. Approvals identified by the Review for bottled water extraction in
the Northern Rivers region date from 1993.
Allocations
 The WSP determines the allowable extraction limit, set from the recharge value of
each aquifer, with an amount of the recharge reserved for the environment and the
reminder determining the Upper Extraction Limit or the LTAAEL
 Under the North Coast Fractured and Porous Rock Water Sharing Plan (WSP),
environmental water and basic landholder rights are given priority over licensed

v
 water extraction. Among licensees, priority is given to water utilities and licensed
stock and domestic over commercial licensed purposes.
 At the commencement of the WSP for the four groundwater sources, 100% of
storage is reserved for the conservation of the groundwater system.
 Water available for extraction is a portion of the estimated recharge value for each
groundwater source. This is determined by the WSP. An amount of the recharge is
reserved for the environment. The amount reserved for the environment equates to
97% of the estimated recharge value for New England Fold Belt Coast, 96% for
North Coast Volcanics, 82% for Alstonville Basalt Plateau and 48% for Clarence
Moreton Basin.
 The remaining amounts can be allocated for licensed purposes. Of these amounts,
38.0% of the New England Fold Belt Coast is allocated, 51.3% of the North Coast
Volcanics and 1.7% in the Clarence Moreton Basin. Alstonville is fully allocated.
 These are average values over the groundwater source areas; which means that the
environment is not protected to these levels in locally impacted areas.
Water Sharing Plan assumptions and uncertainty
 In groundwater studies and management, there will always be a level of uncertainty
associated with predictions (e.g. recharge rates) and a precise value may not be
achieved due to the complex and heterogeneous nature of groundwater movement.
This is particularly evident in fractured rock systems that are difficult to characterise
fully.
 The WSP plan was developed based on the best available data at hand and followed
a standard procedure. The assumptions made in the WSP are practical, reasonable
and in agreement with standard practice. In general, the WSP incorporates a
reasonable level of conservatism for extraction limits based on the risks identified.
 The portion of the estimated recharge value available for extraction is a function of
rainfall recharge over low environmental value areas together with an assessment of
environmental and socio-economic risk.
 Calculating recharge is complex due in part to the variability and complexity of the
hydrogeology and limited knowledge of the systems. Based on the analysis, the
Review considers the recharge rates used in the WSP are reasonable and
conservative. This statement is made with a relatively low level of confidence due to
lack of data for the groundwater sources of interest.
 In practical terms the groundwater sources are treated as geologically homogenous
which adds uncertainty and would benefit from further work. The Review recognises
that the complexity of the geology makes it difficult to incorporate heterogeneity into
the WSP recharge calculations. Particular attention should be given to the effects of
geological variability within groundwater sources, and soils and vegetation overlying
aquifer outcrops. The Review acknowledges the conservatism incorporated into the
current WSP through the allowable allocation figures.
 The application of the sustainability index appears to be a cost and time effective risk
tool that is applied as an additional means to protect resources where limited
information is available.
 The WSP incorporates a reasonable level of conservatism for the extraction limits
when the groundwater sources are not fully allocated and where they are fully
allocated at Alstonville, monitoring is applied.

vi
  Additional monitoring in strategic locations in the Tweed would help inform gaps in
knowledge on a regional scale and provide a path towards better conceptual
understanding of aquifer flows.
 The overall system is managed with some level of adaptive management, including
an annual determination of the volume of water per licence share and WSP are
subject to an interim review at five years with a full review at ten years.
 Impacts of climate change should be considered in future WSP methodologies.
Development of Regional Water Strategies, which are incorporating climate change
in calculations, may provide a valuable source of information.
Sustainability of Water Sharing Plan extraction limits
 Due to limited extraction levels (where known allocations in the Tweed region are
much lower than the extraction limits contemplated in the WSP), limited data and
uncertainties described regarding the WSP parameters, it is not possible to conclude
whether the extraction limits are currently sustainable. However, the Review found no
evidence at this point in time that current WSP extraction limits are not sustainable.
 Water levels in Department piezometers should be regularly assessed to ensure
periods of any sustained water level decline are identified early.
 Analysis of the last thirteen years’ data at Alstonville found lagged rainfall an
important variable for understanding water levels. This was observed in shallow-sited
piezometers and in deeper piezometers sited in systems that are well connected to
surface waters and upper aquifers. Observations from deep piezometers showed a
greater stability and a steady upward trend over time of groundwater levels and/or
pressures. In contrast, readings from shallower piezometers showed greater
variability and appear to be recharged regularly with rainfall.
Assessment and management of potential impacts from water extraction
 Based on the review of available information, there is no measured evidence that
current bottled water extractions have impacts on other properties’ bores, surface
water or GDEs in the Northern Rivers region. This is at least partly due to the
relatively low current levels of extractions, hydrogeological conditions and lack of
monitoring detecting these impacts.
 While all groundwater extractions have impacts, the magnitude of those impacts and
potential consequences will vary. Assessment of risks and measurement of local
impacts is complex due to the spatial and temporal variability of the hydrogeology of
fractured and porous rock systems. There are established approaches to measuring
and modelling to better understand local impacts. All have challenges in terms of
accuracy, practicability and cost.
 Bore water extraction can potentially impact water within the same aquifer, within a
connected aquifer, or within a connected surface water body, leading to possible
changes in water quantity and quality. The pump test is a common field technique,
used in hydrogeological assessments, to derive local scale aquifer properties and to
indicate proposed impacts of the extraction. In fractured rock systems, the fracture
network that intersects the point of extraction will determine the response to
pumping, which is complex and requires hydrogeological investigations and
interpretation of results in order to design the pump test. Impacts may be proximate
to or at distance from the point of extraction, and occur vertically as well as
horizontally.
 Noting the low level of current groundwater monitoring in three of the four relevant
groundwater sources, there would be merit in reviewing the need for additional

vii
 monitoring that will provide the baseline data, conceptual hydrogeological models
and recharge estimates commensurate with potential future risk levels.
 At a regional scale, the cost of traditional monitoring bore infrastructure is likely to be
an ongoing challenge. This is particularly the case in fractured rock systems subject
to high hydrogeological variability. Emerging sensing technologies able to gather
data over large areas and at depth may provide a step-change to the field, subject to
cost and commercial availability. Whether at the local or regional scale the choice of
monitoring will be informed by the level of risk and the cost-effectiveness of the
monitoring.
 Local scale monitoring during extraction operations can assist with better
understanding of local hydrology and extractive impacts and consequences. This
may include piezometric monitoring of the pathway between the point of extraction
and locations where there is perceived risk. The cost of this monitoring is likely to be
a challenge and its requirement should be justified by the risks as identified by an
expert following analysis of pump test data.
 Local scale monitoring during extraction operations could potentially support adaptive
management, for example, through additional reporting and cease-to-pump rules
related to observed groundwater pressures.
 An assessment of hydrological reports submitted to support development
applications by bottled water proponents undertaken by the Review indicates both
industry and decision makers would substantially benefit from greater clarity,
specificity and standardisation of requirements for hydrological reports. Current
technology is available to enable standardised templates and reports to be managed
electronically.
 Robust local assessment of potential connectivity between aquifer and overlying
shallow groundwater and surface water should form part of pump tests and feature in
hydrogeological reports. This is important, as observed in Alstonville, where deeper
aquifers are not necessarily confined and may have connections to surface systems
or shallower aquifers. It is important to increase understanding of how confined the
aquifer is, as assessment criteria of allowable drawdown differs between confined
and unconfined systems. In addition, field verification is an important part of the
process.
 The Review received anecdotal information suggesting bottled water extractors were
generally extracting water at an approximately evenly spaced production rate year-
round compared with other commercial users who extract on a more periodic basis.
The Review was not able to verify these observations. Further, all groundwater users
are subject to future changing environmental conditions, which may influence their
future patterns of use. The implementation of the NSW Non-Urban Water Metering
Policy will provide information about use patterns in the bottled water industry and
enable analyses of interactions and impacts.
 The Review received consistent reports from the community and sometimes
neighbours of bottled water extractors about observed changes including
environmental effects of drying watercourses and loss of water from previously
productive bores. The Review has not identified scientific studies or other evidence
establishing a causal link between these observed effects and extraction specifically
undertaken by the bottled water industry. Going forward, data from extraction bores,
together with monitoring bores (piezometers), local studies and other sources of
information should help improve knowledge of impacts from a range of sources.

viii
Data
 Lack of extraction data is an impediment to establishing appropriate extraction limits
for individual bores, measuring impacts, and at a regional scale, development of
WSP and making determinations of available water. A state-wide metering policy for
qualifying groundwater works with bore diameters of 200mm and above will take
effect in the Northern Rivers region from 2023. Four of the bottled water extractors in
the region are currently required by the regulator to have meters installed.
 The accessibility of any data is central and manual collection can be an impediment
in this regard. Advances in technology to provide robust and tamper-proof
telemetering options that are commercially cost competitive would have a significant
impact.
 Making water extraction and monitoring data available in standardised formats
through open databases would benefit decision-makers, researchers and the general
public to understand better activities and impacts, including cumulative impacts at
local and regional scale. Approvals by relevant state and local government
authorities could include requirements that all hydrogeological data are published.
There are state managed environmental databases (e.g. SEED) that could be
utilised.
Decision-making
 As with any environmental, engineering, resource activity, the proponents and
decision makers and regulators operate in a realm of imperfect information. This
leads to levels of uncertainty around data and information; however, uncertainty need
not prevent decisions being made.
 There are a number of approaches and tools employed to reduce uncertainty with
regard to the assumptions, hydrological domain, impacts, and consequences of
water extraction. These include taking conservative estimates, using multiple lines of
analysis, being judicious in decisions around the type and location of monitoring,
employing adaptive management approaches.
 There is a lack of clarity around water planning, management and decision-making
roles and processes at state and local government level and between relevant
authorities.
 State government agencies and local government should work to clarify roles and
responsibilities to streamline assessment and approval processes, to avoid
duplication of effort, and to address any gaps in the assessment and approvals
process.
 If local government is to undertake hydrogeological assessment as part of the
development application process, then it needs access to relevant expertise to
interpret modelling and technical reports to inform its decision-making, including
requirements for development applications.
 Access to government and industry water data through a common open platform
housing standardised, well-curated and long-term data sets that can be expanded
would assist assessment and decision-making of applications.
 Regional Water Strategies will be developed over the coming months for the 12
catchment regions across the state and will assist to manage the regions’ water
resources. The Greater Hunter Regional Water Strategy is already in place. These
will improve water security within each region and influence decisions about
infrastructure, water reuse, water sharing including during droughts, protect the
regions’ environmental assets as well as addressing community and industry needs.

ix
Truck movements and road impacts
 There are technologies available that can provide accurate, consistent and real-time
data on truck movements, which could be included as a condition of the development
consent.
 Responsibility for governing truck safety, movements and size spans Federal, State
and Local Government authorities. Each of the responsible bodies has measures to
regulate and monitor heavy vehicles through existing legalisation, approval of
applications and technologies.
 Technologies and strategies are available to measure traffic volumes and impacts.
Local government can levy heavy vehicle road users to contribute to the cost of road
maintenance and repair.
Plastics
 The presence and management of plastics is international in scope and management
of the impacts and solutions will be influenced significantly by factors and
developments beyond those extracting water for bottling purposes in the Northern
Rivers region.
 The NSW Government is developing a 20-year waste strategy and plastics plan in
the context of broader Federal Government and inter-jurisdictional commitments to
address waste and transition from linear to circular economies.
 There is a NSW Government container deposit scheme, which has resulted in a one-
third reduction across the state of eligible containers, including bottles entering the
litter stream.
 Research and development efforts to replace, repurpose and recycle plastics is a
fast-moving and evolving space that is predicted to show significant growth within the
next five to ten years.

RECOMMENDATIONS
1. Further work is undertaken to incorporate climate change into the development of
recharge estimates for the Water Sharing Plan.
2. Consideration should be given to incorporate geological heterogeneity and soil and
vegetation types into recharge estimates where practicable. This may be dependent
in part on technological advances, including remote sensing, to characterise
systems.
3. Improved monitoring of piezometric water levels is needed in locations with a
perceived risk and/or lack of knowledge of groundwater responses and flow
directions. This could provide baseline data, conceptual hydrogeological models and
recharge estimates commensurate with potential future risk levels. Additional
investments in monitoring should balance the value of expected improvements in
data availability and data quality against the resources required.
4. Robust local hydrogeological assessments of aquifer connectivity with overlying
shallow groundwater and surface water should be investigated via well-designed
pump tests. This information should feature in hydrogeological reports.
5. Work should continue towards developing practical and comprehensive guidance on
the contents of hydrogeology reports to be submitted by proponents, including
specificity and standardisation of information provided and reporting requirements.
Ideally, these would be able to be lodged electronically and made publically
available.

x
6. State government agencies and local government should work to clarify roles and
responsibilities to streamline assessment and approval processes, to avoid
duplication of effort, and to address any gaps in the assessment and approvals
process. The first step for this would be by February 2020, relevant officers from
Water NSW, DPIE Water, NRAR and Tweed Council convene a workshop for
Northern Rivers region bottled water to discuss and develop an approach between
them.
7. Water extraction and monitoring data should be made available in standardised
formats through open and accessible portals. State managed databases and portals
(e.g. SEED) should be utilised where relevant.

xi
Contents
Executive summary ........................................................................................................... iv
FINDINGS .............................................................................................................................. V
RECOMMENDATIONS ............................................................................................................... X
CONTENTS ............................................................................................................................XII
TABLES ............................................................................................................................ XV
FIGURES ........................................................................................................................... XVI

1 Introduction ....................................................................................................... 18
1.1 BACKGROUND ....................................................................................................... 18
1.1.1 Proposal from Tweed Shire Council to prohibit water bottling facilities ............................. 20
1.2 REVIEW PROCESS ................................................................................................. 20
1.2.1 Updated Terms of Reference .......................................................................................... 20
1.2.2 Experts ........................................................................................................................... 20
1.2.3 Site visits and consultations ............................................................................................ 21
1.2.4 Submissions ................................................................................................................... 21
1.2.5 Workshop ....................................................................................................................... 21
1.3 STRUCTURE OF THIS REPORT ................................................................................. 21

2 The bottled water industry ................................................................................ 22

2.1 BOTTLED WATER INDUSTRY.................................................................................... 22
2.1.1 The bottled water industry supply chain........................................................................... 22
2.1.2 Performance of the bottled water industry in Australia ..................................................... 24
2.1.3 Outlook for the bottled water industry in Australia ............................................................ 24
2.1.4 The bottled water industry in the Northern Rivers region of NSW ..................................... 27
2.1.5 Outlook for the bottled water industry in the Northern Rivers region................................. 29
2.1.6 Policy and regulatory interventions .................................................................................. 32
2.2 W ATER ENTITLEMENTS FOR BOTTLED WATER FACILITIES IN THE NORTHERN RIVERS .. 33
2.3 REGULATION AND CHARACTERISTICS OF EXTRACTION.............................................. 36
2.3.1 Australasian Bottled Water Institute Model Code ............................................................. 36
2.3.2 Characteristics of bottled water extractors ....................................................................... 37
2.4 CONCLUSIONS ...................................................................................................... 38

3 Sustainability of extraction limits in the water sharing plan .......................... 40

3.1 EXTRACTION LIMITS UNDER THE WSP.................................................................... 40
3.1.1 WSPs as regional water management tools .................................................................... 41
3.1.2 Prioritisation of water under WSPs .................................................................................. 42
3.1.3 Regional sustainability and the WSP ............................................................................... 42
3.1.4 Local impacts and the WSP ............................................................................................ 43
3.1.5 Adaptive management and the WSP............................................................................... 44
3.2 SOURCES OF UNCERTAINTIES AND THEIR MANAGEMENT IN WSP .............................. 45
3.2.1 Average annual rainfall ................................................................................................... 46
3.2.2 Surface area and impact of confining geological layers ................................................... 47
3.2.3 Recharge Rate ............................................................................................................... 48
3.2.4 Sustainability Index ......................................................................................................... 49
3.3 TESTING THE WSP RECHARGE RATES AGAINST STUDIES AND OTHER APPROACHES ... 49
3.3.1 Relevant literature on recharge rates .............................................................................. 49
3.3.2 Application of methods to local area ................................................................................ 55
3.3.3 Impact of reducing the recharge rates ............................................................................. 59
3.4 MONITORING DATA FROM ALSTONVILLE BASALT ...................................................... 60
3.4.1 Alstonville monitoring network and aquifer levels to 2006 and 2009 ................................. 60
3.4.2 Alstonville monitoring data between 2006 – 2018/2019 ................................................... 60
3.5 CONCLUSIONS ...................................................................................................... 70

4 Extraction impacts: understanding, assessment and management .............. 73

xii
4.1 EXTRACTION IMPACTS AND CONSEQUENCES ........................................................... 73
4.1.1 Impact mechanisms ........................................................................................................ 73
4.1.2 Extraction consequences ................................................................................................ 75
4.2 COMPLEXITIES IN MEASURING LOCAL SCALE IMPACTS FROM BORE WATER EXTRACTION .
............................................................................................................................ 75
4.2.1 Variability of fractured and porous rock systems .............................................................. 75
4.3 EXISTING TECHNICAL APPROACHES TO MEASURING LOCAL SCALE IMPACTS ............... 77
4.3.1 Groundwater pressure monitoring ................................................................................... 77
4.3.2 Chemical and temperature tracers of water flow .............................................................. 77
4.3.3 Numerical modelling ....................................................................................................... 78
4.3.4 Surface water monitoring ................................................................................................ 78
4.3.5 Groundwater dependent ecosystems (GDE) monitoring .................................................. 78
4.4 ASSESSMENT OF POTENTIAL IMPACTS..................................................................... 79
4.4.1 Applications that require hydrogeological assessment ..................................................... 79
4.4.2 Process for applications requiring hydrogeological assessment ....................................... 79
4.4.3 Applications must satisfy minimum set of conditions ........................................................ 80
4.4.4 Applications are subject to a risk assessment.................................................................. 81
4.4.5 Applications may be subject to a hydrogeological assessment ........................................ 81
4.4.6 Relevant impacts considered in the hydrogeological assessment .................................... 82
4.4.7 Defining the level of acceptable impacts.......................................................................... 84
4.5 MANAGEMENT OF IMPACTS AFTER APPROVAL .......................................................... 85
4.5.1 Conditions imposed on approvals (WALs and works approvals) ...................................... 85
4.5.2 Conditions of development consents ............................................................................... 85
4.5.3 ‘Water allocations’ or ‘available groundwater determinations’ .......................................... 86
4.5.4 Temporary water restrictions ........................................................................................... 87
4.5.5 Collaborative data and approach..................................................................................... 87
4.6 OBSERVATIONS ON PAST HYDROGEOLOGICAL REPORTS .......................................... 88
4.6.1 Review of the available hydrogeological reports for the industry ...................................... 88
4.6.2 Public responses to the hydrogeology reports ................................................................. 93
4.7 FURTHER INFORMATION TO UNDERSTAND SYSTEM IMPACTS FROM BORE WATER
EXTRACTIONS ....................................................................................................... 93
4.7.1 Investment in monitoring and technological advances ..................................................... 94
4.7.2 Metering data.................................................................................................................. 94
4.7.3 Improving clarity on NSW government’s expectations of hydrogeological assessment
reports ............................................................................................................................ 94
4.7.4 Additional requirements concerning shallow groundwater drawdowns ............................. 95
4.7.5 Field verification.............................................................................................................. 96
4.7.6 Improving data collection, accessibility and management ................................................ 96
4.7.7 The potential role of local research studies...................................................................... 96
4.8 CONCLUSIONS ...................................................................................................... 97

5 Technical approaches to socio-economic factors ........................................ 100

5.1 TRANSPORT (TRUCKS AND ROADS) ....................................................................... 100
5.1.1 Frequency and hours of operation................................................................................. 100
5.1.2 Truck noise ................................................................................................................... 102
5.1.3 Truck size and road damage ......................................................................................... 103
5.1.4 Conclusions .................................................................................................................. 104
5.2 PLASTIC BOTTLES................................................................................................ 105
5.2.1 Extent and management of plastic ................................................................................ 105
5.2.2 Bottled water industry – current practices and commitments.......................................... 106
5.2.3 Opportunities and challenges ........................................................................................ 107
5.2.4 Conclusions .................................................................................................................. 109

6 Findings and recommendations ..................................................................... 111

6.1 FINDINGS ............................................................................................................ 111
6.1.1 The bottled water industry ............................................................................................. 111
6.1.2 Allocations .................................................................................................................... 112
6.1.3 Water Sharing Plan assumptions and uncertainty.......................................................... 112

xiii
6.1.4 Sustainability of Water Sharing Plan extraction limits..................................................... 114
6.1.5 Methodological improvements....................................................................................... 114
6.1.6 Assessment and management of potential impacts from water extraction ...................... 115
6.1.7 Data ............................................................................................................................. 117
6.1.8 Decision-making ........................................................................................................... 117
6.1.9 Truck movements and road impacts.............................................................................. 118
6.1.10 Plastics ......................................................................................................................... 118
RECOMMENDATIONS ........................................................................................................... 119

References ...................................................................................................................... 120

Appendix 1: Terms of Reference (Updated Feb 2019) .................................................. 128

Appendix 2: Site visits, consultations, workshop and submissions ........................... 129

Appendix 3: Excerpts from Initial Report – extraction limits and the Water Sharing
Plan .................................................................................................................................. 132
W ATER SHARING PLAN FOR NORTH COAST FRACTURED AND POROUS ROCK GROUNDWATER
SOURCES .......................................................................................................................... 132
EXTRACTION LIMITS ............................................................................................................ 135

Appendix 4: Baseflow filtering technique ..................................................................... 141

Appendix 5: Modelling of Alstonville Piezometric Data ............................................... 153

Appendix 6: Case study - Lumley Park and Convery’s Lane, Alstonville ................... 180

Appendix 7: Setback rules from the WSP ..................................................................... 184

Appendix 8: Decision making and uncertainty ............................................................. 186

ACRONYMS ..................................................................................................................... 192

xiv
Tables
Table 1: Product segments in the bottled water industry ..................................................... 22
Table 2: Trends influencing bottled water production .......................................................... 25
Table 3: Water licences associated with existing and proposed water bottling .................... 27
Table 4: Existing and proposed bottled water licence entitlements as a proportion of total
water requirements (including licenced extraction and BLR) by groundwater source .......... 33
Table 5: Available water, extraction limits and requirements by purpose and groundwater
source (updated October 2019) .......................................................................................... 35
Table 6: Prioritisation of water users under WSPs .............................................................. 42
Table 7: Recharge rates used in WSP for groundwater sources with bottled water extraction
........................................................................................................................................... 48
Table 8: Groundwater sources, recharge rates used in WSP and recharge estimated locally
........................................................................................................................................... 56
Table 9: Flow gauges in the Tweed Area and Alstonville Plateau........................................ 58
Table 10: Fraction Baseflow/Rainfall for flow gauges for period 1960-2018 ........................ 58
Table 11: Hypothetical example scenario - reducing recharge to 20% of the estimated
recharge in the WSP ........................................................................................................... 59
Table 12: DPIE groundwater monitoring piezometer network at Alstonville ......................... 62
Table 13: Types of applications under the Water Management Act 2000 that may require
hydrogeological assessment ............................................................................................... 79
Table 14: Acceptable level of impacts for porous and fractured rock groundwater sources . 84
Table 15: Contents of five bottled water industry hydrogeological reports from the Tweed and
Ballina Shire compared against standardised table of contents in the coastal guidelines .... 90
Table 16: Summary of conditions of approval for truck movements ................................... 101
Table 17: Summary of regulation relating to truck noise .................................................... 103
Table 18: Summary of conditions of approval for truck size............................................... 103
Table 19: Site visit consultations ....................................................................................... 129
Table 20: Consultations .................................................................................................... 129
Table 21: Expert Workshop Participants – 6 September 2019 .......................................... 131
Table 22: Submissions ...................................................................................................... 131
Table 23: Northern Rivers region groundwater and surface water sharing plans ............... 133
Table 24: Groundwater sources and descriptions ............................................................. 133
Table 25: Recharge rates recommended by DPI Water (2015) ......................................... 136
Table 26: Rainfall recharge rates adopted in the Water Sharing Plan ............................... 137
Table 27: Sustainability index matrix, with an example calculation of a high aquifer, medium
socio-economic risk sustainability index of 25% ................................................................ 137
Table 28: Sustainability index for relevant groundwater sources ....................................... 138
Table 29: LTAAEL in fractured rock aquifers as reported in February 2019 (Initial Report) 139
Table 30: LTAAEL for porous rock aquifers as reported in February 2019 (Initial Report) . 139
Table 31: Recharge amount reserved for the environment as reported in February 2019
(Initial Report) ................................................................................................................... 140
Table 32: Details for TWS bores, the allocation, depth of the bore, and nearby DPIE
groundwater monitoring piezometers ................................................................................ 180
Table 33: Minimum distance rules to minimise interference between bores in fractured rock
groundwater sources (Alstonville Basalt Plateau, New England Fold Belt Coast, and North
Coast Volcanics) ............................................................................................................... 184
Table 34: Minimum distance rules to minimise interference between bores in porous rock
groundwater sources (Clarence Morton Basin) ................................................................. 184
Table 35: Minimum distance rules to minimise contamination ........................................... 184
Table 36: Minimum distance rules to minimise impacts on GDEs and environmentally
sensitive areas .................................................................................................................. 185
Table 37: Minimum distance rules to minimise impacts on groundwater-dependent culturally
significant sites.................................................................................................................. 185
Table 38: Acronyms .......................................................................................................... 192

xv
Figures
Figure 1: Map of Northern Rivers region showing boundaries of the four groundwater
systems............................................................................................................................... 19
Figure 2: High level supply chain in the bottled water industry ............................................ 23
Figure 3: Bottled water consumption in Australia ................................................................. 24
Figure 4: Consumption of bottled water in 2017 in developed economies ........................... 26
Figure 5: Water requirements in the Northern Rivers region by groundwater source ........... 28
Figure 6: Water requirements, unassigned water, and groundwater recharge reserved for the
environment in the Northern Rivers region .......................................................................... 29
Figure 7: Bottled water industry growth scenarios 1 (probable) and 2 (unlikely) .................. 30
Figure 8: Bottled water industry growth scenarios 3 and 4 .................................................. 32
Figure 9: Breakdown of total estimated annual aquifer recharges in the Northern Rivers
region by water requirements, unassigned water, groundwater recharges reserved for the
environment and LTAAELs ................................................................................................. 34
Figure 10: Recharge areas of confined and unconfined aquifers ......................................... 48
Figure 11: Mean recharge estimates for the Clarence Moreton bioregion ........................... 51
Figure 12: Data used for estimating recharge for the Clarence Moreton Bioregional
assessments ....................................................................................................................... 52
Figure 13: Relations between long-term average rainfall and chloride mass balance
estimates of recharge for the hydrogeological units of the Clarence Moreton bioregion - (a)
the Cenozoic Volcanics (Lamington Volcanics and Main Range Volcanics), (b) Walloon Coal
Measures, (c) other sedimentary rock formations. ............................................................... 53
Figure 14: Screenshot of AWRA-L deep drainage results, showing map for typical year 2018
........................................................................................................................................... 55
Figure 15: Flow monitoring points in the Northern Rivers region and groundwater
management areas ............................................................................................................. 57
Figure 16: Map of Alstonville showing the 13 sites for the Alstonville monitoring piezometer
network ............................................................................................................................... 61
Figure 17: Hydrographs showing the levels of 12 ‘shallower’ piezometers in the Alstonville
Plateau from 2006 to present. ............................................................................................. 66
Figure 18: Hydrographs showing the levels of 16 ‘deeper’ piezometers in the Alstonville
Plateau from January 2006 to present. ............................................................................... 67
Figure 19: Hydrograph for Piezometer GW041008, Hole 1, Pipes 1 and 2 with 120 day
rainfall average plotted in the lower panel. .......................................................................... 68
Figure 20: Panels a, b and c showing, respectively, the estimated rain effect, the overall time
effect and the seasonal effect estimated from fitting a GAM model ..................................... 69
Figure 21: Area of influence and cone of depression in the aquifer due to groundwater
pumping .............................................................................................................................. 73
Figure 22: Process for assessing applications for new water supply works approvals and
water licence dealings ......................................................................................................... 80
Figure 23: Return and Earn collection volumes of sorted PET plastics in select areas ...... 106
Figure 24: Allotment of estimated recharge to Recharge Amount Reserved for the
Environment (total volumes differ between aquifers) as reported in February 2019 (Initial
Report).............................................................................................................................. 140
Figure 25: Baseflow filtering results - monitoring point 201001: Oxley River at Eungella ... 142
Figure 26: Baseflow filtering results monitoring point 201005: Rous River at Boat Harbour
No. 3 ................................................................................................................................. 143
Figure 27: Baseflow filtering results monitoring point 201012: Cobaki creek at Cobaki ..... 144
Figure 28: Baseflow filtering results - monitoring point 201015: Tweed River D/S Palmers
Road Crossing .................................................................................................................. 145
Figure 29: Baseflow filtering results - monitoring point 201900: Tweed River at Uki .......... 146

xvi
Figure 30: Baseflow filtering results - monitoring point 202001: Brunswick river at Durrumbul
(sherrys crossing) ............................................................................................................. 147
Figure 31: Baseflow filtering results - monitoring point 202002: Burringbar creek at
Burringbar ......................................................................................................................... 148
Figure 32: Baseflow filtering results - monitoring point 203012: Byron Creek at Binna Burra
......................................................................................................................................... 149
Figure 33: Baseflow filtering results - monitoring point 203014: Wilsons River at Eltham .. 150
Figure 34: Baseflow filtering results - monitoring point 203057: Houghlahans Creek at
upstream Teven ................................................................................................................ 151
Figure 35: Baseflow filtering results - monitoring point 203059: Maron Creek at Graham road
......................................................................................................................................... 152
Figure 36: Convery's Lane TWS bore usage (ML/y) with GW036702 (deeper and shallow)
levels and rainfall data. ..................................................................................................... 181
Figure 37: Lumley Park TWS bore usage (black columns) and water level and/or pressure
observed at DPIE monitoring bores GW081006 (shallow) and GW081005 (deeper) from
2002 to 2009. .................................................................................................................... 182
Figure 38: Groundwater levels from DPIE monitoring bores near Lumley Park plotted with
rainfall ............................................................................................................................... 183

xvii
1 INTRODUCTION

In November 2018, the (then) Minister for Primary Industries, Minister for Regional Water
and Minister for Trade and Industry requested the NSW Chief Scientist and Engineer to
conduct an independent review and provide expert advice on the impacts on groundwater
quantity arising from extraction by the bottled water industry in the Northern Rivers region of
NSW.
In February 2019 an Initial Report was submitted that focused on the local bottled water
industry identified in the first phase of work, the geology and hydrogeology of the region
and local groundwater systems, the regulatory framework in which activities are undertaken
and issues raised in submissions and during consultations and site visits.
This second and final report addresses the sustainability of extraction limits and the impacts
and consequences of groundwater extraction. This includes a review of how extraction
limits are assessed at the macro level through the Water Sharing Plan (WSP) and locally
through water access licences (WAL) and development applications; an assessment of
knowledge and data gaps; and sources of uncertainty and how these are managed. An
update of the entitlements of the local bottled water industry is provided having regard to
total access rights, as is comment on factors influencing demand trends and growth
scenarios considered by the Review. Technological approaches to managing issues of
truck movements associated with the industry and plastics are provided.

1.1 BACKGROUND
The North Coast Fractured and Porous Rock Groundwater Sources Water Sharing Plan
(the WSP) sets out extraction limits and rules for all four groundwater sources in scope
within the Northern Rivers region:
 Alstonville Basalt Plateau Groundwater Source – fractured rock aquifer
 Clarence Moreton Basin Groundwater Source – porous rock aquifer
 New England Fold Belt Coast Groundwater Source – fractured rock aquifer
 North Coast Volcanics Groundwater Source – fractured rock aquifer (Figure 1)
The Alstonville source stands in contrast to the three sources further north in the Tweed
valley. The Alstonville source is the only fully allocated groundwater system of the four
groundwater sources; it has higher use, and it has a network of state operated monitoring
bores or piezometers. This Report analyses the monitoring data from these bores from
2006 to present.
Overall use of the other three groundwater systems, in the Tweed, is very low compared
with the size of the aquifers, but monitoring data on these systems is also sparse.
For all four aquifers, the parameters used to determine the extraction limits are considered
by the Review and focus is also put on local impacts and protection of features.
.

18


Figure 1: Map of Northern Rivers region showing boundaries of the four groundwater systems

19
1.1.1 Proposal from Tweed Shire Council to prohibit water bottling
facilities
In November 2018, the Tweed Shire Council resolved to prepare a planning proposal to
remove clause 7.15 from the 2014 Tweed Local Environment Plan (LEP) with the intended
outcome that a water bottling facility will no longer be permitted in RU2 Rural Landscape
Zone (TSC, 2019b).1
In May 2019, a Gateway Determination allowed the amendment to the LEP to proceed
subject to a set of conditions (DPE, 2019).
Tweed Shire Council put a revised proposal on public exhibition in July 2019, accepting
submissions until September 2019. At the time of drafting this report, the Tweed Shire
Council had not finalised the amendment to the LEP and was in the process of meeting the
set of conditions set out in the Determination.

1.2 REVIEW PROCESS

1.2.1 Updated Terms of Reference
On 28 February 2019, the Review Terms of Reference were expanded to include advice
on:
 scientific and technical approaches to examining socio-economic factors and
impacts and possible solutions using locally relevant examples
 localised environmental consequences related to extraction for bottled water.
The full TOR are at Appendix 1.

1.2.2 Experts
The Office of the NSW Chief Scientist & Engineer (OCSE) engaged technical experts in a
range of fields to assist the Review, including hydrology, groundwater, surface water,
groundwater and surface water interactions, modelling, monitoring, statistics and
uncertainty analysis. Experts engaged included:
 Associate Professor Will Glamore, Water Research Laboratory, School of Civil and
Environmental Engineering, UNSW Sydney (Phase 1 and 2)2
 Alice Harrison, Project Engineer, Water Research Laboratory, School of Civil and
Environmental Engineering, UNSW Sydney (Phase 1 and 2)
 Dr Mahmood Sadat-Noori, Research Associate, Water Research Laboratory, School
of Civil and Environmental Engineering, UNSW Sydney (Phase 2)
 Professor Neil McIntyre, Principal Research Fellow, Centre for Water in the Minerals
Industry, Sustainable Minerals Institute, University of Queensland (Phase 1 and 2)
 Dr Liliana Pagliero, Postdoctoral Research Fellow, Centre for Water in the Minerals
Industry, Sustainable Minerals Institute, University of Queensland (Phase 1 and 2)
 Dist. Professor Louise M. Ryan, Professor of Statistics, School of Mathematical and
Physical Sciences, University of Technology Sydney (Phase 1 and 2)

1
Section 7.15 of Tweed Local Environmental Plan 2014 states, Water bottling facilities in Zone RU2 Rural Landscape
(1) Despite any other provision of this Plan, development may be carried out with development consent for the purposes of a
water bottling facility on land in Zone RU2 Rural Landscape if the consent authority is satisfied that development will not h ave
an adverse impact on natural water systems or the potential agricultural use of the land. (2) Despite any other provision of
this Plan, development may be carried out with development consent for the construction of a pipe or similar structure on any
land for the purposes of conveying groundwater to a water bottling facility. (3) In this clause— water bottling facility means a
building or place at which groundwater from land in Zone RU2 Rural Landscape is extracted, handled, treated, processed,
stored or packed for commercial purposes.
2
Phase 1 is work prior to Initial Report; phase 2 is work from Initial Report to Final Report.
20
Over the course of the Review, information and advice on data, policy, guidelines,
monitoring, modelling and regulation, was sought from agencies with roles and
responsibilities relevant to the TOR. This included DPIE Water; Water NSW; the Natural
Resource Access Regulator; NSW Environmental Protection Agency; Energy, Climate
Change & Sustainability (DPIE); Rous County Council, Tweed Shire Council and others.

1.2.3 Site visits and consultations

The Review made two site visits to the Northern Rivers region, in December 2018 and
January 2019. These initial site visits concentrated on the areas near and around Dungay,
Urliup, Murwillumbah, Uki, Mt Warning, Ballina and Alstonville. Stakeholder meetings were
conducted with representatives from local government, the local community and industry.
Requests for relevant reports, additional information and data, and details of other
interested stakeholders were made to all stakeholders as part of these site visits and
meetings. In the second phase of work, the Review Team held various consultations with
community members and industry representatives.
Further details are at Appendix 2.

1.2.4 Submissions
Fourteen submissions were received over the course of the Review. Submissions are
available on the website of the NSW Chief Scientist & Engineer.

1.2.5 Workshop
To inform the Review, OCSE hosted a one-day multi-disciplinary expert workshop in
Sydney on 6 September 2019. The workshop brought together experts in hydrogeology,
groundwater hydrology and modelling, groundwater ecology, surface water, climate
modelling, geology, planning, uncertainty and statistics. Discussion encompassed:
 Science to inform extraction limits in Water Sharing Plan
 What the data from the Alstonville Basalt Plateau Groundwater Source tells us
 Assessing local impacts – hydrogeological studies
 Assessing impacts – research needs and approaches
Diverse views were presented at the workshop. However, all agreed that characterising
fractured rock systems, in particular, are complex due to their heterogeneous structure and
there are significant knowledge gaps in the region. High-level observations from the
sessions are referred to in relevant parts of the report. A list of participants that attended the
workshop is at Appendix 2.

1.3 STRUCTURE OF THIS REPORT

 Chapter 2 analyses growth trends and potential growth scenarios for the industry in
the Northern Rivers region, as well as the allocations for the industry compared with
overall allocations in the WSP.
 Chapter 3 addresses the TOR seeking advice on the sustainability of the extraction
limits in the Water Sharing Plan.
 Chapter 4 addresses impacts and environmental consequences from the bottled
water industry extraction, including impact mechanisms, the challenges to
understanding impacts and consequences, how impacts are assessed and
managed, and further information to assess impacts.
 Chapter 5 considers technical approaches to the socio-economic issues raised in
the review, including the use of plastic bottles and truck movements.
 Chapter 6 provides findings and recommendations.

21
2 THE BOTTLED WATER INDUSTRY

The Review TOR include data on the bottled water industry entitlements and extractions
having regard to total water access rights and WSPs; and advice on potential impacts on
groundwater resources arising from current industry activities and proposed or potential
expansion.
An analysis was undertaken of economic factors influencing supply and demand, which were
used to develop growth scenarios to predict future growth of the industry in the region.
Consideration was also given to whether the industry has different extraction patterns to
other users that could result in different extraction impacts. This chapter includes findings on
industry pumping characteristics as well as monitoring and data collection undertaken by
industry.
Issues specifically related to sustainability and impacts and environmental consequences
are addressed in subsequent chapters.

2.1 BOTTLED WATER INDUSTRY

2.1.1 The bottled water industry supply chain
The bottled water industry extracts, manufactures and bottles spring water, mineral water,
purified water and bulk water for sale (Table 1).3 Spring and mineral water represent 75% of
the bottled water market. The water is extracted from groundwater sources, processed,
bottled and primarily sold through grocery and convenience stores, restaurants, bars and
vending machines (IBISWorld, 2019).
Table 1: Product segments in the bottled water industry
Product segment Description Market share
Spring water Sourced from underground wells and contains a unique mix of 50%
minerals based on the characteristics of the well.
Mineral water Sourced from underground wells and contains a higher 25%
concentration of dissolved salts. No minerals are added.
Purified water Generally sourced from standard town water supply. Purified to 12.5%
remove impurities and contaminants.
Bulk and packaged Large bottled water products (>3 litres), generally sold to 12.5%
water businesses.
Source: (IBISWorld, 2019)
Note 1: Market share measured by reference to segment revenue versus overall market revenue.
Note 2: Bulk and packaged water suppliers source water from both standard water supply and groundwater.

The supply chain in the water bottling industry is generally divided between extractors
(operators specialising in water extraction) and water bottlers (operators who process, bottle
and distribute to retailers) (Figure 2). This is because most water bottlers (particularly major
bottlers) outsource the extraction of water to small local operators who hold the relevant
water access licences and approvals. However, some extractors also have small bottling
plants.

3
The bottled water industry typically does not include suppliers of water for soft drinks and other beverages, (for example
alcoholic beverages), as this water is generally supplied from standard water supplies (for example town water).

22
 Bars Cafes
Some businesses Bottling
Extractor Restaurants
both extract and plant Convenience
stores
bottle water. They Most bottling
Other
may sell any Bottling retail
plants are located
surplus water to Extractor plant close to major
Supermarkets
major bottlers. markets to reduce
distribution costs.
Most major water
bottlers outsource Bottling Plant Water bottlers sell
extraction to a most water
number of smaller domestically, with
Bottling Plant
local facilities. <10% exported.
Extractor
Extractor
Most extractors Extractor Australia is a net
are located close importer of water.
to bottling plants Most imports
to reduce comprise
transport costs. Trade premium brands.

Figure 2: High level supply chain in the bottled water industry

There are approximately 50 water bottlers in Australia, of which approximately 35% are
based in NSW (IBISWorld, 2019). As bottled water products are heavy and of low value
relative to their weight and size, the distribution of bottled water manufacturing facilities
largely reflects population distribution to minimise transportation costs, with most facilities
located near major population centres. As a result, larger manufacturers will often collocate
water bottling facilities with other bottling facilities (for example soft drinks) to reduce
manufacture and distribution costs through economies of scale and scope (IBISWorld,
2019).
Most bottled water manufactured in Australia is sold domestically, with only 8% of revenue
generated from exports (IBISWorld, 2019). Australia is a net importer of bottled water.4 Due
to the high costs of transport most of these imports comprise premium bottled water brands
which charge higher retail prices. These include brands such as Evian, Voss, Fiji Water,
Perrier, San Pellegrino and Vittel (IBISWorld, 2019).
Market share amongst water bottlers is highly concentrated. Coca-Cola Amatil Ltd (CCA),
Asahi Holdings (Australia) Pty Ltd and Nu-Pure Pty Ltd represent approximately three-
quarters of market share (IBISWorld, 2019). CCA’s major spring and mineral bottled water
brands manufactured in Australia include Mount Franklin, Pump, Neverfail Springwater,
Peats Ridge, and Glaceau Smartwater. Asahi Holdings’ major spring and mineral bottled
water brands manufactured in Australia include Cool Ridge and Frantelle. However, the
industry is also characterised by many small businesses, with 50% of businesses earning
less than $200,000 revenue per annum (IBISWorld, 2019).5
Market share amongst extractors is more fragmented. Some extractors operate a network of
extraction facilities across the country. There are also small extractors who operate only a
single site and compete with other extractors to supply water to water bottling plants.
Competition between extractors is driven by the quality and price of water that each extractor
can supply. Water quality is affected by the characteristics of the groundwater source. Price
is influenced by the costs of extraction, processing, transport, economies of scale and
margins. Extractors are often geographically proximate to water bottling facilities to reduce
transport costs.

4
In 2018-19, Australia imported approximately $340m of bottled water for domestic consumption and exported approximately
$56.1m of bottled water for international markets. Imports have been increasing faster than exports and Australia’s net trade
deficit of bottled water is expected to exceed $300m by 2022 (up from $283.9m in 2018-19) (IBISWorld, 2019)
5
Only 7% of businesses earn more than $5m per annum.

23
The industry reports that the manufacturing of bottled water contributes $186M to the NSW
economy per annum, and supports 1,047 full time jobs.6

2.1.2 Performance of the bottled water industry in Australia

Between 2014 and 2018, consumption of bottled water in Australia grew at an average rate
of 10% per annum by water volume, reaching approximately 1,100 ML in 2018 (Figure 3).
This growth in consumption has been driven by several factors, including by consumer
preferences trending towards snacks and takeaway meals (IBISWorld, 2019); water being
perceived as a healthier beverage alternative to soft drinks (Asahi Group Holdings, 2019);
and major grocery stores launching low cost private label bottled water products. 7

1200
CAGR +10 %
1000
Million litres

800

600

400

200

0
2014 2015 2016 2017 2018

Figure 3: Bottled water consumption in Australia

Source: 2014 to 2018 (Asahi Group Holdings, 2019)

While consumption of bottled water has grown in recent years, overall revenue of the
Australian bottled water industry has not been growing. Industry revenue is expected have
fallen by 0.4% per annum between 2014 and 2019 to $706.8m (IBISWorld, 2019). This
reflects the fact that low cost private label bottled water brands have grown volumes at the
expense of the higher priced brands.8 Overall industry revenue is expected to be flat or grow
modestly (at less than 1.5% per annum) over the period 2019-24 (IBISWorld, 2019; Statista,
2019).

2.1.3 Outlook for the bottled water industry in Australia

The Australian bottled water industry is influenced by a number of trends. The extent to
which these trends will influence future production is expected to vary. Table 2 sets out the
likely trends based on currently available forecasts. Based on these trends, the following
section explores a range of future scenarios.

6
Economic and employment contribution of manufacturing comprises direct manufacturing contribution of $73M and 281 FTEs,
and manufacturing supply chain contribution of $113M and 766 FTEs (from activities including extraction, other manufacturing,
transport, infrastructure, utilities and professional services) (ABC, 2019).
7
While bottled water consumption has been growing, this growth has been largely concentrated in sales of private label bottled
water products. Major bottled water brands have struggled to grow volumes due to this aggressive price competition from
private labels (Asahi Group Holdings, 2019; IBISWorld, 2019)
8
For example, despite growth in overall Australian bottled water consumption from 2014 to 2018, CCA has reported falling still
beverage sales since 2014 and sales volume declines of bottled water of 1.7% from FY17 to FY18. See Coca Cola Amatil Full
Year Results 2018 (Coca-Cola Amatil, 2019a), 2017 (Coca-Cola Amatil, 2018a); 2016 (Coca-Cola Amatil, 2017); 2015 (Coca-
Cola Amatil, 2016)); and 2014 (Coca-Cola Amatil, 2019b). At the same time, sales of private label bottled water have grown
from 21.2% of all bottled water sales in Australia in 2014 to 50% in 2018. However, the growth in market share of private lab el
bottled water has also slowed significantly from 12.2% growth between 2014 and 2015, to 3.2% growth between 2017 and
2018. (Asahi Group Holdings, 2019).

24
Table 2: Trends influencing bottled water production
Trends impacting bottled Past impact on Forecast future impact on production
water production in Australia production
(a) Increasing consumer ▲ ▲▲ Consumer preferences for takeaway meals are
preferences for takeaway expected to increase as disposable incomes
meals increase (RBA, 2019) and economies of scale
reduce takeaway cost.
(b) Increasing consumer ▲ ~ Sugar free, artificially sweetened beverage
preferences for healthier alternatives are now widely available and
beverages accepted.
(c) Increased availability of ▲ ~ Private label availability is already high and is
private label bottled water unlikely to increase further, evidenced by growth in
private label market share having been strong
(12.2% between in 2014 and 2015) but having
slowed significantly (3.2% between 2017 and
2018) (Asahi Group Holdings, 2019).
(d) Population growth ▲ ▲ Population growth is forecast to remain constant at
1.4-1.8% p.a. to 2027. From 2027 to 2042
population growth is expected to reduce to 0.9-
1.5% p.a (ABS, 2018).
(e) Increasing international ▲ ▲ Exports are forecast to continue to increase
demand for Australian modestly, particularly if the Australian dollar
bottled water exports remains depreciated (IBISWorld, 2019).
(f) Increasing Australian ▼ ▼▼ The trade imbalance in bottled water is forecast to
demand for imported bottled grow as imports increase much faster than exports
water (IBISWorld, 2019).
(g) Increasing consumer ▼ ▼▼ Consumer awareness is expected to increase
awareness of the rapidly due to media campaigns and incentives
environmental impacts of (e.g. container deposit schemes) (IBISWorld,
plastic 2019).
(h) Wide availability of tap water ▼ ~ Drought may impact the availability of water and
in Australia reduce disposable incomes in certain communities
(RBA, 2019).
(i) High quality of Australian tap ▼ ▼ Quality of Australian tap water is expected to
water remain high (IBISWorld, 2019).
Note: ▲= increased production; ▼ = decreased production; ~ = no impact on production

2.1.3.1 Bottled water consumption and production scenarios

The trends set out in Table 2 were used to assess potential future growth in the Australian
bottled water industry. Two scenarios were considered – the most likely (‘probable’) scenario
and an ‘unlikely’ scenario using available data over the next five years to 2024. Two more
scenarios are considered later in this Chapter – a ‘highly unlikely’ scenario and an ‘extremely
unlikely’ scenario using available data over the next 15 years to 2034.

Scenario 1 (probable) – 2% growth per annum

Industry forecasts predict that the most likely scenario is that the Australian bottled water
industry experiences modest revenue growth of approximately $11 million per annum (1.5%)
over five years from 2019 to 2024 (IBISWorld, 2019). Under this scenario, growth in
consumption of bottled water in Australia will slow to 2% per annum (IBISWorld, 2019).
Based on the trends in Table 2, under this scenario production growth is likely to slow to less
than 2% per annum; in particular because:
 Although consumption of bottled water in Australia grew at an average rate of 10%
per annum by water volume between 2014 and 2018, this growth has been trending
down, from 13% per annum in 2015 to 6% in 2018, and industry forecasts predict
average consumption growth of approximately 2% per annum to 2024 (Asahi Group
Holdings, 2019; IBISWorld, 2019).

25
  Low cost private label bottled water products have grown market share strongly (from
21.2% of all bottled water sales in Australia in 2014 to 50% in 2015), and driven
bottled water consumption growth by aggressive price competition. However, the
availability of private label products is now high, and the growth in market share of
private label bottled water has slowed significantly from 12.2% growth in 2015, to
3.2% growth in 2018 (Asahi Group Holdings, 2019; IBISWorld, 2019). As a result,
private labels are unlikely to drive significant consumption growth as they have done
in the past.
 The trade deficit in bottled water is expected to grow as imports increase faster than
exports. Growth in imported bottled water is forecast to exceed approximately $15
million per annum over five years from 2019 to 2024, while growth in exported water
is expected to be more modest at approximately $3 million per annum over five years
from 2019 to 2024 (IBISWorld, 2019). Therefore, growth in Australian consumption is
likely to be served increasingly by imported bottled water.
At a production growth rate of 2% per annum, the market would reach approximately 1,200
ML per annum by 2024 (a 10% increase).

Scenario 2 (unlikely) – 10% growth per annum

An alternative but unlikely scenario is that growth in consumption and production of bottled
water continues at 10% per annum. This scenario is unlikely because of the trends
described in Table 2 and discussed above, in particular because of the slowing growth in
bottled water consumption; the slowing growth in market share of private label brands; the
increasing trade deficit in bottled water; the increased availability of sugar free beverages;
and increased consumer concerns over plastic waste (IBISWorld, 2019).
At this growth rate, the market would reach approximately 1,900 ML per annum by 2024 (a
70% increase). Under this scenario, Australia’s per capita consumption of bottled water
would not reach that of the United States until 2034. The United States has one of the
highest levels of bottled water consumption per capita amongst developed economies – four
times greater than Australia (Figure 4).
200
Litres per capita

150
100
50
0

Figure 4: Consumption of bottled water in 2017 in developed economies

Source: (Beverage Marketing Corporation, 2018; Asahi Group Holdings, 2019; EFBW, 2019)

This scenario is also unlikely to occur in Australia because many of the factors that influence
high consumption in certain developed economies (e.g. Italy, United States, Germany and
France) are not present in Australia. These factors include that:
 High quality tap water is widely available in homes, restaurants, bars and workplaces
in Australia. Many other countries do not have high quality tap water or do not offer
free tap water in bars and restaurants, and other public spaces (IBISWorld, 2019).
 Many of these other countries have major premium international bottled water
brands, such as Evian in France and San Pellegrino in Italy. These brands contribute
to increased consumption in those countries through consumer loyalty.

26
It is also important to note that consumption in many of these markets appears to have
reached saturation and growth has peaked. For example, Italy and Germany have
experienced bottled water volume growth of only 0.1% and 0.7% respectively from 2012 to
2017 (Beverage Marketing Corporation, 2018). This suggests that there is an effective limit
on consumption volumes, even in high consumption markets. This is theoretically sound
since consumers can only drink a certain amount of water per day.
Under this scenario, where growth continues at 10% per annum until Australian annual
consumption reaches a limit of 160 L per capita (equivalent to the United States), the bottled
water market would reach approximately 4,500 to 5,000 ML per annum (depending on
population growth). Assuming that the percentage of Australian bottled water demand met
by imports remains constant, this would represent an approximately five fold increase in
Australia’s bottled water production by 2034.

2.1.4 The bottled water industry in the Northern Rivers region of NSW
The Review has identified seven operators, with allocations totalling 240.5 ML/y, that are
actively extracting for water bottling purposes from four groundwater sources in the Northern
Rivers region (Table 3).9 In addition to these, the Review also identified:
 two operators that have had their development applications approved but are not yet
extracting,
 one application to expand existing operations progressing through the development
application process,
 one application that has had the development application refused.
These four proposals, if approved and actioned, would amount to additional extraction
volumes of 168 ML/y across the four groundwater sources (Table 3). The Review also
identified one application which had been progressing through the development application
process but was subsequently withdrawn.

Table 3: Water licences associated with existing and proposed water bottling
Groundwater source Water licences associated with Water licences associated with
the bottled water industry (ML/y) proposed expansions in bottled
water industry (ML/y)
Alstonville Basalt Plateau 7.5 -
(fractured)
Clarence Moreton Basin (porous) 50 100
New England Fold Belt Coast 163 68
(fractured)
North Coast Volcanics (porous) 20 -

Water licences associated with the current and proposed future expansions in the bottled
water industry represent a very small portion of the overall water requirements10 from
groundwater sources in the Northern Rivers region (Figure 5). These licences represent an
even smaller portion of total groundwater recharge in the Northern Rivers region, as most
groundwater recharge is reserved for the environment or is not licenced (Figure 6).

9
The Review identified active and proposed water bottling operations in the Northern Rivers region through a search of
secondary sources, including local council development applications and related documents, internet searches and information
from state agencies, the community and councils. This is because the public water access licence registers (for example, the
NSW Water Register and the NSW Water Access Licence Register) do not record the purpose of each licence.
10
‘Requirements’ is a term used in the WSP, which is the sum of the estimated basic landholder rights, town water supply and
all other licenced entitlements (ML/yr).

27
 Clarence Moreton Basin water requirements

Potential future water requirements [a]

Existing water requirements

0 1,000 2,000 3,000 4,000 5,000 6,000

ML/y

New England Fold Belt Coast water requirements

Potential future water requirements [a]

Existing water requirements

0 5,000 10,000 15,000 20,000 25,000

ML/y

North Coast Volcanics water requirements

Potential future water requirements [a]

Existing water requirements

0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000

ML/y

Alstonville Basalt Plateau water requirements

Potential future water requirements [a]

Existing water requirements

0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000
ML/y

Bottled water licences All other licences Local water utility access licences Basic landholder rights

Figure 5: Water requirements in the Northern Rivers region by groundwater source

[a] Potential future water requirements includes proposals for water extraction for water bottling where the proponent (1) has
already obtained water allocations or works approval; and (2) is seeking, or has indicated that they will seek a development
application, or will commence operation. This analysis assumes that no additional non-bottled water industry water
requirements are created – however if this occurred, then the fraction of water requirements associated with the water bottling
industry would be even smaller.

28
 Potential future water requirements [a]

Existing water requirements

0 10000 20000 30000 40000 50000

ML/y

Potential future water requirements [a]

Existing water requirements

0 500000 1000000 1500000 2000000 2500000 3000000

ML/y
Bottled water licences All other licences Local water utility access licences
Basic landholder rights Unassigned water Recharge reserved for the environment

Figure 6: Water requirements, unassigned water, and groundwater recharge reserved for the
environment in the Northern Rivers region
[a] See note [a] at Figure 55 above.
Note: Northern Rivers region comprises Alstonville Basalt Plateau, Clarence Moreton Basin, New England Fold Belt Coast, and
North Coast Volcanics groundwater sources

2.1.5 Outlook for the bottled water industry in the Northern Rivers region
Based on the scenarios described above for potential future growth in the Australian bottled
water industry,11 forecasts have been made for growth in the bottled water industry in the
Northern Rivers region to 2024 and 2034. These forecasts assume that growth rates are
likely to be evenly distributed across Australia close to major population centres, and are
unlikely to be concentrated in a single region. Bottled water products are bulky and low value
relative to their weight and size, and therefore incur relatively high transport costs. Extractors
and water bottlers are close to major population centres to reduce these costs through
economies of scale in both production and transport (IBISWorld, 2019). For this reason,
most large water bottlers have processing facilities across Australia near major capital cities.
It is likely any future expansion in the bottled water industry would also be undertaken at
multiple sites close to major markets to minimise distribution costs.
As a result, the rate of growth of the bottled water industry (including production and
extraction) in the Northern Rivers region is unlikely to significantly exceed the rate of growth
bottled water consumption in nearby markets. Furthermore, if the bottled water industry in
the Northern Rivers region obtains significant additional water allocations beyond future local
demand for bottled water, it is likely that these allocations would not be fully utilised, or would
offset existing uncompetitive extraction – under the probable (Scenario 1) and unlikely
(Scenario 2) scenarios discussed below.

2.1.5.1 Forecast groundwater extraction by the bottled water industry to 2024

Scenario 1 (probable) – 2% growth per annum to 2024

As discussed, the most likely scenario for growth in bottled water production in Australia is
that it is likely to slow to less than 2% per annum, representing a 10% increase to 2024
(Scenario 1). Based on this forecast, and assuming that the 240.5ML/y of water allocated to
the current bottled water industry in the Northern Rivers region are fully utilised, then 2%
growth in demand for water extraction is likely to be approximately 265 ML/y by 2024 (Figure
7).

11
Note: scenarios considered were Scenario 1 (probable) – 2% production growth per annum, and Scenario 2 (unlikely) – 10%
production growth per annum.

29
Under this scenario, the 168 ML/y of water licences associated with proposed expansions in
the bottled water industry are well in excess of the approximately 25 ML per annum
necessary to meet anticipated demand in 2024. Therefore, these water licences would likely
be underutilised and/or offset approximately 145 ML/y of water licences associated with the
existing bottled water industry in the region (Figure 7).

Scenario 2 (unlikely) – 10% growth per annum to 2024

An unlikely scenario would be that growth in the bottled water industry in the Northern Rivers
region will be 10% per annum, representing a 60% increase to 2024 (Scenario 2). Under this
high growth scenario, the 168 ML/y of water licences associated with proposed expansions
in the bottled water industry will still be sufficient to fully meet growth in demand of
approximately 150 ML/y in the region by 2024 and no additional expansions would be
expected (unless they offset existing extractions) (Figure 7).

450
Water licences associated
400
Water licences in excess of forecast with proposed expansions in
350
demand increases will likely offset bottled water industry
existing licences and not (408.5 ML/y)
300 signficiantly increase overall
extraction (Scenarios 1 and 2)
Volume (ML/y)

250 Water licences associated

200
with the current bottled
water industry in 2019
150 (240.5 ML/y)

100

50

0
2019 2024
Estimated water demand for bottled water industry under Scenario 1 (probable)
Estimated water demand for bottled water industry under Scenario 2 (unlikely)

Figure 7: Bottled water industry growth scenarios 1 (probable) and 2 (unlikely)

2.1.5.2 Highly unlikely and extremely unlikely groundwater extraction scenarios

to 2034
While the expected growth in the bottled water industry in the Northern Rivers region is
expected to be modest to 2024 (see Scenario 1), it is useful to consider the impact of highly
unlikely scenarios where the industry experiences high growth over a long timescale. These
scenarios assist in understanding the theoretical limits on the industry in this region. Two
extreme growth scenarios are considered:

Scenario 3 is highly unlikely and considers the impact of 10% growth per annum in
the bottled water industry in the Northern Rivers region until 2034, under this
scenario the growth rate of Scenario 2 continues an additional 10 years.
 Scenario 4 is extremely unlikely and considers the impact of a major bottled water
exporter being established in the Northern Rivers region.
The impacts of Scenario 3 and Scenario 4 are analysed below.

Scenario 3 (highly unlikely) – 10% growth per annum to 2034

Scenario 3 assumes that the bottled water industry in the Northern Rivers region grows at
10% per annum, reaching approximately 1,000 ML/y in 2034, but does not exceed this level
(Figure 8). Under this scenario, Australian bottled water consumption peaks at four times

30
current volumes in 2034 – similar to current per capita consumption volumes in the United
States (Figure 4).12 This scenario also adopts the same assumptions as those in Scenarios 1
and 2: that growth rates are likely to be evenly distributed across Australia close to major
population centres, and are unlikely to be concentrated in a single region.

Scenario 4 (extremely unlikely) – a major bottled water exporter is established in the

region
Although the size of the bottled water industry in a region is generally influenced by bottled
water consumption in nearby markets, there are a number of international premium bottled
water brands that extract water from groundwater sources in a single area, and bottle and
distribute it to international markets.13 This scenario assumes that a major premium bottled
water brand is established and that it:
 obtains the necessary licences and approvals to extract water from groundwater
sources in the North Rivers Region,
 extracts solely from sources in this region to satisfy its entire customer demand,
 establishes world leading brand recognition and availability in multiple major global
markets (e.g. Europe, the Americas, Asia, the Middle East),
 overcomes well-established international competitors to build up significant market
share, and
 becomes one of the world’s largest producers of bottled water.
If a brand were able to overcome all of these hurdles, this process would take many years.
Most premium bottled water brands exporting to international markets have long histories
(San Pellegrino, 2019; VOSS, 2019). This is because it takes a significant amount of time to
build up an international brand identity that can charge a premium necessary to cover the
cost of large scale distribution to international markets.
Evian is one of the largest bottled water brands in the world and sells premium bottled water
in more than 140 countries (Danone, 2019). The company has extracted and sold water from
groundwater sources near a single bottling plant in Evian-les-Bains in France since 1826
(Evian, 2017). Evian produced 1,441 ML of bottled water in 1999 (Danone, 2000). Assuming
this scenario occurred, and a brand of the scale of Evian (i.e. of approximate production of
1400 ML/y) was able to be established, it would still represent a small fraction of
groundwater extraction, unassigned water and total estimated groundwater recharge in the
region (Figure 8).

12
Bottled water consumption per capita in the United States is currently four times greater than Australia. This scenario models
a four-fold increase in Australian consumption volumes – equivalent to a 3.2-fold increase in Australia’s per capita consumption
plus forecast population growth in Australia to 2034.
13
For example, Evian in France, San Pellegrino in Italy, and Voss in Norway.

31
 Scenario 4 (extremely unlikely)
Scenario 3 (highly unlikely)
Potential future water requirements [a]
Existing water requirements

0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000

Under Scenarios 3 and 4), bottled
water licences are a small fraction ML/y
of overall estimated annual
groundwater recharge

Scenario 4 (extremely unlikely)

Scenario 3 (highly unlikely)
Potential future water requirements [a]
Existing water requirements

0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000

ML/y
Bottled water licences All other licences Local water utility access licences
Basic landholder rights Unassigned water Recharge reserved for the environment

Figure 8: Bottled water industry growth scenarios 3 and 4

Note 1: Northern Rivers region comprises Alstonville Basalt Plateau, Clarence Moreton Basin, New England Fold Belt Coast,
and North Coast Volcanics groundwater sources.
Note 2: For Scenario 4, this analysis assumes that the impact of a newly established bottled water plant producing 1,400 ML/y
14
would result in an increase in extraction volumes by an additional 1,610 ML/y.
[a] See note [a] at Figure 55 above.

2.1.6 Policy and regulatory interventions

The bottled water industry is subject to local council15 and NSW Government16 policy,
planning and regulatory requirements. These are discussed in detail at Appendix 4 of the
Initial Report. It is not possible to forecast possible future changes in laws or regulations
which impact groundwater extraction. While regulatory intervention may directly or indirectly
restrict groundwater extraction for water bottling in particular areas, these restrictions would
be unlikely to impact underlying consumer demand for bottled water or overall bottled water
consumption.

14
As a conservative assumption, this analysis assumes a Water Use Ratio of 1.46L/L (WUR, the litres of water including
product water used to make one litre of bottled water) and assumes 30% of the additional water required to produce 1,400 ML/y
of bottled water is extracted groundwater, and that the remaining additional water comes from other sources. The WUR for the
bottled water industry is 1.32L/L in North America and 1.46L/L internationally (Antea Group, 2015). Note that the Australian
bottled water industry anecdotally reports lower WURs. The WUR includes all water used by the facility, including product
water, and water used for facility processes (e.g., treatment, cleaning, maintenance).
15
Local councils can indirectly control the volume of water extraction in their local government areas (LGAs) through
restrictions in the development consent, for example, by imposing limits on truck movements. Local councils can also (within
limits) control the types of businesses in certain zones through the terms of the relevant LEP, which govern the conditions and
types of development that can occur in their LGAs.
16
The NSW Government, through the Water Management Act 2000 and WSPs, controls the volume and manner of water
extraction from particular groundwater sources. These instruments also regulate the water licencing, allocation and trading
process with the objective of ensuring the economically efficient allocation of water to the commercial uses of highest economic
value. These instruments also establish priorities between environmental, domestic and stock, and industrial and commercial
extraction in particular groundwater sources. However, within the category of industrial and commercial extraction there is no
distinction made between the purposes of the groundwater extraction. As a result, the laws and regulations do not incentivis e
or disincentivise water extraction for water bottling over extraction for other commercial purposes (for example, farming, other
manufacturing).

32
2.2 WATER ENTITLEMENTS FOR BOTTLED WATER
FACILITIES IN THE NORTHERN RIVERS
The Initial Report outlined challenges in determining the scope of the bottled water industry,
in part due to commercial water access licences specifying allowable extraction rates and
not the intended use of the extraction. Other means, including sourcing and reviewing
development applications and consents, are required to confirm if groundwater extraction is
being used for bottled water.
Table 5 provides an updated overview of the total available water for all purposes by
groundwater source. It includes landholder rights and entitlements, as well as an estimate of
the water entitlements held by the bottled water industry in the Northern Rivers area. It is
emphasised that Table 5 provides a summary of the licence entitlements, not a record of
actual water taken. It also does not reference any additional restrictions imposed on water
take imposed through the development consent or any self-imposed limits on water take, so
it is likely to overestimate actual water extraction. The Review is aware of several examples
where the licence entitlement is higher than the allowable extraction through the
development consent; this table reflects the entitlement volume only.
This section examines the volume of existing and proposed total licence allocations for the
bottled water industry against the total water requirements, including all licences and basic
landholder rights, for each of the four groundwater sources. The percentages of existing and
proposed licences used for the bottled water industry for each sources is at Table 4. Under
the WSP, environmental water and Basic Landholder Rights are given priority over licensed
water extraction. Among licensees, priority is given to water utilities and licensed stock and
domestic over commercial licensed purposes.

Table 4: Existing and proposed bottled water licence entitlements as a proportion of total water
requirements (including licenced extraction and BLR) by groundwater source
Groundwater source Existing Existing and proposed
Alstonville Basalt Plateau 0.1% 0.1%
Clarence Moreton Basin 1.0% 3.0%
New England Fold Belt Coast 0.7% 1.0%
North Coast Volcanics 0.3% 0.3%

Figure 9 illustrates the total estimated annual aquifer recharge for each relevant groundwater
source in the WSP, the amounts of the recharge that are reserved for the environment in the
WSP and the long-term annual average extraction limit (LTAAEL), including the breakdown
by licence, Basic Landholder Rights and unassigned water. This figure does not consider the
Upper Extraction Limit (UEL), which in the case of the New England Fold Belt Coast and the
North Coast Volcanics, as shown in Table 5, are higher than the LTAAEL.
In the case of the New England Fold Belt Coast and the North Coast Volcanics, the vast
majority of the recharge is reserved for the environment (97% and 96% respectively), with
only a small proportion that can be allocated to licences. This is because the LTAAEL for
these water sources is set as the current entitlement plus estimated future water
entitlements for the term of the plan rather than as a percentage of recharge. In these water
sources the percentage of recharge protected from extraction (75%) is defined in the UEL.
For Clarence Moreton Basin, the amount reserved for the environment is 48% of the
recharge, but for this groundwater source, only 1.7% of the amount which can be allocated
for licences is allocated. For Alstonville, 82% of the recharge is reserved for the
environment, but the remainder that is available for licences is fully allocated. The total
volume of storage of the four aquifers is fully reserved for the conservation of the
groundwater system.

33
Of the recharge that is not reserved for the environment (i.e. the LTAAEL in the case of
these four aquifers), some is allocated to Basic Landholder Rights and licences, while the
remainder remains unassigned water. At present, approximately 38.0% of LTAAEL in the
New England Fold Belt Coast is allocated, 51.3% of the North Coast Volcanics and 1.7% in
the Clarence Moreton Basin is allocated. Alstonville is fully allocated. A fraction of these
allocations are for licences for the bottled water industry (Table 4 and Table 5).

LTAAEL

Alstonville Basalt Plateau

0 10000 20000 30000 40000 50000 60000

ML/y
LTAAEL

Clarence Moreton Basin

0 100000 200000 300000 400000 500000 600000 700000

ML/y
LTAAEL
New England Fold Belt

0 250000 500000 750000 1000000 1250000 1500000 1750000 2000000

ML/y
LTAAEL

North Coast Volcanics

0 50000 100000 150000 200000 250000 300000 350000

ML/y
Bottled water licences All other licences Local water utility access licences
Basic landholder rights Unassigned water Recharge reserved for the environment

Figure 9: Breakdown of total estimated annual aquifer recharges in the Northern Rivers region by water
requirements, unassigned water, groundwater recharges reserved for the environment and LTAAELs

34
 Table 5: Available water, extraction limits and requirements by purpose and groundwater source (updated October 2019)
A (ML) B (ML/yr) C (ML/yr) D (ML/yr) E F (ML/yr) G (ML/yr) H (ML/yr) I (ML/yr) J (ML/yr) K (ML/yr) L (ML/yr) M
Groundwater Estimated Estimated Recharge Upper (ML/yr) Un- Total Basic Local All other Bottled Bottled Total no.
Source Total total amount Extraction assigned requirements landholder water aquifer Water Water water
3,4
Aquifer annual reserved for Limit LTAAEL Water (BLR and rights utility access Licences Licences access
1,a,b 3, 2,3 2,3,d l
Storage aquifer environment (UEL) 2,3,c licences) (BLR) access licences (Existing (Proposed licences
a e,f f,g 4,k
recharge licences Industry) Industry) (WALs)
3 4

h h 5,i
Alstonville 640,000 50,079 41,184 na 8,895 0 9,086 2,014 1,230 5,842 7.5 0 192
Basalt
Plateau
4
Clarence na 576,000 276,000 na 300,000 294,847 5,153 2,341 31 2,781 50 100 136
Moreton
Basin
4
New England 24,000,000 1,980,000 1,920,000 375,000 60,000 37,227 22,773 9,605 240 12,928 163 68 558
Fold Belt
Coast
4
North Coast 4,380,000 310,000 297,000 55,000 13,000 6,327 6,673 3,402 0 3,271 20 - 205
Volcanics
j
Relationship B=C+E (C=B-E) D E F =E-G G=H+I+J Subset of J Subset of J
between
columns
Sources and notes:
1 Estimated based on total area, porosity, average saturated thickness of source (EMM, 2018)
2 WSP NCFPR (July 2016)
3 WSP NCFPR – Background document (Sept 2016)
4 NSW Water Register https://waterregister.waternsw.com.au/water-register-frame - data used is from 19/20 year for each groundwater source
5 Supplied by DPIE Water (DOI Water, 2019)
a. Sources 2 and 3 define Planned Environment Water as equal to the total recharge minus the LTAAEL plus the portion of storage not available for extraction. At the commencement of the WSP
NCFPR, 100 percent of groundwater storage is reserved as planned environmental water.
b. Reserved as part of Planned Environment Water – allocations made only on recharge
c. LTAAEL is long-term average annual extraction limit.
d. BLRs comprise domestic and stock but do not include volumes for Native Title Rights due to difficulty predicting volumes used
e. Column K is based on the full volume of a licence entitlement where all or part of that licence may be extracted for bottled water. This number does not reflect any other restrictions on the licences,
e.g. through development consent conditions or voluntary etc.
f. Due to difficulty in confirming bottled water industry participants, there may be some small extractions captured in ‘all other aquifer licence entitlements’ column J that are not yet captured in
columns K and L.
g. Under Column L, WALs, water supply approvals or general terms of approval may have been issued; and there is some indication in the public domain of either works approval or development
application in process to start/expand extraction.
h. There is no unassigned water in Alstonville Basalt Plateau Groundwater Source. The total requirements do not represent actual take. When considering AWDs, actual water take is assessed
against LTAAEL to determine volume or percentage of unit share.
i. DoI Water noted that two licences were handed back to the Water Administration Ministerial for a total of 10 ML/yr. The figure reflects this (DOI Water, 2019)
j. The WSP NCFPR (2016) reflects unassigned water as LTAAEL minus total requirements (p. 35). This method is used with updated figures.
k. Includes Water Utility Licences
l. Total Share Component

35
2.3 REGULATION AND CHARACTERISTICS OF EXTRACTION
There is a requirement for all food for sale in NSW, including bottled water, to be safe and
suitable for human consumption.17 Bottled water also is subject to the Australian New
Zealand Food Standard Code.18 The Code defines bottled water (the term ‘packaged’ water
is used), its composition and labelling requirements. Bottled water is required to comply with
certain limits on with microbiological and chemical contaminants, toxicants and processing
aids. For example, the Code specifies a microbiological limit for packaged water, whereby E.
Coli should not be detected in a sample of 100mL of water.19
Although, the Code requires that the bacteriological and physical quality of bottled water
comply with these criteria, it does not specify frequency of analysis. However, if samples are
found to be contaminated or exceed limits specified in the Code, then appropriate action
would be taken by the manufacturer or the relevant regulatory agency. As the Code applies
nationally but is enforced locally, the responsible agency will vary in each case but may be
the local council, the NSW Food Authority or interstate agencies.

2.3.1 Australasian Bottled Water Institute Model Code

Some members of the bottled water industry within the Northern Rivers region are members
of the Australian Beverages Council (ABC), which is an industry body representing the non-
alcoholic beverage industry. The Australasian Bottled Water Institute (ABWI), a division of
the ABC has developed a Model Code with guidelines relating to quality and safety,
including Good Manufacturing Practices (GMP) (ABWI, 2018).
Members of the ABWI are required to comply with certain provisions of the Model Code
when producing bottled water. The Code specifies key testing requirements relating to the
extraction of water, the transport of water, and the processing of water to minimise the risk
from biological, chemical or physical contaminants. In relation to water extraction, the Code
also requires that a review be undertaken by a hydrogeologist that demonstrates the integrity
of the source and the safety of the catchment operations. This review should also include
(ABWI, 2018):
“(a) An evaluation of the chemical, physical, microbiological, and radiological
characteristics of the source.
(b) A report on the regional geology surrounding the site and the specific site geology
… to define the recharge area of the aquifer, or in the case of regional aquifers, the
zone of influence of the subject source.
(c) A report detailing the development of the source …
(d) A watershed survey of the recharge area or zone of influence of subject source
that identifies and evaluates actual and potential sources of contamination, including
any reported discharge that may affect the source.
(e) Based on the findings in item (d), a plan for special monitoring of any significant
contaminant source and for taking restrictive preventive or corrective measures as
appropriate to protect the source water.”
In addition to this hydrogeology review, other tests are required including microbiology tests
(for example, for Coliforms, and/or E. Coli, yeasts and moulds) and chemistry tests (physical,
radiological, inorganics, organics and volatile organics). Testing frequency varies from
weekly to every four years (ABWI, 2018).

17
Food Act 2003, s 3(a).
18
Australia New Zealand Food Standards Code. Standard 2.6.2 Non-Alcoholic Beverages and Brewed Soft Drinks
19
Australia New Zealand Food Standards Code – Schedule 27 – Microbiological limits in food, cl 4.

36
2.3.2 Characteristics of bottled water extractors
It appears that extractors in the bottled water industry in the Northern Rivers region are using
bores which range from a depth of approximately 25 to 90 metres, but more typically range
from 30-50 metres.20 In contrast, stock and domestic bores tend to be shallower as they are
generally sunk until water is reached (to minimise costs of drilling and construction).
Anecdotally, the Review was advised the purpose is to draw from deeper aquifers to attempt
to minimise or avoid interaction with surface water and other shallower bores. Interaction
between groundwater sources and surface water can have negative impacts on water quality
and quantity.

2.3.2.1 Managing water quality

As discussed, the bottled water industry is subject to the Food Standards Code which
regulates the safety and quality of the water. Some operators are also subject to obligations
under the ABWI Model Code which control the extraction, transport and processing of
bottled water. However, groundwater extractors supplying water bottling plants also have
direct commercial incentives in monitoring the quality of their product to ensure that it is of
sufficient quality for sale. Unsafe or contaminated water poses issues of legal liability and
brand reputation for the extractor and bottled water producer.
Groundwater from depth is generally microbiologically safe and chemically stable, however,
contamination of the groundwater bores can still occur. Sources of contamination can
include seepages associated with septic tank discharges (potentially introducing pathogens
and nitrates), agricultural practices (potentially introducing pathogens, nitrates and
pesticides) and industrial wastes (NHMRC & NRMMC, 2011). Contamination can also occur
from natural sources including water flowing into the target aquifer from a connected aquifer
that has different physical or chemical characteristics. If interference from other sources
causes the water quality to change, the water may no longer be suitable for sale for bottling.
For example, a connection between a groundwater source and an E. Coli contaminated
surface stream could contaminate groundwater water for bottling purposes.

2.3.2.2 Managing water quantity

In addition to ensuring water is free from contamination, bottled water extractors have a
commercial interest in ensuring a steady supply of water to meet contractual obligations
and/or manage commercial risk. Sustainable extraction requires that groundwater levels
recover during and after pumping.
The Review received anecdotal information that suggested that bottled water extractors
were generally extracting water at an approximately evenly spaced production rate year-
round compared with many other commercial users who extract on a more periodic basis.
While theoretically sound given supply chain arrangements in the bottled water industry, the
Review was not able to independently verify these observations. Further, all groundwater
extractors are subject to changing environmental conditions which may influence future
patterns of use.
As part of the pump testing requirements, extractors are required to conduct assessments
on groundwater levels, pumping rates, drawdown and recovery, and water quality.
Conditions of licence also require the collection of information about water extraction
volumes and rates of extraction.21 The implementation of the NSW Non-Urban Water
Metering Policy (DOI, 2018b), with specific requirements around groundwater metering for
groundwater extraction works over a specified size, will provide improved information about
use patterns in the bottled water industry and enable analyses to be undertaken on
interactions and impacts.

20
Strictly, the depth at which the bore casing is slotted defines the depth of the extraction, however this tends to be close t o the
bore depth.
21
Conditions of licences are discussed further in Section 4.5.1
37
2.4 CONCLUSIONS
The industry
 Available industry data indicates that across Australia, over three-quarters of bottled
water is sourced from underground wells, and the remainder from standard
reticulated water supplies. Approximately 8% of Australian bottled water production is
exported.
 The Review identified seven operators in the Northern Rivers region with allocations
of 240.5 ML/y who are actively extracting for water bottling purposes, representing
0.55% of water licences and basic landholder rights (together defined in the WSP as
‘total water requirements’) and 0.008% of estimated total annual aquifer recharge in
the four groundwater sources.
 Four further proposals, if approved, would amount to an additional 168 ML/y, being
an additional 0.38% of estimated total water requirements and 0.006% of total annual
aquifer recharge.
 Changing consumer preferences, trade imbalances, the availability of tap water and
private (‘no name’) brands and population growth are expected to impact future
bottled water production and consumption volumes.
 Scenario analyses conducted by the Review suggest the Australian bottled water
industry is most likely to grow at a rate of less than 2% per annum to 2024 and that
growth in the Northern Rivers region is likely to be consistent with this trend. Under
most scenarios to 2024 considered, the 168ML/y of additional proposed bottled water
operations would be sufficient to meet fully projected growth in demand.
 The Review also considered ‘highly unlikely’ and ‘extremely unlikely’ scenarios to
2034, being growth continuing at the current rate of 10% per annum and
establishment of a major premium bottled water exporter in the Northern Rivers,
respectively.
o If the ‘highly unlikely’ scenario occurred, the bottled water industry would
represent less than 2.3% of ‘total water requirements’ and 0.034% of
estimated total annual aquifer recharge.
o If the ‘extremely unlikely’ scenario occurred, the bottled water industry would
represent less than 4.6% of ‘total water requirements’ and 0.069% of
estimated total annual aquifer recharge.
 As the scenario analyses considered an unchanged regulatory and policy
environment, these forecasts may be affected by regulatory intervention which
directly or indirectly impacts the bottled water industry in this region.
 For the purposes of water extraction licensing, the bottled water industry is treated
the same as other prospective commercial users. However, development consent
under the Environmental Planning and Assessment Act 1979 is required for water
bottling activities. Approvals for bottled water extraction in the Northern Rivers region
identified by the Review date from 1993.
Water entitlements and allocations
 The WSP determines the allowable extraction limit, set from the recharge value of
each aquifer, with an amount of the recharge reserved for the environment and the
reminder determining the UEL or the LTAAEL.
 Under the North Coast Fractured and Porous Rock Water Sharing Plan (WSP),
environmental water and basic landholder rights are given priority over licensed
water extraction. Among licensees, priority is given to water utilities and licensed
stock and domestic over commercial licensed purposes.

38
  At the commencement of the WSP for the four groundwater sources, 100% of
storage is reserved for the conservation of the groundwater system.
 Water available for extraction is a portion of the estimated recharge value for each
groundwater source. This is determined by the WSP. An amount of the recharge is
reserved for the environment. The amount reserved for the environment equates to
97% of the estimated recharge value for New England Fold Belt Coast, 96% for
North Coast Volcanics, 82% for Alstonville Basalt Plateau and 48% for Clarence
Moreton Basin.
 The remaining amounts can be allocated for licensed purposes. Of these amounts,
38.0% of the New England Fold Belt Coast is allocated, 51.3% of the North Coast
Volcanics and 1.7% in the Clarence Moreton Basin. Alstonville is fully allocated.
 These are average values over the groundwater source areas, which means that the
environment is not protected to these levels in locally impacted areas.
Characteristics of extraction
 The Review received anecdotal information suggesting that bottled water extractors
were generally extracting water at an approximately evenly spaced production rate
year-round compared with other commercial users who extract on a more periodic
basis. The Review was not able to verify these observations.
 All groundwater users are subject to future changing environmental conditions, which
may influence their future patterns of use.
 The implementation of the NSW Non-Urban Water Metering Policy will provide
information about use patterns in the bottled water industry and enable analyses of
interactions and impacts.

39
3 SUSTAINABILITY OF EXTRACTION LIMITS IN THE WATER
SHARING PLAN

This Chapter reports on the analysis of how sustainability is assessed and managed under
WSPs, and the Water Sharing Plan for the North Coast Fractured and Porous Rock
Groundwater Sources (the WSP) in particular.
The Water Management Act 2000 makes provision for WSPs to be developed which apply to
groundwater and surface water systems in NSW and to implement the overall objective of
the Act of “sustainable and integrated management of the State’s Water” (DLWC 2001).
The WSP sets extraction limits and rules about groundwater extraction for the four
groundwater sources in scope of this review based on multiple factors. These factors include
groundwater recharge, a risk assessment, planned environmental water and current and
future water requirements (including Basic Landholder Rights and licenced entitlements).
Groundwater sustainability relates to limiting the current (environmental and anthropogenic)
use of the resource to ensure long-term water security in an aquifer. Short-term changes in
an aquifer (e.g. water level declines) do not necessarily indicate unsustainable use due the
variability of these natural systems which are inherently dynamic. As such, understanding
the impact of groundwater extractions on long-term sustainability of an aquifer can be a
complex issue even when substantial data is available.
The extraction limits in the WSP are designed to ensure the sustainable on-going use of the
groundwater resource. The aim of this chapter is to understand whether the extraction limits
adopted are considered ‘sustainable’ for the entire groundwater source. However, potential
local scale social, environmental or economic effects are also important. These additional
considerations are considered separately in Chapters 4 and 5.
This Review considers the impact of uncertainty arising from the variables used to determine
extraction limits, particularly the recharge rates which ultimately drive the limits in the WSP,
and compares the values to those available for comparable aquifers, as well as alternative
methods of calculating recharge. A hypothetical scenario is also presented in this section, in
which the impact of reducing the recharge rate by 80% is considered to illustrate the impact
of the rate on sustainable groundwater extractions.
For most of the groundwater sources considered in this review, groundwater extraction per
year is relatively small compared to the total size of the groundwater source and monitoring
data is sparse. The exception to this is in the Alstonville Plateau which has a high (relative to
its size) annual extraction, but is also equipped with a network of state operated monitoring
piezometers or bores. This section analyses the available monitoring data from 2006 to
present to assess the general levels of the aquifer and provide commentary on whether this
data can be used to assess the sustainability of current groundwater extractions.

3.1 EXTRACTION LIMITS UNDER THE WSP

The Review’s Initial Report included the legislative and policy framework under which water
sharing plans are developed, operationalised and reviewed, including that for the Northern
Rivers region, and the Initial Report also described the technical approach to assessing and
determining extraction limits under the WSP. For convenience, relevant components are
reproduced in Appendix 3 of this report. This Section describes the Review’s analysis of the
use of WSPs as regional water management tools, how sustainability is conceptualised and
integrated into the WSP and how the impact of the WSP is monitored, reviewed and can be
amended as part of an adaptive management approach.

40
3.1.1 WSPs as regional water management tools
WSPs support the regional scale implementation of the objectives of the Water Management
Act 2000: “to provide for the sustainable and integrated management of the water sources of
the State for the benefit of both present and future generations”.22 Generally, each WSP for
groundwater will cover a number of different regional groundwater sources. Management
zones within certain groundwater sources may be defined where specific access and trading
rules are required to control extraction and prevent localised impacts in sensitive areas. An
example of this is the Alstonville Basalt Plateau Groundwater Source which is comprised of
the Alstonville-Tuckean Management Zone and the Bangalow-Wyrallah Management Zone.
WSPs are influenced by a number of other legislative instruments and policies guiding the
management of water in NSW. These include:
The Water Management Act 2000: which governs the management and extraction of
water in NSW.23
 The Access Licence Dealing Principles Order 2004: which provides guidance and
rules for water access licence dealings which are reflected in the WSPs.
 The National Water Initiative: an intergovernmental agreement to implement “a
nationally-compatible, market, regulatory and planning based system of managing
surface and groundwater resources for rural and urban use that optimises economic,
social and environmental outcomes”. The NWI relates to water management
elements including water access entitlements, environmental and other public benefit
outcomes, adaptive management, risk management, managing over allocation,
regional variability and allocation decisions.
 The Standard for Quality Natural Resource Management: developed by the Natural
Resource Commission to inform natural resource management (including monitoring,
evaluation and reporting arrangements) at all scales in NSW including state, regional,
catchment and local level.
 Catchment Action Plans: which are non-regulatory plans prepared by Local Land
Services jurisdiction that set a strategic direction for the sustainable use and care of
natural resources in each region.
 DPIE Water is leading the development of Regional Water Strategies for each of 12
regions across NSW. The strategies will inform plans and management of a region’s
short and long-term water needs. They look at how much water a region will need to
meet future demand and determine ways to manage risk to water availability and
security. They will incorporate new data to improve understanding of climate risk,
including consideration of climate change and the probability of extreme events.
 Other water planning policies and guidelines developed by NSW Government
agencies which support the development and implementation of WSP plans and
guide critical aspects of water management including WSP rule changes (DOI,
2018c), consultation (DPI Water, 2015), licensing (DOI, 2018a), assessment
(WaterNSW, 2017), and metering (DOI, 2018b).
Most of these legislative instruments and policies apply on a state-wide basis, however
some, such as the Catchment Action Plans are regionally focused. These legislative
instruments and policies are critical in setting out the framework for water management in
NSW and in specific regions, which is then codified in WSPs for each region. This enables a
set of water management rules that aim to reflect the particular environmental, social and
economic situations of different regions.
The critical elements of the water management approach for each region covered by the
WSP are:

22
Water Management Act 2000 s 3.
23
The overall objective of the Water Management Act 2000 is to “provide for the sustainable and integrated management of the
water sources of the State for the benefit of both present and future generations”. See Water Management Act 2000 s 3.

41
  Prioritisation – the prioritisation of water to the environment and between water
users (Table 6) and purposes.
 Regional sustainability – the establishment of water extraction limits to protect
surface and groundwater sources at a macro level.
 Local impacts – the establishment of a set of rules to protect specific groundwater
dependent ecosystems, significant sites and other water users and the
implementation of take limits on particular bores to manage local impacts.

3.1.2 Prioritisation of water under WSPs

A critical element in the management of water under the WSPs is the prioritisation of water
to the environment and between water users and purposes. WSPs establish priorities of
allocation with environmental first, Basic Landholder Rights second, town water supply and
stock and domestic licences third, and all other licensed extraction for industrial and
commercial purposes fourth (Table 6).
Table 6: Prioritisation of water users under WSPs
Priority Water use Detail
1 Environmental water Water reserved for the environment. WSPs reserve ‘storage
water’. They also reserve a portion of groundwater recharge for
the environment before determining available licensed water
extraction.
2 Basic Landholder Rights (BLRs) BLRs have a higher priority over licensed water extraction and
 Stock and domestic rights cannot be limited through available water determinations.
 Harvestable rights However, they can still be subject to temporary water
 Native Title rights restriction order.24 WSPs must allow for BLRs before
determining available licensed water extraction.
3 Local water utility licences Priority is given to groundwater extracted for local and major
Major water utility licences utilities for town water supply and to licensed stock and
Stock and domestic licences25 domestic bores over other licensed water users.26 Licenced
extraction under these licences can be limited through
available water determinations and temporary water restriction
orders.
4 Water Access Licences Extraction under these licences can be limited through
available water determinations and temporary water restriction
orders. No distinction is made between the end use of the
water (agriculture, industrial production, etc). The WAL is also
separable from the property on which it is being used, so a
WAL can be traded within management areas

3.1.3 Regional sustainability and the WSP

WSPs share water and manage the sustainability of extraction from particular groundwater
sources through extraction limits. WSPs for most groundwater sources set extraction limits
through calculations of the expected groundwater recharge for each groundwater source
based on average rainfall, recharge rates, the sustainability index, the type of groundwater
source (e.g. fractured versus porous rock aquifer), and the distribution of high conservation
areas (e.g. National Parks) where water is excluded from the calculation.27 The groundwater
recharge is then used to calculate the UEL, the LTAAEL and Planned Environmental Water

24
Water Management Act 2000 s 324.
25 Under section 52 of the Water Management Act 2000, an owner or occupier of a landholding is entitled to take water from
an aquifer that is underlying their land for domestic consumption and stock watering, without the need for a water access
licence. However, a domestic and stock access licence may be required for the taking of water for domestic or stock watering
purposes where the land does not overly the aquifer from which water would be taken.
26
Water Management Act 2000 s 58.
27
The sustainability index (SI) considers both the environmental risk (based on the prevalence of high priority groundwater
dependent ecosystems, water quality, ecology, aquifer integrity and potential for mitigation) and social economic risk (based on
the dependence of local communities on the groundwater sources, including alternative water sources and the contribution of
groundwater to the local economy).

42
(PEW).28 These calculations and the related assumptions as applied in the development of
the WSP are described in detail in Appendix 3.
The UEL, LTAAEL, and PEW for each groundwater source are calculated with the intention
of reducing the risk of unsustainable extraction from particular groundwater sources in the
long term and to inform the distribution of water to achieve environmental, social and
economic wellbeing in the region according to the prioritisation of water uses as described
above:
 For fractured rock groundwater sources under the WSP, the UEL represents the
upper limit of extraction that could occur from a groundwater source under the WSP
each year taking into account the rainfall recharge over non-high environmental value
areas and the sustainability index for that groundwater source. 29
 For fractured rock groundwater sources under the WSP, the LTAAEL represents the
long term maximum average volume of water that can be extracted from a
groundwater source under the WSP each year taking into account the lower of the
UEL, or the current and estimated future requirements for groundwater and a
conservative buffer.30
 For porous rock groundwater sources under the WSP, the LTAAEL represents the
long term maximum average volume of water that can be extracted from a
groundwater source under the WSP each year taking into account the rainfall
recharge over non-high environmental value areas and the sustainability index for
that groundwater source.31
 The PEW represents the portion of water to be reserved for environmental
purposes.32
Extraction limits are a critical part of the ongoing adaptive management of groundwater
sources under the WSP. For example, during the term of the WSP, as the demand for
extractions changes over time and the understanding of the relevant groundwater source
improves, the LTAAEL may be increased to be closer to the UEL in those water sources
where the LTAAEL is below the UEL.33 Where growth in water take is assessed to have
increased more than 5% above the LTAAEL on average over a three-year period, the water
allocation may be reduced to less than 1ML per unit share to bring extraction levels back
down to the LTAAEL.34 This information will also critically inform reviews of WSPs at the end
of their 10 year period and the development of replacement WSPs. Adaptive management is
described in more detail below.

3.1.4 Local impacts and the WSP

WSPs include rules to address local impacts of water extraction. These rules set limits on
the proximity of groundwater extractions to certain assets including groundwater dependent
ecosystems (GDEs), other groundwater users, aboriginal heritage sites and major water
supply bores (Appendix 7). This approach allows local impacts to be addressed on a case-
by-case basis by:

28
Note that the UEL is not calculated for all groundwater sources.
29
Note that the sustainability index for high environmental value areas under the WSP is 0%, and therefore 0% of recharge
over high environmental value areas is considered available for extraction. The UEL is not calculated for porous rock
groundwater sources as the WSP indicates a high level of confidence in the calculation of the LTAAEL for these groundwater
sources. See WSP p.31.
30
These calculations and the related assumptions as applied in the development of the WSP are described in detail in Section
3.2 of the Initial Report.
31
Note that as high conservation areas are excluded from the extraction limit calculations, effectively 100% of recharge over
high conservation areas is reserved for the environment.
32
Planned environmental water (PEW), as defined in the WSP, comprises a portion of groundwater held in storage and a
portion of groundwater generated from recharge. At the commencement of the WSP, 100% of groundwater storage is reserved
as planned environmental water. However, this may be reduced to 99.998% for some porous rock groundwater sources in
accordance with the NSW policy for Managing Access to Buried Water Sources (NSW Office of Water, 2011).
33
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 60(1).
34
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 29(2)-(3)

43
 a risk assessment which identifies medium and high risk proposed extractions, and
if appropriate, an assessment that takes into account the particular hydrogeological
characteristics of the groundwater source and the local area (including proximate
sensitive areas).
The assessment of local impacts under the WSP is described in more detail in Chapter 4.

3.1.5 Adaptive management and the WSP

Adaptive management is a critical process in managing environmental sustainability of
groundwater extraction. Adaptive management relies on the collection of information (for
example, through monitoring and studies), the assessment of that information against
indicators of performance, and if necessary, changes in rules and management actions in
response to that information.

3.1.5.1 Monitoring and performance indicators

The WSP identifies monitoring as key to understanding whether the WSP is meeting its
environmental, social and economic objectives. To guide the appropriate range of monitoring
and to assess the outcomes of those monitoring activities, the WSP specifies a number of
performance indicators for all groundwater sources. These are intended to measure the
success of strategies in the WSP in reaching its objectives. The WSP performance
indicators are the changes in:
 groundwater extraction relative to the LTAAEL
 water quality
 the ecological condition of these groundwater sources and their dependent
ecosystems
 the extent to which domestic and stock rights and native title rights requirements
have been met
 the economic benefits derived from water extraction and use
 the extent to which water has been made available in recognition of the Aboriginal,
cultural and heritage values of these groundwater sources.35
However, the Background Document to the WSP notes that it is not practicable to monitor all
of these indicators in all groundwater sources covered by the WSP (DPI Water, 2016f). In
March 2011, the Department released the Environmental flow response and socio-economic
monitoring. North Coast - progress report 2009 while in 2006 a detailed hydrology report was
published on the Alstonville monitoring data (Green, 2006). More information is available in
relation to the Alstonville Groundwater source (Section 3.4.1).

3.1.5.2 Amendments and reviews of the WSP

WSPs allow for adaptive management responses at the level of individual licences 36 and
through amendment of some provisions of the WSP during its term. The Water Management
Act 2000 permits the Minister to amend the WSP if it is in the public interest to do so, or is in
accordance with the terms of the WSP, which allows for certain amendments, including to:
 Modify or add groundwater sources or management zones37
 Vary the amount of recharge reserved for the environment as planned environmental
water38
 Increase the LTAAEL39

35
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 12.
36
For example through available water determinations and temporary water restrictions discussed in further detail in Chapter 4
– Local Impacts.
37
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 58.
38
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 59.
39
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 60.

44
  Allow for the granting of supplementary water (storage) access licences under a
controlled allocation order in particular groundwater sources and the granting of
Aboriginal Community Development licences40
 Establish access rules for managing major utility access licences and for access
licences in particular groundwater sources41
 Add, remove or modify minimum distance rules42
 Impose rate and time restrictions on groundwater extraction 43
 Amend the map of high priority groundwater dependant ecosystems. 44
Most of these amendment provisions that allow for certain aspects of the WSP to be revised
require evidence (for example hydrogeological studies), that evidence supporting an
improved understanding of the system and justifying an adjustment to the WSP.
The WSP is also subject to a range of periodic and end of term reviews. For example, the
WSP is subject to an initial review and audit by the NRC within five years of enactment, and
a renewal or replacement review at the end of its 10-year life. These reviews focus on the
water management principles specified in the Water Management Act 2000,45 as well as the
environmental, social and economic outcomes.46 These review requirements are discussed
in further detail in Appendix 4 of the Initial Report.

3.2 SOURCES OF UNCERTAINTIES AND THEIR

MANAGEMENT IN THE WSP
The WSP estimates recharge (and ultimately the allowable extraction rates) for each
groundwater source covered by the WSP using the equations outlined in Appendix 3. This
section investigates the uncertainty associated with each parameter and what approaches
are taken in the calculations to manage this uncertainty, with the aim to provide a comment,
as far as possible, on the long-term sustainability of the extraction limits in the WSP as per
the terms of reference.
There are a number of relevant environmental parameters that introduce levels of
uncertainty to the extraction figures in the WSP including:
 Average annual rainfall
 Surface area and impact of confining geological layers
 Recharge rate
 Sustainability index.
This section analyses the uncertainties that have been identified through the review of the
WSP. The term uncertainty is used to recognise a range of possible values that could be
assigned to a given attribute, such as the recharge rate or the sustainability index. When
there is a substantial body of research (or lines of evidence) on a given topic, the uncertainty
is typically lower and a higher degree of confidence in the adopted values. Conversely, when
there is limited available data and/or research, the uncertainty may be high, thereby
including a large range of possible values.

40
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 61.
41
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 62.
42
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 63(a).
43
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 63(c).
44
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 63(d).
45
Water Management Act 2000 ss 5 and 43(2).
46
Water Management Act 2000 ss 43A(1),(3).

45
The section below describes how each of the terms used in Equation 1 (Average Recharge)
and 2 (UEL) were addressed and if improvements could be made.
Average Recharge (ML/yr) = Average Rainfall (mm/yr) x Area (km2) x Recharge Rate
(%)/100 (Equation 1)
UEL (ML/yr) = Recharge over non- high environmental areas (ML/yr) x SI(%)
(Equation 2)

3.2.1 Average annual rainfall

Annual rainfall is variable over the relatively large area (76,000 km 2) covered by the WSP.
Rainfall data used by the WSP was sourced from the Bureau of Meteorology (BOM) gridded
rainfall data (approximately 5 km2 grid) from 1901 – 2011. This rainfall model uses
algorithms to estimate a weighted average rainfall in each grid based on the observations at
the nearest BOM stations. This accounts for the spatial variability of rainfall throughout large
catchments and is based on the best available data. More information on the rainfall data
can be found on BOM (2015).
Rainfall data provided by BOM are considered reliable and the spatial distribution of BOM
weather observation stations in NSW is considered sufficient to capture the spatial
distribution of rainfall and represent the average rainfall over the WSP area. There are 94
BOM rain gauges currently operating in the Northern Rivers region and a further 229
historical gauges.

3.2.1.1 Climate change projection and its effects on rainfall

The earth’s climate is expected to change globally in the future (IPCC, 2013). The DPIE’s
Energy, Environment and Science Branch, formerly Office of Environment and Heritage, has
provided a detailed assessment of climate change effects on the North Coast region of NSW
(OEH, 2014). The report uses information from the NSW and ACT Regional Climate
Modelling project (NARCliM) to make climate change predictions. The NARCliM analysis
uses 12 predictive models and report what the majority of models agree on.
The report projects that the North Coast region will see increase in temperatures in the near
future (2020–2039) and far future (2060–2079), compared to recent years (1990–2009). The
warming is projected to be on average about 0.7°C in the near future, increasing to about
2°C in the far future. An increase in temperatures with climate change could lead to greater
or reduced evapotranspiration (evapotranspiration is a function of humidity, irradiance and
wind). If evapotranspiration increases, this would contribute to drying over some land areas.
With potentially more extreme rainfall (storms), but also longer dry spells in between, storm-
affected areas are likely to experience increased risk of flooding, while also seeing
increased risk of drought.
The majority of climate models agree that autumn and spring rainfall in the region will
increase in both the near and far future, and that winter rainfall will decrease for both
timeframes. For summer rainfall, the majority of models predict little change in the near
future and an increase in the far future. High intensity rainfall results in fast moving surface
water which may not have sufficient time to infiltrate the soil matrix to effectively recharge
groundwater aquifers. While rainfall may be predicted to increase over some time horizons, if
the increased rainfall is as a result of more frequent high intensity rainfall, recharge may not
increase.
Climate modelling results should be interpreted with care as large uncertainties are
associated with the direction of change in the region’s predicted rainfall. In a CSIRO and
Bureau of Meteorology report (Dowdy et al., 2015), projected changes in the region’s annual
rainfall over the next 20 years range from -15% to +10% relative to 1986–2005. That study
also concluded for the region that intensity of heavy rainfall events will increase (with high
confidence) and there will be longer periods of meteorological drought by late in the 21st
century (with medium confidence). However, natural climate variability will likely remain the

46
major driver of rainfall changes in the next decades. Regardless of the direction in rainfall
change, rainfall patterns in the region will change in the future which will have a direct impact
on recharge. This should be considered as part of future WSP planning.

3.2.2 Surface area and impact of confining geological layers

The method to determine the UEL in coastal fractured and porous rock groundwater sources
assumes a spatially uniform recharge rate as a percentage of rainfall, which is contrary to
the high variability in land use and cover, soils, geology and recharge processes that exists.
While the uniform rate assumption is justified given lack of data, care must be taken when
considering the groundwater source to ensure that the recharge rate is sufficiently
conservative to protect the groundwater source, yet reasonably approximated for the region
to provide a valid end point.
The WSP provides a surface area for high and non-high environmental value areas for each
groundwater source but does not describe how the surface area was delineated. Typically,
this information, as derived from a digitised aerial image and/or high-resolution satellite data,
is widely used and acceptable.
Modelling reports and hydrogeological reports (e.g. (Bilge, 2003), (Eco Logical, 2016, 2018a,
2018b) and (Kobus Argent, 2018)) suggest that parts of the relevant aquifers are at least
locally confined or partially confined in some areas. If the aquifer is confined or partially
confined, it is overlain by a layer of low permeability material that inhibits rainfall infiltration
over some parts of the aquifer, as illustrated in Figure 10. As a result, a confined aquifer with
the same surface area as an unconfined aquifer may receive less recharge for the same
direct rainfall. In most cases, unless there has been significant and detailed geological
mapping, the area where semi-confined aquifers are recharged is often uncertain. This is
addressed in the WSP by not permitting overlaps in the recharge areas of the confined and
unconfined aquifers defined in the WSP. However, the variable presence of other confining
layers (e.g. local confining clay layers) not being considered in the WSP introduces
additional uncertainty in the recharge estimates.
It is unclear whether the percentage recharge rate used in the WSP refers to only the
identified outcrops of the aquifers that make up the groundwater source or also the sub-
crops beneath a confining layer. The variations in connectivity (permeability) in the different
layers in the aquifer can cause local variations in recharge rates. Although, this is a
ubiquitous issue for recharge studies, integrating the many layers into one lumped aquifer
introduces some level of uncertainty as the fractured basalts are the dominant source for
bottled water extractions.
Developing a better conceptual understanding of the geological strata in the WSP, possibly
via a 3D geological modelling tool (e.g. Leapfrog Geo) where there is sufficient data would
help to reduce the level of uncertainty in the estimated recharge values and assist in
understanding whether the assumptions made in the WSP are appropriate. This would
require a large scale detailed geological mapping survey or the collation of the existing core
log data and geophysical measurements, where available. It is worth noting that this level of
detail has not been typically undertaken in other WSPs of a similar nature for easterly
flowing systems.

47
Figure 10: Recharge areas of confined and unconfined aquifers
Note: confined in red highlights; unconfined in purple highlights
Source: UNSW Sydney (2017)

3.2.3 Recharge Rate

The estimates of recharge provide the basis for Planned Environmental Water, LTAAEL or
UEL in the WSP and are a source of uncertainty in the WSP (Table 7). While uncertainty in
the recharge estimate may be acceptable if the estimates are suitably conservative, or there
is limited consequence as a result of possible error bands, it is important to recognise how
ambiguity may impact the plans. The systems considered in the WSP are considered to be
‘less highly-connected groundwater sources’ (DPI Water, 2016f), which means there is
limited recharge through creek beds, and thereby, only recharge through rainfall was
considered.
Recharge estimates in fractured rock are typically poorly understood due to the spatial
variability and complexity of these environments (Crosbie et al., 2010). This may result in
over (or under) estimation of recharge rates. Due to the practicality of treating the water
sources as being homogenous, it is considered necessary to be conservative in the recharge
rates assumed, to account for inherent inaccuracies and uncertainties in this calculation.
The values adopted for each of the groundwater sources considered in the WSP are
typically based on estimates dependent on the hydrogeological type. The Coastal Porous
Rock Rainfall Recharge Study showed that while the NSW default value of recharge rate in
coastal porous rock is 6%, the values reported in calibrated models across the state ranged
between 1 – 6%. This indicates that the NSW default value could potentially overestimate
recharge in the porous rock groundwater systems in scope of this Review. The Clarence
Moreton Basin adopted the value of 6% in lieu of other estimates.
Table 7: Recharge rates used in WSP for groundwater sources with bottled water extraction
Groundwater source Recharge used in WSP (% of long-term rainfall)
Clarence Moreton Basin (porous rock) 6
New England Fold Belt Coast (fractured rock) 4
North Coast Volcanics (fractured rock) 8
Alstonville Basalt Plateau (fractured rock) 8

48
More accurate estimates of rainfall recharge could be developed through multiple lines of
evidence including field data analyses such as chloride mass balance calculations, water
table fluctuation or base flow analyses. As the recharge rate is integral to the allocation of
the LTAAEL and planned environmental water, the impact of the adopted values warranted
further investigation and are considered further by the Review at Section 3.3.

3.2.4 Sustainability Index

The WSP, which is based on recommendations within DPI Water (2016f) and DPI Water
(2015), acknowledges that regional estimates of recharge for large aquifers is not an exact
science. As such, the WSP highlights that due to this uncertainty a precautionary approach
is warranted. DPIE Water has advised that a precautionary approach was subsequently
formulated using 0% recharge estimates for high environmental value areas, with no
allowance for recharge other than direct rainfall and a sustainability index, which is further
discussed below (DOI Water, 2019).
The sustainability index is a simple method for risk accounting that endeavours to manage
the balance between environmental risk, economic and social growth. While the allocation of
high, medium or low risk is subjective, the index is a measure that can be applied to all
catchments with relatively limited information. Further work could be undertaken to assess
whether the risk ratings given to specific catchments are appropriate, but the index appears
to be a cost and time effective means to protect resources where limited information is
available.
It would be useful to undertake sensitivity testing of the index to assess the implications of
changing conditions or additional research that may result in a change in the risk ratings.
Where this may result in the aquifer being over-allocated, more research may be required.
Estimates of the risks associated with each groundwater source could be better detailed as
information comes to hand.

3.3 TESTING THE WSP RECHARGE RATES AGAINST

STUDIES AND OTHER APPROACHES
3.3.1 Relevant literature on recharge rates
A review of literature was undertaken to compare the recharge rates adopted in the WSP
with existing studies in the same or comparable areas. This section provides an overview of
the various approaches to calculating recharge rates in the region presented in the literature.

3.3.1.1 Coastal Porous Rock Rainfall Recharge Study

A primary source of information about recharge across NSW is the Coastal Porous Rock
Rainfall Recharge Study prepared for NSW DPI Water (EMM, 2015). The data in that study
are estimated from groundwater modelling studies, none of which are in the Tweed or
Alstonville areas, or in the groundwater sources of interest, but rather the report considered
advice from experts through interviews and a workshop. It also considered the Bioregional
Assessment program (Raiber et al., 2016), which was then unpublished but in the process of
undertaking recharge estimates for the Clarence Moreton bioregion.
The Coastal Porous Rock Rainfall Recharge Study recommended adopting 6% for the
Clarence Moreton Basin groundwater source. It recommended that the ongoing work of the
Bioregional Assessments should be considered (as published in 2016). Further, in high-risk
locations (i.e. where a groundwater dependent ecosystem could be impacted or a surface
water source influenced) and high-demand locations, the report recommended detailed
recharge investigations.

49
3.3.1.2 The Bioregional Assessment
The Clarence Moreton Bioregion sits on top of the geological Clarence Moreton Basin. The
Basin includes the Bundamba Group, overlain by the Walloon Coal Measures, overlain by
the Lamington and Main Range volcanic rocks and alluvial aquifers. Other minor formations
are present.
The Lamington Volcanics geological group encompasses the North Coast Volcanics. Both
the Lamington Volcanics and Alstonville Plateau aquifers are fractured basalt aquifers within
the Cenozoic basalt group. Based on this geology, recharge estimates for the Lamington
Volcanics outcrops may be useful for North Coast Volcanics and Alstonville Plateau
groundwater sources (although differences in climate, soils and vegetation are expected to
play a dominant role in controlling recharge rates).
The Bioregional Assessment of the Clarence Moreton Bioregion used the chloride mass
balance method to produce long-term average recharge estimates. The results from the
Bioregional Assessment of the Clarence Moreton Bioregion are provided in Figure 11
(Raiber et al., 2016). The results are based on 3632-point estimates of groundwater chloride
at points shown in Figure 12 below. The recharge estimates developed for these points are
interpolated and extrapolated spatially over the bioregion, based on the identified
relationship between long-term average recharge and long-term average rainfall, as detailed
in Figure 13.
Averaging the 172 point estimates of recharge available for the Lamington Volcanics within
the Bioregional Assessment data set, results in a 16% recharge rate. Based on Figure 11
and Figure 13, this appears to represent the spatially interpolated average for the Lamington
Volcanics (the actual spatially interpolated value is not available from the report or the
publicly accessible data set).
It is worth noting that the chloride measurements shown in Figure 13 provide minimal
coverage of the groundwater sources of interest. Although 16 % is likely to be the best
available estimate for the North Coast Volcanics and Alstonville Plateau groundwater
sources, it cannot be used with confidence. Furthermore, while the chloride mass balance
method is an established approach for approximating recharge. However, various
assumptions inherent within the method mean that it is not generally considered an accurate
approach. As such, these results are best complemented with alternative methods and
should not be used as a single line of evidence.

50
Figure 11: Mean recharge estimates for the Clarence Moreton bioregion
Source: Raiber et al. (2016)

51
Figure 12: Data used for estimating recharge for the Clarence Moreton Bioregional assessments
Source: Raiber et al. (2016)

52
Figure 13: Relations between long-term average rainfall and chloride mass balance estimates of recharge
for the hydrogeological units of the Clarence Moreton bioregion - (a) the Cenozoic Volcanics (Lamington
Volcanics and Main Range Volcanics), (b) Walloon Coal Measures, (c) other sedimentary rock formations.
Note: The black line is the line of best fit through the data points, the dashed grey line is recharge as half of rainfall and the blue
line is the range of annual average rainfall within the bioregion for the surface geology class Source: (Raiber et al., 2016).

The other groundwater source of primary interest, the New England Fold Belt Coast, is not
clearly represented by the Clarence Moreton bioregion, as it underlays the Clarence Moreton
Basin.
The Review concluded from the Bioregional Assessment:
 The 6% assumed in the WSP for the Clarence Moreton Basin groundwater sources
is a reasonable basin-average estimate. However, there is large spatial variability of
recharge within the Basin, and 6% should not be assumed a safe value for particular
aquifers or locations.
 The results for the Lamington Volcanics indicate a spatial average recharge value of
approximately 16% of rainfall. This is the best available published value to support
the WSP values for fractured volcanic basalt aquifers including the Alstonville
Plateau and North Coast Volcanics. However, the value of 16% is subject to the
various uncertainties in the chloride mass balance method and is based on data
points that are well outside the Tweed and Alstonville areas of interest.
 The other groundwater source of interest, the New England Fold Belt Coast, is not
represented in the Bioregional Assessment results.

3.3.1.3 The University of Queensland recharge study

The University of Queensland undertook a study “Recharge estimation in the Surat Basin”
(West et al., 2018). From discussion with the report authors, the study included recharge
estimates in the Main Range Volcanics. These fractured basalt aquifers may be considered
comparable in hydrogeological properties to the fractured rock aquifers of interest. However,
the Main Range Volcanics are approximately 150 km north-west of the Tweed Valley, in a
much drier region (approximately 700 mm annual average rainfall versus 1800 mm in the
upper Tweed). Chloride mass balance and baseflow analysis methods were also used in this
study and are provided below for comparison.
The recharge using the two methods, spatially averaged over the Main Range Volcanics
subcrop, was estimated to be 0.7% and 0.9% of the long-term rainfall. The low values were
attributed to relatively impermeable soils. Another recent estimate using the chloride mass
balance method (DNRME, 2016) provided an estimate of 1.2%. Further, a broader review of
previous recharge studies in this region (West et al., 2018) suggested that recharge results
range from 1.9 to 2.3% of rainfall. This range of estimates (from 0.7% to 2.3%) are broadly
consistent with the results found for the Main Range Volcanics in the Bioregional

53
Assessment of the Clarence Moreton Bioregion (the point estimates provided for the
Bioregional Assessment ranged from 0.2 to 6.3% with an average of 2.4%). The relatively
low percentage values in the Main Range Volcanics compared to the Lamington Volcanics
are expected due to the high sensitivity of recharge to the rainfall and potentially attributed to
different soil and vegetation covers. While it can be concluded that The University of
Queensland recharge study results are not comparable to the aquifers of interest for the
Tweed and Alstonville areas, the results illustrate the range of approaches that may be
undertaken to estimate recharge in hydro-geologically comparable aquifers.

3.3.1.4 Australian Landscape Water Balance

The Australian Landscape Water Balance website of the Bureau of Meteorology includes a
Australian Water Resources Assessment Landscape model (AWRA-L) which is the near-
surface component of Bureau of Meteorology’s national water balance modelling system.
One of the AWRA-L outputs is deep drainage, which is the water moving vertically
downward at a depth of 6 m. In most situations, this can be considered to be an estimate of
groundwater recharge.
AWRA-L focusses on surface hydrology and is calibrated to river flow gauges. Its accuracy
for recharge estimation has been assessed by CSIRO although not specifically for aquifers
of interest (Shi, Vaze, & Crosbie, 2015). It was concluded that “overall, the groundwater
recharge assessment in this report indicates that AWRA is able to provide reasonably
reliable spatial and temporal (e.g. annual) estimates of recharge across Australia”.
Deep drainage results for the Tweed River catchment and surrounding region are shown in
Figure 14. The mapped results are for 2018, which has a modelled deep drainage close to
the long-term median value, therefore may be considered typical. The modelled median
deep drainage for the catchment is approximately 40 mm/year, or 2.2% of long-term average
rainfall. Due to the grid sizes, it is not practical to pick out particular values for the
groundwater sources of interest.
It may be concluded from this study that surface water modelling tools, which represent the
effects of soil and vegetation and calibrated to surface flows, may give significantly lower
recharge rates than assumed in the WSP. The difference between the recharge rates
estimated by AWRA-L and the WSP suggest that additional investigations are needed to
ensure that the rates used in the WSP are not over-estimated or the AWRA-L are not
underestimated.

54
Figure 14: Screenshot of AWRA-L deep drainage results, showing map for typical year 2018
Source: BOM (2019)

3.3.1.5 The National Review of Recharge Studies by CSIRO

In 2010, CSIRO conducted a review of previous recharge studies in Australia (Crosbie et al.,
2010). The NSW studies included in the review were for Coastal Alluvium, Inland Alluvium
and the Western Slopes of the Great Dividing Range. None of these are relevant to the
groundwater sources of interest. Fractured rock aquifers were found to be very poorly
represented in the available data.
Vegetation type and soil type were found to be critical determinants of recharge. Relating
recharge to surface geology had mixed results. It may be concluded from this review that
there is likely to be high uncertainty in generalising recharge rates from surface geology (as
used in WSP), especially in fractured rock aquifers.

3.3.2 Application of methods to local area

There is a range of different techniques available to quantify recharge, with the choice of
method dependent on the goal of the study, the spatial and temporal scales, and level of
information available. Each approach has uncertainties so often multiple techniques are
used to increase the reliability of the outputs (Scanlon, Healy, & Cook, 2002)
Two methods were used to estimate groundwater recharge within this Review. These are
application of the chloride mass balance method to chloride measurements from the
proponents’ hydrogeology reports; and filtering baseflow from available surface flow
measurements. Both methods have considerable assumptions and uncertainties.

55
3.3.2.1 Application of chloride mass balance
The chloride mass balance method (Anderson, 1945) is a simple method for estimate
groundwater recharge from rainfall based on the following assumptions (Wood, 1999):
 Chloride in groundwater is only sourced from rainfall
 Chloride is conservative in the system (no sources or losses).
 The chloride flux does not change over time (steady state conditions).
 There is no recycling of chloride in the system.
Then,
1000 𝐷
𝑅=
(𝐶𝑙 − )𝑔𝑤
where R: recharge [mm/year];
D: Chloride deposition rate [kg/m2/year];
(Cl-)gw: Chloride concentration in groundwater [kg/m3].
The chloride mass balance method was applied to the area under study using values of
chloride deposition estimated from the Bioregional Assessment (assuming no Chloride
losses) and chloride concentration in groundwater from the proponents’ hydrogeology
reports (when available).
Long term recharge estimations obtained were averaged per aquifer system and are
presented in Table 8. Table 8 also shows WSP values for comparison.
Table 8: Groundwater sources, recharge rates used in WSP and recharge estimated locally
Groundwater source Recharge used in WSP Recharge estimated locally (% of long-
(% of long-term rainfall) term rainfall)

New England Fold Belt 4 31

Coast

Clarence Moreton Basin 6 10

Alstonville Plateau 8 (*)

Note (*): No local data was available for the Alstonville Plateau.

The estimates based on the local chloride measurements are considerably higher than those
used in the WSP, however, it should be noted that:
 there is uncertainty in these estimates because they are based on a very small
number of chloride measurements
 if losses of chloride due to surface runoff are considered, these values would be
lower, but there were no available data to include them
 if the source aquifers are confined, these estimates may be interpreted as estimates
at the outcrop of the aquifers
 if the source aquifers are not confined the recharge may come from overlying
aquifers and/or surface water sources.

3.3.2.2 Baseflow filtering as an estimate of recharge to the Northern Rivers

bottled water industry
Baseflow is the sustained contribution to river flows. Conceptually, baseflow is often
considered to be the groundwater contribution to river flows, although may contain other
sustained flow contributions. Where baseflow may be assumed to be the groundwater
contribution, and considered as the outflow from the groundwater reservoir, it provides an
alternative estimate of recharge, although will only provide estimates of recharge to aquifers
that discharge upstream of the gauge. Long-term stream flow gauges are needed, like those
maintained by the DPIE Water.

56
At catchment scale, the stream flow components can be broadly grouped in classes based
on the different orders of magnitude of the subflow responses to rainfall. Most stream flow
series show quick flow and slow flow components (Willems, 2009). The slow flow component
corresponds to baseflow.
Baseflow filtering is a data analysis technique that allows users to numerically separate out
baseflow contributions from total measured/gauged river flow (Tallaksen, 1995). The method
excludes any recharge that does not appear as baseflow at the gauge, and thus may
exclude recharge that feeds deep groundwater systems and excludes recharge through the
river bed that does not re-emerge before the gauge.
Baseflow filtering was used to estimate groundwater recharge at a number of flow gauging
stations within the area of interest. The filtering method used is described in Appendix 4.
Figure 15 shows the location of flow monitoring points and groundwater sources.

Figure 15: Flow monitoring points in the Northern Rivers region and groundwater management areas

57
Table 9 shows the flow gauges in the Tweed and Alstonville Plateau Areas. The daily flow
time series for monitoring points in Figure 15 and Table 9 were obtained from the BOM.
Rainfall is from daily SILO (Scientific Information for Land Owners) data downloaded from
(Queensland Government, 2019), SILO uses mathematical interpolation techniques to infill
gaps in gauged rainfall time series and constructs spatial grids, and for the current purpose
is equivocal to the alternative BOM gridded rainfall data. Rainfall falling over each gauged
catchment was averaged from all SILO cells within or overlapping the drainage area.
Table 9: Flow gauges in the Tweed Area and Alstonville Plateau
Code Area Name Latitude Longitude Catchment
Area km2
201001 Tweed Oxley River at Eungella -28.3537 153.2931 213

201005 Tweed Rous River at Boat Harbour No. -28.3096 153.336 111
3
201012 Tweed Cobaki Creek at Cobaki -28.200871 153.458926 10

201015 Tweed Tweed River D/S Palmers Road -28.433857 153.291908 156
Crossing
201900 Tweed Tweed River at Uki -28.413522 153.334927 275(**)

202001 Brunswick Brunswick River at Durrumbul -28.531174 153.458183 39 (*)

(Sherrys crossing)
202002 Brunswick Burringbar Creek at Burringbar -28.43774 153.476761 39 (*)

203012 Alstonville Byron Creek at Binna Burra -28.706611 153.497897 39

203014 Alstonville Wilsons River at Eltham -28.755574 153.394827 223

203057 Alstonville Houghlahans Creek at upstream -28.801169 153.494846 10 (*)

Teven
203059 Alstonville Maron Creek at Graham Road -28.874925 153.36741 39 (*)

(*) Catchment areas with * were estimated (**) Influenced by upstream reservoir so lower value for baseflow analysis

Appendix 4 presents aggregated monthly baseflow for the available period of data for these
flow gauges, together with the rainfall over the catchment.
For each monitoring point, aggregated annual values are also presented in millimetres for
rainfall, total streamflow (Q) and baseflow (BF). Ratios BF/Rainfall and BF/Q are calculated,
and their tendency is analysed by calculating a 5-year moving average. Table 10 shows the
average BF/rainfall ratios over the available data periods.
Table 10: Fraction Baseflow/Rainfall for flow gauges for period 1960-2018
Code Area No of years with information Average Baseflow/Rainfall
201001 Tweed 59 0.08
201005 Tweed 37 0.09
201012 Tweed 36 0.10
201015 Tweed 9 0.05
201900 Tweed 37 0.05**
202001 Brunswick 47 0.05
202002 Brunswick 9 0.08
203012 Alstonville 40 0.22
203014 Alstonville 58 0.17
203057 Alstonville 8 0.14
203059 Alstonville 7 0.14
** Influenced by upstream reservoir so lower value for baseflow analysis

58
From the figures in Appendix 4 and data in Table 10, it is observed that fractions of
baseflow:rainfall and baseflow:total streamflow are variable annually. Long-term
baseflow:rainfall ratios are in the range of 5-15% in the Tweed, 5-10% in Brunswick and 10-
25% in the Alstonville system. These values may be interpreted, with considerable
uncertainty, as recharge rates of aquifers that discharge upstream of the gauging locations.
Given that these values are likely to be under-estimates of total recharge because they only
capture the groundwater that is discharged above the gauge, and due to the uncertainty in
the filtering method, it is concluded that there is no evidence here that the WSP values of
recharge are not conservative

3.3.3 Impact of reducing the recharge rates

Despite the available data or research, there is likely to always be some uncertainty in the
extraction limits and planned environmental water calculated in any WSP. One approach to
assess whether the current level of uncertainty is acceptable is to undertake sensitivity
analyses of the results. Sensitivity analyses examine ‘what-if’ cases for a range of possible
values for an attribute in question. In the case of the WSP, a sensitivity analysis could be
undertaken to see whether a change in the recharge or sustainability index might result in
the aquifers being stressed at the current level of extraction or at the prescribed level of
allowable recharge.
For example, in a scenario where recharge was just 20% of that calculated in the WSP,
however there was no change to the LTAAEL or current assignment of water, the recharge
amount reserved for the environment (RRE) would be as shown below in Table 11.
Table 11: Hypothetical example scenario - reducing recharge to 20% of the estimated recharge in the
WSP
Groundwater Hypothetica LTAAEL Current RRE* RRE as a Assigned
Source l Recharge - (ML/yr) Assigned (ML/yr) percentage of Water as a
20% of WSP Water estimated percentage of
value (ML/yr) recharge recharge
(ML/yr)
New England 396,000 60,000 35,468 336,000 85% 9%
Fold Belt
Coast
North Coast 62,000 13,000 5,907 49,000 79% 10%
Volcanics
Alstonville 10,016 8,895 8,895 1,121 11% 89%
Plateau
Clarence 115,200 300,000 4,562 -184,800* -160%* 4%
Moreton Basin
*RRE is calculated as Average Recharge (ML/yr) - LTAAEL. A negative RRE implies the extraction limit is greater than the
expected recharge.

As shown in Table 11, if the recharge rate was reduced to 20% of the WSP value, the RRE
in the New England Fold Belt Coast and the North Coast Volcanics remains relatively high
(approximately 80%) and the present LTAAEL is likely to remain reasonable. In the
Alstonville Plateau, RRE (as a percentage of recharge) would be significantly reduced if the
recharge rate had been overestimated. However, there is a network of monitoring bores
(piezometers) within the Alstonville aquifer which can be inspected to assess whether there
has been any long term changes in water levels throughout the groundwater system. The
availability of this data provides a way to scientifically evaluate long term trends in aquifer
water level and to identify signs of the aquifer becoming stressed (e.g. systematic decline of
water levels over a significant time period).
Table 11 also shows that the reduced recharge would result in the LTAAEL being larger than
the annual recharge. However, extraction in this region is presently very low. If there is
significant concerns that the recharge rate in this region has been overestimated, the
LTAAEL could be reduced without significant impacts to current license holders.

59
3.4 MONITORING DATA FROM ALSTONVILLE BASALT
The NSW Government operates 29 monitoring piezometers on the Alstonville Plateau
(including two in North Coast Volcanics) which continuously measure water levels in various
locations at multiple depths throughout the region. Analysis of this data provides a useful
way to understand the groundwater system, as well as to identify periods in which water
supplies in the aquifer may be stressed.
This section is separated into two distinct parts based on the time of monitoring:
 Pre 2009, a number of reports were released that analysed the existing monitoring
data to date;
 Post 2009, the Review hasn’t found an updated status report on the groundwater
levels. As such, monitoring data was accessed by the review team to provide
comment on groundwater levels in the Alstonville Plateau over the last decade.

3.4.1 Alstonville monitoring network and aquifer levels to 2006 and 2009
In 2006, the Department published a status report on the groundwater levels in Alstonville.
At the time, the monitoring network consisted of 11 monitoring piezometers measuring both
the deep and shallow aquifers, across five sites at Alstonville. This report included the
development of conceptual models for the Alstonville GW sources showing the system
comprises two major aquifers – a shallow unconfined aquifer (less than 50 metres deep) and
a deeper semi-confined/confined aquifer (generally >50 m) (Green, 2006; DECCW Water,
2011).
Green (2006) stated that the shallow unconfined aquifer is rapidly recharged by rain, while
the deeper semi-confined/confined aquifer takes longer to recharge after rainfall events.
While the deep aquifers do experience periods of drawdown and recovery, the recovery
period can be substantially longer than that observed in the shallow aquifers, most likely due
to the limited and slow recharge processes associated with these semi-confined systems.
Green (2006) specifically addresses groundwater levels during a drought period in the early
2000s. During this period, surface water runoff was minimal and groundwater extractions
were high. As a result, the deep groundwater levels were at some of their lowest observed
levels and the aquifer was noted in several reports as stressed over this period of time
(Green, 2006; DECCW Water, 2011). While limited monitoring data is available to quantify
groundwater extractions during this period, Green (2006) speculated that this was likely
associated with the over-extraction of groundwater due to limited surface water availability
during the drought. The drawdown covered an area more than 3 km wide with drawdown
levels varying from 8 to 19 metres (Green, 2006). An embargo on new licences was also
imposed on the Alstonville aquifer in 2000 to prevent further stress on the aquifer (DPIE
Water 2019, pers comm., 30 October).
From early 2003 onward when the drought ceased, the levels in the deep aquifer system
started to recover. From that time until the data was analysed for the 2006 status report, the
deep aquifer levels rose by 8 to 25 metres across the aquifer (Green, 2006; DECCW Water,
2011). By 2009, the Department noted that groundwater levels had recovered to levels seen
in the 1980s when monitoring commenced. In addition to rainfall slowly providing recharge to
these aquifers, it is likely that extraction of the groundwater reduced significantly when
surface water became more plentiful. However, as groundwater use in the Alstonville region
is largely un-metered, conclusions about the impact of pumping and climate are difficult to
differentiate in the deeper aquifers.

3.4.2 Alstonville monitoring data between 2006 – 2018/2019

In 2005-2006, after completion of the groundwater status report discussed in the section
above, the Department expanded the monitoring network to consist of 29 monitoring
piezometers across 13 sites measuring both shallow and deep aquifer levels. While no

60
 additional reporting had been undertaken analysing the Alstonville Plateau data since the
detailed 2006 report (Green, 2006) and a brief update in 2009 (DECCW Water, 2011),
groundwater level data has been continually collected on the expanded monitoring network.
A basic analysis of the data from the previous thirteen years (2006-2019) has been
undertaken by the Review to provide insight into the state of the Alstonville aquifer today and
the impact of environmental variables (e.g. rainfall and seasons) on groundwater levels in
the region. Data was accessed for all 29 monitoring piezometers in the Alstonville Plateau.
The location of each piezometer (including depth) is summarised in Figure 16 and Table 12.
In summary, the analysis detailed below shows, during the period from 2006 onward, the
readings from shallower piezometers, tend to be more variable and show increases following
periods of rainfall. The deeper piezometers (depths around 50m+) tend to be more stable,
showing a tendency towards a steady upward trend over time. The analysis below has
identified lagged rainfall as an important variable for understanding piezometer water levels
in the Alstonville Plateau, especially in shallow piezometers and deeper piezometers that are
indicating connection to surface waters and upper aquifers. There were no strong seasonal
effects shown on water levels or pressures, but almost all the piezometers showed temporal
effects.
.! .! .!
.!
.!

.! .!
*# .!
.!
.! .! .!
.! .! *#
.!.!
.!
.! *#
*#

.!
.!
.!
.! .! *# .!
*#
.! *#
*#*# .! *#
.! *# *# ALSTONVILLE .!
.!
*# .! .! .!
.!
*#
.!
*#
.!
.!
.!
.!

± 0 1 2 4 6

Data sources: NSW Department of Planning,

Industry and Environment | Water;
8
KM
.! Northern Rivers Region Boundary Clarence Morton Basin Groundwater Source
New England Fold Belt Coast Groundwater Source
Licence data extracted from WLS database 7/12/2018 *# DPIE Groundwater monitoring bores
Licences attributed to works approvals where possible Alstonville Basalt Plataeu Groundwater Source
.! BOM rainwater gauge historic (closed)
Alstonville-Tuckean Management Zone
.! BOM rainwater gauge status open Bangalow-Wyrallah Management Zone
Map produced by NSW Department of Planning, Industry and Environment | Water
October 2019

Figure 16: Map of Alstonville showing the 13 sites for the Alstonville monitoring piezometer network
:

61
Table 12: DPIE groundwater monitoring piezometer network at Alstonville
1
Shows depth classification (shallow, medium or deep) , year drilled, the amount of variability of the levels
explained by rain, season and time through a GAM analysis.
1 #
Piezometer Hole Pipe Depth Year Lag Variance Explained (post 2006)
&
ID drilled (days)

Rain Season Time

GW040999 1 1 Shallow 2005/6 30 57%* 8%* 10%*

GW040999 2 2 Deep 2005/6 40 1% 7% 98%*

GW041001 1 1 Shallow 2005/6 30 68%* 8%* 8%*

GW041001 2 2 Deep 2005/6 40 3% 2% 99%*

GW041002 1 1 Deep 2005/6 200 61%* 5% 75%*

GW041003 1 1 Shallow 2005/6 30 39%* 11%* 8%

GW041003 2 2 Deep 2005/6 180 63%* 5% 19%*

GW041004 1 1 Shallow 2005/6 150 50%* 11%* 8%

GW041004 2 2 Deep 2005/6 150 51%* 11%* 8%

GW041005 1 1 Shallow 2005/6 30 53%* 3%* 5%*

GW041007 1 1 Medium 2005/6 100 55%* 7% 19%*

GW041007 1 2 Deep 2005/6 120 51%* 7% 22%*

GW041008 1 1 Shallow 2005/6 120 56%* 10% 37%*

GW041008 1 2 Deep 2005/6 150 58%* 16%* 36%*

GW041000 1 1 Deep 2005/6 280 17% 2% 94%*

GW041000 1 2 Deep 2005/6 280 20%* 3% 95%*

GW081005 1 1 Deep 1999 280 2% 1% 99%

GW081006 1 1 Shallow 1999 120 44%* 5% 42%*

GW081002 1 1 Deep 1999 280 45%* 7% 71%*

GW081003 1 1 Medium 1999 150 66%* 14%* 35%*

GW081004^ 1 1 Shallow 1999 NA NA NA NA

GW081000 1 1 Shallow 1999 150 47%* 4% 15%

GW081001 1 1 Deep 1999 280 2% 11% 99%

GW036702 3 1 Shallow 1987 150 49%* 4% 8%

GW036702 2 2 Shallow 1987 280 29%* 21%* 18%*

GW036702 1 4 Deep 1987 NA NA NA NA

GW036701 1 1 Shallow 1987 240 33%* 11% 41%*

GW036701^ 1 4 ?? 1987 NA NA NA NA

GW036701 2 2 Deep 1987 280 8% 3% 97%

& Lag associated with the best predictive power

# variance explained corresponds to the percentage of variability in the data that can be explained by a GAM model
that includes all three factors (rain, season and time) compared to a model that leaves each respective factor out.
* Significant at p<0.05 based on a likelihood ratio test with block bootstrap
^ This piezometer had only limited data and none past 2006
1
Classifcation of shallow, medium and deep provided by DPIE Water. Generally DPIE Water are classified shallow with screen
interval up to approximately 25 m, medium 25-40 m and deep 40m onwards, with the deepest having a screen interval 150-168
m. The conceptual model indicated the shallow aquifer runs to approx. 50 m with the deep aquifer >50 m

62
The monitoring data alone will not separate the effects of extraction from climate, but
analysis of rainfall patterns over the period can give some indication of the impact of climate
variability on groundwater levels. This complements the qualitative analysis of how long-term
river baseflows relate to climate in the Initial Report, which found a strong and consistent
influence of climate but no evidence of other influences.
The analysis in this section is not intended to be an in-depth hydrogeological analysis, but
rather an empirical statistical analysis to assess the long term changes in water levels in
each piezometer. The water levels in the piezometers over the last decade are shown in
Figure 17 and Figure 18. Piezometer levels here refer to the distance (in metres) from the
measuring point to water.
Figure 17 and Figure 18 also shows interpolated rainfall sourced from the SILO database
(Queensland Government, 2019) to improve the understanding of the relationship between
rainfall and groundwater levels. As evident in these two figures, the groundwater level data is
not continuous at every monitoring station, with several extended gaps in a number of the
monitoring wells, especially in the earlier periods. There are also several notable dips in the
levels from two deep pziemometer. The Review was unable to confirm the cause, but it was
speculated that these may have been due to measurement error, periods when the data
from the logger was being downloaded or the logger was down.
The shallower piezometers are more variable and appear to be recharged regularly with
rainfall. Figure 18 shows that the deeper piezometers (depths around 40-50m+) tend to be
quite stable, with a tendency towards a steady upward trend over time.
The Review team undertook some statistical analyses to consider these observations in
more detail. In particular, a statistical technique called the Generalized Additive Model or
GAM (Hastie & Tibshirani, 1990) was used. GAMs have become very popular as an
exploratory data analysis tool that allows one to assess the relationships between variables
of interest without having to impose strong modelling assumptions such as linearity. They
are popular for modelling environmental data where non-linear effects arise often and where
it is desirable to use analysis tools that let the data drive the results rather than imposing
strong assumptions.
For the purpose of this investigation, the Review team developed GAM models that predict
piezometer level (in metres below measuring point) as a function of time, rainfall and
season. The purpose of this analysis is to investigate how much of the variability in
groundwater levels in each piezometer is associated with each of these three components.
The Review explored a range of options for how to best incorporate rainfall into the model. It
found that ‘lagged rainfall’ averages, where rainfall is reported for each day as the average
rainfall over the previous x days (where x was allowed to range from 10 days up to 240
days) provided a better explanation of variability in the piezometers than daily rainfall. While
other lagged rainfall distributions may be suitable, for the purpose of this simplified analysis,
the unweighted average over the x days is used. For each piezometer, the Review team re-
ran the GAM models to identify the most appropriate lagged rainfall average(x). The sixth
column in Table 12 shows the best identified rain lag variable value for each piezometer.
Following this analysis, the extent to which each component contributes to the variability of
each piezometer was calculated. To ensure that the significance tests were appropriately
adjusted for autocorrelation induced by the time-series nature of the data, a technique called
the block bootstrap (Kunsch, 1989) was used, which has been implemented using the boot
package in the statistical programming environment, R (Canty & Ripley, 2019). This analysis
is an alternative to using the Seasonal Kendall Trend test, which is popular in hydrogeology.
Table 12 shows the percentage variability in measured piezometer levels that can be
explained by lagged rainfall, season and time (shown in columns 7, 8 and 9 respectively).
The data is analysed from 2006 onward in order to boost statistical power to detect effects.
These figures were computed by running models leaving out each factor and comparing the

63
deviance explained by that model to the deviance explained by the model with all three
factors in included. Statistical significance was assessed through use of a likelihood ratio
test, using the block bootstrap to adjust for autocorrelation. Numbers that are statistically
significantly different from 0 at p<0.05 are indicated by an asterix. A higher percentage
indicates that the water levels in that piezometer are more highly correlated to that particular
variable.
Table 12 suggests quite a lot of variability between piezometers and locations. In some
cases, for example piezometer number GW41008, shown in Figure 19, rainfall averaged
over the previous 120 to 150 days could explain quite large amount (>55%) of the observed
variability in day to day piezometer levels. The strong association between the lagged rainfall
and measured piezometer level is visually quite striking in Figure 19. In general, lagged
rainfall was found to explain 30% - 70% of variance in shallow piezometers, which is
consistent with the findings of Green (2006) who noted that these shallow aquifers are
rapidly recharged through direct rainfall. The Review had anecdotal information that
groundwater extraction may be higher during dry periods; this analysis indicates the impact
of low rainfall periods on groundwater levels may be from a variety of factors.
The correlation between rainfall and water level variance in deeper piezometers was less
consistent. Rainfall was observed to account for more the 50% of variability in some deeper
piezometers (such as GW410002_1_1, GW410003_2_2, GW410004_2_2). This indicates
that some of the deeper piezometers have some connection with surface waters (possibly
through upper aquifers) and are unlikely to be within confined aquifers. However, rainfall
effects were less significant in other deep piezometers (see GW040999_2_2,
GW041001_2_2, GW041000_1_1, GW041000_1_2, GW081005_1_1, GW081001_1_1,
GW036702_1_4, GW036702_2_2). This suggests that these aquifers are confined (or semi
confined) as per the observations of Green (2006). Further work could be undertaken, based
on this analysis and existing geological mapping, to potentially identify confined aquifers, but
this analysis has not been undertaken as part of this review.
The majority of the piezometers showed no significant seasonal effect after adjusting for
rainfall. However, almost all the piezometers showed significant temporal effects. By this we
mean that including time in the model provided a statistically significant improvement
compared to models that did not include time. We undertook further analysis to determine
whether these time effects could be described as linear or non-linear. For the most part, we
found significant non-linear effects, though visually these effects were not strong. For the
majority of deep piezometers, the overall time-effect explained the majority of variation in
observed levels. The significance of the time variable may reflect other factors not
incorporated into our modelling, for example changes in patterns of extraction or other
aspects of rainfall not adequately captured with the lagged rainfall variable tested in the
GAM analysis.
The three panels in Figure 19 provide more detail on the GAM analysis for a single
piezometer (GW041008) to further illustrate the outcomes of this analysis. Panel a) reveals a
strong and fairly linear relationship between lagged rainfall and piezometer water levels.
This suggests that as rainfall occurs, piezometer water levels respond. As such, there is a
possible recharge mechanism of the rainfall to the aquifer and the aquifer is unlikely to be
fully confined.
Panel b) shows the relationship between piezometer water levels and time. While there does
appear to be some correlation between time and water level, the pattern is not particularly
linear or systematic (in terms of either a general increase or decrease). As stated above, this
is likely to reflect other environmental factors (e.g. extraction or surface water connections)
that have not been considered in the analysis. Further analysis would be required to identify
other factors that may be significantly impacting groundwater levels throughout the
Alstonville Plateau.

64
Panel c) shows the relationship between piezometer water levels and seasonal effect. While
technically this particular piezometer shows a statistically significant seasonal effect (in
terms of having a p-value less than 0.05), it is not a strong effect and no clear pattern can be
visually discerned. The contribution of seasonal effects to piezometer water level variability
was generally found to be small for most of the piezometers considered in this analysis.
Even in cases where the analysis revealed a statistically significant seasonal effect, the
magnitude of change in piezometer levels over a season tended to be very modest (after
accounting for rain and overall time effects). This suggests that the season has a relatively
minor impact on piezometer water levels in the Alstonville Plateau. Similar detailed analysis
for all piezometers is provided in Appendix 5.
The availability of reliable monitoring data on these piezometers from the Alstonville region
provides an invaluable tool for stakeholders who wish to understand how patterns might be
changing over time or in response to rainfall and other factors. While the simple empirical
analyses presented above are by no mean a replacement for more sophisticated hydro-
geologically based models and might be improved by further or modified inputs, they are a
useful tool for planning and monitoring. This analysis has identified lagged rainfall as an
important variable for understanding piezometer water levels in the Alstonville Plateau,
especially in shallow piezometers and deeper piezometers that are well connected to
surface waters and upper aquifers.
Water levels in these piezometers should continue to be regularly assessed to ensure
periods of sustained water level decline are identified early. If further analysis can identify
how changes in extractions relate to trends in groundwater decline and recovery, it may be
appropriate to set trigger values for water levels in key deep aquifers that allow for adaptive
management of groundwater extractions (e.g. once water levels fall below a certain level,
restrictions may be placed on extractions in that area).
Figure 17 and Figure 18 were created using an on-line tool created by the Review team that
can be used to interactively view hydrographs similar to those reported by Green (2006) and
showing the piezometer levels over time, along with a graphical display of rainfall levels over
the same time period. The online hydrograph tool allows the user flexibility in terms of which
piezometers to plot as well as whether to display daily rainfall or lagged averages (discussed
above). The hydrograph tool also allows flexibility in terms of zooming in to a particular time
period of interest. While this tool does not offer the same breadth of information as available
in Groundwater Explorer, it has the advantage of being simpler to use and is also much
quicker to run. The Review will explore the possibility with DPIE Water of the interactive
hydrograph tool being made more widely available.

65
Figure 17: Hydrographs showing the levels of 12 ‘shallower’ piezometers in the Alstonville Plateau from 2006 to present.
These include GW040999.1.1, GW041001.1.1, GW041003.1.1, GW041004.1.1, GW041005.1.1, GW041008.1.1, GW081006.1.1, GW081004.1.1, GW081000.1.1, GW036702.3.1, GW036702.2.2,
GW036701.1.1. The vertical axis in the top panel shows the distance from measuring point to water for each piezometer while the x-axis shows date. The bottom panel shows daily rainfall
associated with the Alstonville Tropical Research Station.

66
Figure 18: Hydrographs showing the levels of 16 ‘deeper’ piezometers in the Alstonville Plateau from January 2006 to present.
These include, GW040999.2.2, GW041001.2.2, GW041002.1.1, GW041003.2.2, GW041004.2.2, GW041007.1.1, GW041008.1.2, GW041000.1.1, GW041000.1.2, GW081005.1.1, GW081002.1.1,
GW081003.1.1, GW081001.1.1, GW036702.1.4, GW036701.1.4, GW036701.2.2
The vertical axis in the top panel shows the distance from measuring point to water for each piezometer while the x-axis shows date. The bottom panel shows daily rainfall associated with the
Alstonville Tropical Research Station.

67
Figure 19: Hydrograph for Piezometer GW041008, Hole 1, Pipes 1 and 2 with 120 day rainfall average plotted in the lower panel.

68
Figure 20: Panels a, b and c showing, respectively, the estimated rain effect, the overall time effect and
the seasonal effect estimated from fitting a GAM model

69
3.5 CONCLUSIONS
The WSP aims to set sustainable extraction limits for the groundwater sources through
consideration of a number of parameters, including rainfall over the area, recharge rates,
areas of high and non-high environmental value and a sustainability index. These variables
are subject to a level of uncertainty associated with the predictions and a precise value may
not be achieved due to the complex and heterogeneous nature of groundwater movement.
This is particularly evident in fractured rock systems that are difficult to fully characterise.
Given this, a range of practices have been used in developing and managing the WSP to
account for, or reduce, uncertainty – conservative calculations, adaptive management,
sensitivity testing, examining multiple lines of evidence.
The recharge rate, in particular, was considered in detail in this Review as a key technical
variable in the determination of the extraction rates. This was done through a review of
literature of recharge rates on comparable aquifers and application of alternative techniques
as a basis for recharge rates (chloride-mass balance and baseflow filtering).
WSP assumptions
 In groundwater studies and management, there will always be a level of uncertainty
associated with predictions (e.g. recharge rates) and a precise value may not be
achieved due to the complex and heterogeneous nature of groundwater movement.
This is particularly evident in fractured rock systems that are difficult to characterise
fully.
 The WSP plan was developed based on the best available data at hand and followed
a standard procedure. The assumptions made in the WSP are practical, reasonable
and in agreement with standard practice. In general, the WSP incorporates a
reasonable level of conservatism for extraction limits based on the risks identified.
 The rainfall data used and the methodologies are sound and apply limited uncertainty
to the extraction rates.
 The portion of the estimated recharge value available for extraction is a function of
rainfall recharge over low environmental value areas together with an assessment of
environmental and socio-economic risk.
 Calculating recharge is complex due in part to the variability and complexity of the
hydrogeology and limited knowledge of the systems. Based on the analysis, the
Review considers the recharge rates used in the WSP are reasonable and
conservative. This statement is made with a relatively low level of confidence due to
lack of data for the groundwater sources of interest.
 In practical terms the groundwater sources are treated as geologically homogenous
which adds uncertainty and would benefit from further work. The Review recognises
that the complexity of the geology makes it difficult to incorporate heterogeneity into
the WSP recharge calculations. Particular attention should be given to the effects of
geological variability within groundwater sources, and soils and vegetation overlying
aquifer outcrops. The Review acknowledges the conservatism incorporated into the
current WSP through the allowable allocation figures.
 There is evidence to suggest that for the WSP recharge variable, there is a wide
range of values that can be applied as well as a number of different approaches to
calculate it. Limited field data is available to support a single estimate and best
practice is to use more than one estimation method to reduce uncertainty if possible.
 Recharge rates applied to the four groundwater sources in scope in the WSP ranged
from 4% - 8% with studies and alternative methods indicating, with considerable
uncertainty, levels between 1% and 31%. The calculations by the Review using CBM
and baseflow filtering for recharge rates had results mostly above the values used in

70
 the WSP. The Review noted the important contribution that surface conditions and
soil could make to the recharge of the underlying geology.
 The Review tested a scenario in which the recharge rates were reduced by 80%. It
found the recharge reserved for the environment for the New England Fold Belt
Coast and North Coast Volcanics would remain at around 80% of recharge. For
Alstonville, it would be reduced, but the network of monitoring bores provides the
ability to monitor long-term changes in levels. For the Clarence-Moreton Basin, with a
relatively low volume water allocated, the LTAAEL could be reduced with no impact
on licences.
 Based on the analysis, the Review considers the recharge rates used in the WSP are
reasonable and conservative. This statement is made with relatively low level of
confidence due to lack of data for the groundwater sources of interest.
 The application of the sustainability index appears to be a cost and time effective risk
tool that is applied as an additional means to protect resources where limited
information is available.
 The WSP incorporates a reasonable level of conservatism for the extraction limits
when the groundwater sources are not fully allocated and where they are fully
allocated at Alstonville, monitoring is applied.
 Additional monitoring in strategic locations in the Tweed would help inform gaps in
knowledge on a regional scale and provide a path towards better conceptual
understanding of aquifer flows.
 The overall system is managed with some level of adaptive management, including
an annual determination of the volume of water per licence share and WSP are
subject to an interim review at five years with a full review at ten years.
 Impacts of climate change should be considered in future WSP methodologies. A
warming climate can lead to increases or decreases in rainfall at a location,
variations in the timing and frequency and strength of rainfall events, and increases
or decreases in evapotranspiration. The development by the NSW Government of
Regional Water Strategies will provide further insights into the impact that climate
change could have on the region and catchments over the coming decades, which
can further inform management approaches for the region’s water resources.
Sustainability of WSP extraction limits
 Due to limited extraction levels (where known allocations in the Tweed region are
much lower than the extraction limits contemplated in the WSP), limited data and
uncertainties described regarding the WSP parameters, it is not possible to conclude
whether the extraction limits are currently sustainable. However, the Review found no
evidence at this point in time that current WSP extraction limits are not sustainable.
 For the Alstonville Basalt Plateau Groundwater Source, which is fully allocated, and
there is a network of monitoring piezometers, data from 2006 onwards was analysed
by the Review, which concluded:
o The deeper piezometers (depths greater than around 25 m) showed a greater
stability and a steady upward trend over time of groundwater levels and/or
pressures. In contrast, the shallower piezometers showed greater variability
and appear to be recharged more regularly with rainfall.
o Lagged rainfall is an important variable for understanding piezometer water
levels in the Alstonville Plateau. This was observed in shallow-sited
piezometers and in deeper piezometers sited in systems that are well
connected to surface waters and upper aquifers.

71
 o There is limited amount of information available on current actual extraction
volumes. The Review notes that enhanced metering requirements will come
into force in the region in 2023 for eligible groundwater extractors. Given this
lack of data on extraction volumes, it is difficult to separate the effects of
environmental variables (such as rainfall) from the impacts of human
extraction (which tends to increase during dry periods).
Methodological improvements
 The Review considers there is room for improvement in the future assessment of the
variables underlying the extraction limits.
 Impacts of rainfall patterns in the region on recharge should be considered in future
WSP methodologies, including changing patterns associated with climate change.
 Particular attention should be given to assessment of groundwater recharge rates
across broad spatial areas and the associated need to distinguish between confined
versus unconfined aquifers.
 Developing a better conceptual understanding of the geological strata in the WSP to
reduce the level of uncertainty in the estimated recharge values. This could possibly
be undertaken via a 3D geological modelling tool (e.g. Leapfrog Geo) where there is
sufficient data and should include some soil mapping. This would require a large
scale detailed geological mapping survey or the collation of the existing core log data
and geophysical measurements, where available. The Review notes this level of
detail has not been typically applied in similar WSP for easterly flowing rivers and
would require allocation of time and resources.
 Sensitivity testing could be undertaken to see whether a change in the recharge or
sustainability index might result in the aquifers being over allocated or stressed.
 Further work could be undertaken to assess whether the risk ratings given to specific
groundwater sources are appropriate.
 Water levels in the Department’s piezometers should be regularly assessed to
ensure periods of sustained water level decline are identified early. With further
analysis, it may be appropriate to set trigger values for water levels in key deep
aquifers that allow for adaptive management of groundwater extractions (e.g. once
water levels fall below a certain level, restrictions may be placed on extractions in
that area).
 Where the system is fully (or near fully) allocated, additional monitoring/sampling and
routine data analyses could be applied, as was undertaken at Alstonville, within an
adaptive management framework.

72
4 EXTRACTION IMPACTS: UNDERSTANDING, ASSESSMENT AND
MANAGEMENT

The focus of this Chapter is on the mechanisms that cause impacts and consequences; the
complexities in measuring local scale impacts from bore extraction; the hydrogeological
assessments that form part of the development application and licencing processes;
associated assessment challenges and potential management solutions; and additional
information that would help to understand the systems better.

4.1 EXTRACTION IMPACTS AND CONSEQUENCES

For this Review, an impact refers to the physical change that occurs from an action (such as
groundwater extraction), while a consequence (following the analogy) may be the temporary
or permanent loss of water access or loss of environment for GDEs and associated flora and
fauna. An example would be if groundwater extraction results in reduced pressure heads
and groundwater discharge to a local creek (impact), which then affects flora or fauna
dependent upon that water source (consequence). Risk refers to the level of potential
consequence combined with its likelihood of occurring.

4.1.1 Impact mechanisms

There is an extensive volume of literature detailing the impact mechanisms from a bore (or
bore field) on the surrounding aquifer or connected waters. Readers are directed to Acworth
(2019) for a recent and detailed description of processes in Australia. In brief, groundwater
extraction from an aquifer (via a groundwater bore) reduces hydrostatic pressure heads at
the bore, creating a differential pressure gradient that induces water flow towards the lower
pressure area (i.e. towards the bore). The area influenced is called the ‘zone of influence’ or
‘cone of depressurisation’ or ‘cone of depression’ surrounding the bore (Figure 21). The size
of this zone, and the nature and degree of the pressure head change and resultant flow
paths, depends on the hydrogeological properties of the rock or soil matrix, as well as on the
rate and duration of the extraction(s).

Figure 21: Area of influence and cone of depression in the aquifer due to groundwater pumping
Source: Oregon State University (2019).

73
The primary means to calculate flow in porous media is Darcy’s Law. This equation states
that flow is a function of the differential pressure heads across an area and the saturated
hydraulic conductivity. The saturated hydraulic conductivity is a measure of how water flows
through an aquifer and its value varies widely – over at least 10 orders of magnitude -
depending on the rock or soil type, integrity and structure. Large variations in hydraulic
conductivity may apply even in the vicinity of a single zone of influence. Whereas pressure
heads can be directly measured via piezometers, assumptions are often made (based on
our geological understanding) to determine broad acre values of hydraulic conductivity.
In many locations the subsurface geology is complex and difficult to characterise. Natural
geological structures may be heterogeneous, with variations in geology, identity, structure,
and physical properties both laterally and with depth. This can result in important regional
and local-scale variations in saturated hydraulic conductivity and other properties. To
overcome these uncertainties, a variety of field techniques have been developed, including
geophysical methods, to derive local-scale values of hydrogeological properties. The most
commonly used field method is the well pump test, where water is extracted from a bore for
an extended period and adjacent piezometer pressure heads are measured before, during
and post the pumping period (which may be 12 hr, 24 hr, 48 hr or longer periods depending
on the extraction rate and geologic properties). Other commonly used methods to calculate
hydraulic conductivity include slug tests, direct pressure testing or core analysis.
In locations where the aquifer is connected (either directly or indirectly) to other systems,
drawdown may reduce water yields in adjacent bores or induce diversions from other
aquifers. Further, if any adjoining aquifers are unconfined, a decline in the water table
elevations may decrease groundwater discharge to connected surface waters and potentially
influence groundwater dependent ecosystems (GDEs). The induced flow between aquifers
or from surface water may also result in water quality impairments. Importantly, if the zones
of influence of two or more bores overlap, then the drawdown impacts are cumulative.
Mechanisms of impacts in practical terms are different between porous rock (e.g. Clarence
Moreton Basin Groundwater Source) and fractured rock aquifers (e.g. Alstonville Basalt
Plateau, North Coast Volcanics and New England Fold Belt Coast Groundwater Sources).
Flow through fractured rock is dominated by discrete fractures and complex folds that
formed through volcanic activity often tens or hundreds of millions years ago. In these
fractured rock environments the extraction point (the bore location and pumped depth)
depend on the fracture network that intersects that point. This network is difficult to
accurately map without extensive hydrogeological investigations and, in many cases, cannot
be explicitly determined. This means that extractions in fractured rock aquifers may be
unpredictable and do not comply to Darcian theory (i.e. flow is a function of fracture size and
fracture connectivity, versus pressure gradients and hydraulic conductivity). In contrast, flow
through porous rock better conforms to drawdown prediction models and hence, local
impacts may be more accurately predicted.
In either circumstance, expert interpretation of bore logs and pump test results is typically
required to determine aquifer hydrogeologic behaviour. In a practical sense, this variance
suggests that fractured rock aquifers require more investigations as they have higher
uncertainty, although this should not be used as rationale to limit data gathering in porous
systems. Further considerations are addressed below.
Extraction impacts are assessed in the hydrogeological assessment process for some new
water extraction approvals (for example, in water access licence dealings or water supply
works approvals). The range of impacts assessed and the standards for acceptable impacts
are discussed below.

74
4.1.2 Extraction consequences
Bore water extraction can potentially impact connected water within the same aquifer, within
a connected aquifer, or within a connected surface water body, leading to possible changes
in water quantity or water quality.
The range of impacts described in the DOI (2018a) document includes those on the
groundwater source itself (both in terms of quality and quantity), on groundwater dependent
ecosystems (GDEs), surface water, culturally significant sites, and other water supply bores,
as well as the compaction of sediments, and cumulative drawdown from existing approved
water supply works and entitlements (DOI, 2018a). For the purposes of the Review’s report,
for ease of explanation, physical impacts (such as on groundwater quantity and quality) are
distinguished from consequences (such as on culturally significant sites). Depending on the
magnitude and extent of the impact, these changes can result in environmental
consequences both within and outside the aquifer.
GDEs are a type of ecosystem which can be impacted by groundwater extraction. GDEs are
generally recognised as “ecosystems that require access to groundwater to meet all or some
of their water requirements so as to maintain their communities of plants and animals,
ecological processes and ecosystem services” (Kuginis et al., 2016). Not all GDEs draw on
groundwater directly and not all are solely reliant on groundwater, and the groundwater
dependence of GDEs will vary due to climate, geology and land use.
The extraction of groundwater can harm GDE ecology if it impacts the amount of water
entering wetlands and water courses, may lead to alteration of the ecosystem and loss of
ecosystem services (Eamus et al., 2006). For example, where extraction increases the depth
to groundwater, the availability of water within the root zone of terrestrial groundwater
dependent vegetation will be reduced, which could lead to stress and the potential loss of
vegetation (Eamus et al., 2015). Further monitoring and studies are required to improve
understanding of tolerance thresholds; and the relationships between drawdown, changes to
the groundwater environment and impacts on biota. Impacts to GDEs are considered in the
hydrogeological assessment process for groundwater extraction approvals. This process
seeks to minimise impacts on groundwater levels associated with GDEs and prevent
unacceptable consequences on GDEs.47

4.2 COMPLEXITIES IN MEASURING LOCAL SCALE IMPACTS

FROM BORE WATER EXTRACTION
4.2.1 Variability of fractured and porous rock systems
There are well established methods for measuring and/or predicting local scale groundwater
and surface water impacts. However, in the complex systems of interest here, considerable
technical challenges and costs are expected. Addressing these challenges can lead to
conservative generalisations or assumptions that idealise the system features, which may or
may not be conservative. A brief discussion of these challenges is provided in this section.
The region of interest is characterised by complex geology, with three fractured rock
systems (New England Fold Belt Coast, North Coast Volcanics and Alstonville Basalt
Plateau) and the Clarence Moreton Basin porous rock. As noted above, fractured rock
systems may be highly variable and it is often difficult to assess the size, shape or
connectivity of the fracture network. Typically, this uncertainty cannot be resolved via field
data collection and assumptions are required to predict sustainable yields.
In contrast, the Clarence Moreton Basin generally consists of porous rock aquifers that are
easier to conceptualise. Nevertheless, aquifer property variability within the Clarence

47
Other potential activities that can impact GDEs include contamination, salinisation, vegetation clearing and filling or draining
of wetlands.

75
Moreton Basin can lead to fractured rock-like behaviour. Overall, the complexities,
challenges and costs involved when working in these hydrogeologic systems must be
recognised when considering new monitoring guidelines or requirements.
There are considerable challenges in measuring impacts from groundwater extractions.
Spatial and temporal variability means that existing groundwater pressure monitoring is
unlikely to be sufficient, and monitoring should consider the need for an adequate number of
appropriately located measurements. Analytical or numerical modelling may be required to
extrapolate the measured pressure drawdowns to the asset locations but these models are
only as sufficient as the data used to conceptualise the system. Therefore, it is not a trivial
decision to request or require comprehensive assessments/modelling especially where there
is already available field (including local) information to indicate that the risks are low.

4.2.1.1 Spatial variability of hydrogeology

Spatial variability in aquifer hydrogeology exists over both regional and local scales. Over
regional scales, there will be significant variability in the source aquifer properties and their
overlying (potentially confining) layers. This suggests that broad-scale generalisations about
aquifer properties, and their degree of confinement (including those made in the WSP),
cannot be directly applied to a local scale impacts assessment. Similarly, using
hydrogeological properties measured in other water catchments and aquifers, even those of
the same classification, is of limited value. Over local scales, fractures and other variations in
hydrogeological properties in the area around an extraction bore mean that single or even
multiple observations of hydrogeological properties or drawdowns (e.g. during pump tests)
may not adequately characterise the Representative Elementary Area (REA). The concept of
the REA suggests that the area under observation should be of sufficient size to adequately
represent the broader characteristics. In porous media the REA can be relatively small,
whereas in fractured rock areas the REA must be much larger to be equally representative.
Another key factor is the connectivity between deep and shallow aquifers. Bore logs showing
clay layers (aquitards) between shallow and deep aquifers are sometimes used to support
the view that shallow aquifers and surface water are unlikely to impact one another (e.g.
aquitards or aquicludes). Moreover, while it may be correct that this supports the argument
that there is no connectivity along the profile of the bore log, the spatial continuity of the
aquitard is typically uncertain and the area in question may not be characteristic of its REA.
The local-scale spatial variability means that consequences for GDEs and water supply
works will not necessarily be at the locations closest to the proposed extraction point.
Further, flow pathways between the extraction point and the assets may deviate
substantially from straight lines. It is therefore challenging to determine appropriate
monitoring locations for measuring depressurisation as the extraction point propagates in
various means towards the assets at risk.
The complexities of evaluating and ascertaining potential connections between the deeper
groundwater system, the shallower aquifers and impacts on local assets, as a result of
spatial variability, is detailed in a case study of town water supply bores at Lumley Park and
Convery’s Lane which draw from the deeper aquifer in the Alstonville Basalt Plateau
Groundwater system. This case study is included at Appendix 6.

4.2.1.2 Temporal variability

Temporal variability of groundwater pressures and surface flows is related to climate,
extractions and potentially other human influences. These influences on pressures and flows
can have time-lags varying between minutes to years and have complex interactions that
make measurements difficult to interpret. This complexity is illustrated in the analysis of the
Alstonville groundwater data in Chapter 3. In many cases (but not confined aquifer systems),
there is a cone of depressurisation that encompasses assets, but there may also be assets
where the impact or consequence is present but more difficult to detect and unambiguously

76
attribute to the extraction. Continuous monitoring technology such as weather stations,
pressure transducers and extraction meters facilitate detailed analysis of responses and
potential drivers/influences. However, months, years or even decades of data may be
required to desegregate the various influences at sites of interest.

4.3 EXISTING TECHNICAL APPROACHES TO MEASURING

LOCAL SCALE IMPACTS
4.3.1 Groundwater pressure monitoring
Groundwater pressure or level monitoring is fundamental to impact monitoring. This requires
the drilling and construction of monitoring bores called piezometers. This technology is well
established. Generally, costs increase with the depth and numbers of bores. As much of the
cost is associated with hiring a drill rig and operator, installing additional monitoring bores
may not induce significant additional costs when other bores are being drilled
simultaneously. Nonetheless, the location and depth of the piezometer should be based on
the potential impacts being assessed and the hydrogeological conditions.
Monitoring of groundwater pressures continuously in time is undertaken in two main ways:
1) Conventional water level monitoring. Slotted pipes are installed at the required monitoring
depth, whereby the groundwater stabilises to a level equivalent to the water pressure head
in the aquifer. A pressure transducer installed below this level or ultrasonic transducer
installed at the top of the casing can accurately measure the groundwater level at a desired
time interval. This is an accurate method and has a secondary benefits as it can also be
used as a sampling piezometric well to collect water quality data;
2) Vibrating wire piezometers allow pressure heads to be monitored at multiple depths, but
with lower accuracy, higher cost and limited flexibility regarding the function of the bore. In
any case a traditional monitoring bore would also be required to take validation
measurements. Hence, conventional technology is typically more applicable in small-
medium enterprises, although in principle, the later would be useful if costs reduce and
accuracy rises.

4.3.2 Chemical and temperature tracers of water flow

Chemical and temperature tracers to identify groundwater flow patterns are sometimes used
to supplement groundwater pressure data. The presence of multiple drivers of water
pressures and flows (climate, extractions and potentially others) means that there is often
ambiguity in the cause-effect relationship. In hydrogeology, this is often addressed by
investigating water chemistry. For example, if the river water chemistry changes from the
typical background surface water chemistry to include components of a local aquifer’s
groundwater chemistry, this may signal a groundwater discharge point. Vice-versa, the
chemistry can be used to identify where surface water is being drawn into (recharging)
groundwater.
Temperature may also be used in the same way, since groundwater temperatures are
usually distinct from surface water temperatures. In this context, the chemistry or
temperature are called ‘tracers’ of water flow. The combination of pressure data and tracers
provides further lines of evidence to suggest connectivity between aquifers, or between
aquifers and the surface water. Many tracers are straightforward and low cost to implement
(although isotope tracers that provide more precise results and give additional information
about groundwater age require specialist methods). Indeed, Radon (222Rn) is a commonly
used natural environmental tracer as it provides an indicator of the volume of groundwater
within a surface waterbody or it can be used to identify groundwater sources.

77
4.3.3 Numerical modelling
Numerical modelling is often employed where limited field data is available to make
adequate predictions. In these cases, hydrogeological, surface hydrology and ecological
numerical models may be used based on monitored sites/periods. Numerical models may
also be used to simulate scenarios that test the effects of a single existing or proposed
extraction bore field if sufficient data is available to adequately characterise the aquifer
properties. The use of numerical models introduces many new technical challenges
including:
 the value of the numerical model is dependent on the accuracy of the underlying
conceptual model and the numerical calibration and validation process,
 the quality of the model depends on the quantity, quality and relevance of available
measurements; and,
 the time, data and expertise required for numerical modelling can increase
assessment costs.
The limited existing conceptual models and data sets to support modelling is a particular
challenge in the Northern Rivers and even more so in the Tweed Shire due to limited
previous projects that have warranted the investment. The cost of developing complex
numerical models means that they are generally used for larger projects where the potential
risks are considerable and the data is limited.

4.3.4 Surface water monitoring

Surface water levels can be measured continuously in time in a similar manner to
groundwater. For rivers, the water levels can be converted to flow rates using a rating curve.
Unless a hydraulically suitable natural site can be found, achieving accurate rating curve will
require an intrusive structure (weir or flume) to be built in the river as well as calibration and
regular maintenance.
Alternative non-intrusive river flow measurement technologies are also available. Another
limitation of using flow measurements to detect impacts of groundwater extractions is that
river flows tend to be dominated by climate influences. Along with the accuracy limitations of
the flow gauge, this can make it more difficult to discern the impacts of extractions unless
they are large compared to river flows. Extractions near to headwater streams may be
relatively large; however further downstream the catchment (where most existing flow
gauges are situated) the impacts of extractions of the magnitude relevant to this Review are
unlikely to be discernible.

4.3.5 Groundwater dependent ecosystems (GDE) monitoring

Assessing the potential impact of an extraction bore relies on monitoring data measuring the
local groundwater level or pressure heads. Applying information about the groundwater level
to a terrestrial GDE requires an understanding of the GDE’s groundwater use from surficial
and deeper aquifers. In many cases there is limited information available to know the exact
water requirements of the species of concern.
Methods for monitoring terrestrial GDEs include soil moisture, evapotranspiration and
various ecological indices (Richardson et al., 2011). Various technologies and theoretical
approaches combined with other data sources can be used to measure soil moisture and
evapotranspiration. Both soil moisture and evapotranspiration can be highly variable in
space and accurate estimates at one measurement do not always mean sufficient
representation of potentially impacted areas. Terrestrial GDE studies conducted in other
areas of NSW have generally used a multiple lines of evidence approach combining various
monitoring approaches.48 However, it is rare for this full range of monitoring approaches to

48
Including measuring the stress of groundwater dependent vegetation through tree growth point dendrometers (stem gauges
that monitor tree growth increment at small timescales), sapflow gauges and isotopic analysis of leaf samples (Eamus et al.,
2015)

78
be considered in impact assessments except when high priority GDEs are considered to be
at risk.

4.4 ASSESSMENT OF POTENTIAL IMPACTS

The WSP requires hydrogeological reports for certain extraction and development
applications. This section provides further detail on the assessment processes associated
with these hydrogeological reports. The next section considers the substantive content of
these hydrogeological reports.

4.4.1 Applications that require hydrogeological assessment

Hydrogeological assessments may be required under the Water Management Act 2000 for
applications that will change the authorised groundwater extraction volumes from new or
existing bores, for example, applications for water supply works approval or applications for
water licence dealings. The types of applications that may require a hydrogeological
assessment are summarised in Table 13 below.

Table 13: Types of applications under the Water Management Act 2000 that may require hydrogeological
assessment
Application type Description
s 92: Water supply works approval Approval to construct a new or additional groundwater work
s 71P: Subdivision and Division of a licence into two or more licences (usually so a portion
consolidation of access licences can be sold); or combining of licences
s 71Q: Assignment of rights under Reduction of the share component on a licence and the increase by
access licence dealing the same amount on another (previously referred to as a permanent
trade)
s 71R: Amendment of share Cancel an access licence and grant a new licence in another water
component of access licence source or management area
s 71S: Amendment of extraction Change the times or rates at which water can be extracted (not
component of access licence generally applicable to groundwater)
s 71T: Assignment of water Reduction of allocation in a licence account and increase by the same
allocations amount in another (previously known as a temporary transfer)
s 71U: Interstate transfer of access Same as 71Q dealing except it is between two interstate access
licences licences
s 71V: Interstate assignment of Same as 71T dealing except it is between two interstate access
water allocations licences
s 71W: Nomination of water supply Nomination of a works removed from or added to an access licence,
works to access licence irrespective of ownership and location.
Source: (DOI, 2018a)
Note: Applications for interstate transfer of access licences (s 71U) or water allocations (s 71V) will not require a
hydrogeological assessment if they do not impact bores in NSW.

4.4.2 Process for applications requiring hydrogeological assessment

Applications for water access licence dealings or water supply works approvals are lodged
with WaterNSW.49 WaterNSW may refer these applications to DPIE Water for
hydrogeological assessment if required. DPIE Water has the necessary expertise to conduct
hydrogeological assessments of applications as required. Figure 22 outlines the DPIE Water
process for assessing applications for water access licence dealings or water supply works
approvals.

49
WaterNSW is responsible for responsible for granting and managing water licences and approvals for rural landholders, rural
industries, developments which are not SSDs or SSIs. However, NRAR is responsible for granting and managing water
licences and approvals for government agencies, state owned corporations, water utilities, licensed network operators, mining
companies, irrigation corporations, Aboriginal communities, floodplain harvesting, state significant developments (SSD), stat e
significant infrastructure (SSI), schools and hospitals

79
Figure 22: Process for assessing applications for new water supply works approvals and water licence
dealings

4.4.3 Applications must satisfy minimum set of conditions

Prior to determining if a hydrogeological assessment is required, all applications must firstly
satisfy a minimum set of conditions including:
 Consistency with the WSP rules – for example, the WSP may specify conditions
on minimum distances to certain environmentally sensitive features (refer to
Appendix 7 for further detail);50
 Local management or trade area rules – for example, certain water dealings may
be subject to additional restrictions;51 and
 Sufficient water for trading – for example, the seller must have sufficient
shares/account water to trade.

50
For example, the Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 40-3
specifies rules for minimum distances between water supply works to minimise interference.
51
For example, the Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 48(1)-(2)
specifies prohibitions on trading between groundwater sources in the WSP and on certain assignments of rights from an access
licence in the Alstonville Basalt Plateau (Bangalow-Wyrallah) Management Zone to an access licence with an extraction
component that specifies the Alstonville Basalt Plateau (Alstonville-Tuckean) Management Zone. Note that this prohibition only
applies if it would cause the sum of the share components of all access licences in the Alstonville Basalt Plateau (Alstonville-
Tuckean) Management Zone to exceed the sum of the share components of all access licences in the Alstonville Basalt
Plateau (Alstonville-Tuckean) Management Zone at the commencement of this Plan.

80
The Access Licence Dealing Principles Order 2004 also specifies general principles and
rules for managing dealings in rights and allocations under water access licences. The core
principles expounded in the Order focus on minimising negative interference with other water
users and the environment, and maximising the value to society gained by extracting the
water. These principles include:
 Dealings should not adversely affect environmental water, water dependent
ecosystems, or geographical and other features of indigenous, cultural, heritage or
spiritual significance.52
 Dealings should not adversely affect the exercise of basic landholder rights. 53
 Dealings should have no more than a minimal effect on the ability of a person to take
water using an existing approved water supply work and any associated access
licences.54
 Dealings should maximise social and economic benefits of access licences to the
community. Access licence dealings rules should allow maximum flexibility in
dealings to promote this objective.55

4.4.4 Applications are subject to a risk assessment

Applications that satisfy minimum conditions are then assessed by DPIE Water for risk to
determine what level of hydraulic analysis is required to support the application. Applications
can either be considered to be:
 Low risk – no further hydrogeological impact assessment is required;
 Medium risk – assessment of drawdown impacts using a simple analytical hydraulic
model undertaken; or
 High risk – assessment of drawdowns using a detailed analytical hydraulic model a
is undertaken (DOI, 2018a).
As a general rule, applications which request approval to take larger volumes of water,56 or
which are proximate to other bores, groundwater dependent ecosystems, or other sensitive
areas will generally require hydrogeological evidence to support them and validate that the
impacts will be acceptably minor. For example, the WSP rules require a hydrogeological
report to establish evidence of acceptably minor impacts to approve proposed bores within
certain minimum distances from GDEs, groundwater dependent culturally significant sites,
other water supply works, and contamination sources (see Appendix 7).57

4.4.5 Applications may be subject to a hydrogeological assessment

Based on the risk assessment process, medium and high risk applications are subject to
hydrogeological assessment.58 To inform this hydrogeological assessment, the applicant
seeking approval for a dealing or water supply work should supply a hydrogeological report.
These reports are prepared by a groundwater consultancy, and will generally comprise a
pump test (see further detail below) and a hydrogeological study that includes a technical

52
Access Licence Dealing Principles Order 2004 cl 7, 8.
53
Access Licence Dealing Principles Order 2004 cl 9(1).
54
Access Licence Dealing Principles Order 2004 cl 9(2).
55
Access Licence Dealing Principles Order 2004 cl 10.
56
The Review notes that the licences entitlements for bottled water operators in the region cover a broad range from 5 ML up to
greater than 100 ML/year, with some operators drawing on their licences from multiple bores.
57
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 40-3. Note that the
standard of acceptability varies depending on the protected feature the works would be proximate to, for example another bore
versus a GDE or an environmentally sensitive area versus a source of contamination. WSP applies minimum distances
determined by Water Management Committees based on aquifer type and technical advice from the Department (DPIE,
2019a). The Department is developing a fact-sheet detailing the assessment process including assessment of potential impacts
on GDEs (DPIE, 2019a). Minimum distance rules may also be influenced by the application of more accurate GDE mapping,
developed from the PCT vegetation mapping (DPIE Water, 2019a).
58
Note that applicants submitting applications for low risk dealings or works would generally not required to provide a
hydrogeological report or conduct a pump test.

81
analysis of the pump test data; and assesses the potential drawdown of the proposed
extraction on neighbouring water uses and environmental assets.

4.4.5.1 Applications over 20ML/year will require a pump test

A pump test is required to support an application for approval to construct and use a bore to
extract over 20ML/y of water from a groundwater source within the coastal management
area of NSW, including areas within the WSP, for irrigation, industrial, recreation or other
commercial purposes. This pump test should be conducted by a groundwater consultant and
form part of the hydrogeological report consistent with WaterNSW guidelines (see Table 15),
including providing a “technical analysis of the pumping test information; and identification of
the potential drawdown impacts of the proposed operation on neighbouring users and
surrounding sensitive environmental assets” (WaterNSW, 2017).
Pump tests will generally require a test bore licence from WaterNSW prior to drilling, and
must be conducted in accordance with Australian Standards. The proposed volume of water
extraction will determine the minimum duration of pump testing,59 and drawdown and
recovery measurements from observation bores may be required (WaterNSW, 2017).

4.4.5.2 Applicability of the Aquifer Interference Policy (AIP)

The NSW Government’s AIP focuses on proposed high risk aquifer interference activities
where the purpose of water extraction is for disposal, not for use, for example, mine and
construction project dewatering involving hundreds of megalitres. WaterNSW and DPIE
Water have advised the Review that they consider all applicable policies when considering
applications for licences and approvals or providing hydrogeological advice, respectively.
However, they do not refer to the AIP for activities that are not defined as high risk under the
AIP or do not involve large volumes of water.60

4.4.6 Relevant impacts considered in the hydrogeological assessment

As discussed above, the hydrogeological assessment consider the potential drawdown
impacts of the proposed water dealing or works on neighbouring water users and
environmental and cultural assets. This includes impacts:
 on the groundwater source in question,
 on groundwater dependent ecosystems (GDEs),
 on connected surface water sources,
 on culturally significant sites,
 on neighbouring water supply bores,
 on groundwater quality,
 of compaction of sediments,
 of cumulative drawdown from existing approved water supply works and entitlements
(DOI, 2018a).
A key consideration is the impact of any drawdown on GDEs and culturally significant sites.
This reflects the objectives of the Water Management Act 2000, including to “protect,
enhance and restore water sources, their associated ecosystems, ecological processes and
biological diversity and their water quality” as well as to recognise and foster social and
economic benefits from the sustainable and efficient use of water.

59
For applications seeking a 21 to 50ML/year entitlement, a minimum one day pumping duration and recovery will be required.
For applications seeking a 51 to 100 ML/year, a minimum seven day pumping duration and recovery will be required. For
applications seeking more than 100 ML/year entitlement, a minimum 70 day pumping and recovery will be required
(WaterNSW, 2017)
60
See Appendix 4, Initial Report.

82
4.4.6.1 Groundwater dependent ecosystems (GDEs)
To be able to protect and monitor the health of GDEs, knowledge of their broad-scale
distribution, location and vulnerability to changes in groundwater level is required. The NSW
Government has undertaken work to identify groundwater dependent ecosystems to meet its
legislative requirements under the WMA and WSPs, which require spatial mapping of GDEs
to apply the minimum distance rules as discussed above.61
The report used remote sensing analysis, including vegetation and groundwater mapping
data to identify a model for high, medium and low probability terrestrial vegetation GDEs.
However due to the conservative nature of the decision rules, the report acknowledges that
some GDEs may have been inadvertently filtered out.62 The GDE model generated was
refined and tested against other existing literature, knowledge of underlying geology and
environment, and ground-truthing in other areas of NSW.63 This mapping will be further
updated for the coastal region of NSW following an update of the underlying vegetation
classification scheme, resulting in more accurate GDE maps (DPIE 2019, pers. comm., 02
August).
While the probability GDE mapping layers are used by DPIE, the GDE mapping used in the
WSP is a point layer map that is not electronically available and only recognises GDEs
considered ‘high priority’64. DPIE has indicated that new WSPs under development or being
re-made in NSW will refer to a full dataset of all GDEs identified by NSW DPIE and link to
the National GDE Atlas (DPIE 2019, pers comm., 23 October).65 The WSP allows for the list
of high priority GDEs to be amended after year five of the plan as further GDEs are
identified, or during the life of the plan following approval by the Minister (DPI Water, 2016f).

4.4.6.2 Culturally significant sites

Six different Aboriginal nations occupied the NSW north coast prior to European settlement
due to the high diversity and abundance of natural resources in the area, particularly around
the Northern Rivers region. The Tweed and Northern Rivers region is the traditional home of
the Bundjalung Nation. The area has a number of culturally significant coastal sites, special
meeting places, middens, campsites, hunting and gathering sites, crafting sites, and
ceremonial places. The water and vegetation in the area provided the people with critical
resources including flora and fauna.
The WMA and hydrological assessment process recognise and protect culturally significant
sites and aim to foster social and economic benefits to culture and heritage. The WMA
specifically aims to recognise and foster “benefits to the Aboriginal people in relation to their
spiritual, social, customary and economic use of land and water.”
These objectives are reflected in the WSP, which recognises the multidimensional
relationship Aboriginal people have with land and water – including spiritual, customary and
economic. This means that not only should certain sites be protected because of their
spiritual and environmental significance, but also because the ecosystems and flora and
fauna they support should be preserved to enable ongoing traditional practices including
hunting, fishing and gathering. The WSP recognises this multidimensional relationship by
establishment and prioritisation of Native Title rights within Basic Landholder Rights as well

61
This work is summarised in the report, Methods for the identification of high probability groundwater dependent vegetation
ecosystems (Kuginis et al., 2016).
62
This technique recognised some but not all wetlands; other potential GDEs such as stygofauna (groundwater invertebrates
important to maintaining water health) and groundwater baseflow contributions to surface water were not included.
63
Note that this mapping will be updated to provide a more comprehensive and accurate GDE map for the coastal region of
NSW following an update of the underlying vegetation classification scheme (DPIE Water, 2019b)
64
Identified in ‘The GDE Map’ attached in Appendix 10 of WSP Background Document (DPI Water, 2016f) and Appendix 3 of
the WSP
65
The Atlas includes data obtained through a national assessment process as well as NSW. GDEs will be prioritised according
to the High Ecological Value Aquatic Ecosystem (HEVAE) Framework developed as part of the Aquatic Ecosystems Toolkit
(2012).

83
as permitting Aboriginal communities to apply for a WAL for cultural purposes or community
development purposes. The WSP identifies a number of groundwater sources that represent
areas of spiritual and cultural significant to Aboriginal people.
The hydrogeological assessment process also recognises and protects culturally significant
sites, including sites of significance to Aboriginal people, by considering the potential
drawdown impacts of proposed water dealing or works on these sites (including with respect
to the traditional uses of these sites). As discussed above, the WSP specifies minimum
distance rules to groundwater-dependent culturally significant sites to avoid unacceptable
impacts on these sites. These rules apply not only to sites of significance to Aboriginal
people, but also to other sites of cultural and heritage significance.
Information on groundwater-dependant culturally significant sites is constantly updated by
DPIE through engagement with Aboriginal peoples (DPIE 2019, pers comm., 25
September). This is an ongoing activity as even once a site has been identified, further
information may be required to understand the cultural significance of the site, its
interrelationship with the groundwater source and the associated water requirements of the
site.

4.4.7 Defining the level of acceptable impacts

For different groundwater sources, the magnitude of acceptable impacts currently applied by
DPIE Water for groundwater dealings and water supply work approvals on water drawdown
varies. Table 14 details the acceptable impacts on the water table and groundwater pressure
for most porous and fractured rock groundwater sources in NSW.66 These criteria apply to all
four groundwater sources within the scope of this Review.
Table 14: Acceptable level of impacts for porous and fractured rock groundwater sources
Type of impact Level of acceptable impacts

Impact on water 1. Less than 0.1 metre cumulative drawdown in the water table 40 metres from any:
table (unconfined a. High-priority, groundwater dependent ecosystem, or
aquifers) b. High-priority, culturally significant site.
2. An additional drawdown of not more than 10% of the pre-development Total
Available Drawdown (TAD) to a maximum of 2 metres at any:
a. 3rd or higher order surface water source measured at 40 metres from the
high bank.
b. Water supply works (excluding those on the same property), subject to
negotiation with impacted parties.
3. A cumulative drawdown of no more than 10% of the pre-development TAD of the
unconfined aquifer at a distance of 200 metres from any water supply works
including the pumping bores.
Impact on 1. A cumulative drawdown of not more than 40% of the pre-development TAD at a
groundwater distance of 200 metres from any water supply works including the pumping bores.
pressure 2. An additional drawdown of not more than 3 metres at any water supply works
(confined/semi- (excluding those on the same property) subject to negotiation with impacted
confined aquifers) parties.
Source: (DOI, 2018a)

When assessing expected impacts against the acceptable level of impacts, the impact
period considered varies from one year for temporary trades,67 to 10 years for permanent
trades.68

66
Except for porous and fractured rock groundwater sources within the Great Artesian Basin for which different criteria apply.
67
For example, assignment of water allocation to another licence under s 71T of the Water Management Act 2000 (DOI,
2018a)
68
For example, assignment of share component of a water access licence under s 71R or 71Q of the Water Management Act
2000 (DOI, 2018a)

84
4.5 MANAGEMENT OF IMPACTS AFTER APPROVAL
The primary mechanisms for managing the impacts of groundwater extraction under existing
groundwater entitlements are conditions on works approvals or access licences, conditions
on development consents through the councils, water allocations (or ‘available water
determinations’) and temporary water restrictions.

4.5.1 Conditions imposed on approvals (WALs and works approvals)

Approvals may be subject to conditions, including:
 ‘Mandatory conditions’ – conditions imposed by the Water Management Act 2000,
Water Management Regulations 2018, or the relevant WSP,69
 ‘Discretionary conditions’ – conditions specific to the particular approval and location,
for example to give effect to agreements between an applicant and an objector, or to
protect the environment.70 These conditions would be informed by the
hydrogeological assessment that may identify particular areas of risk that can be
managed through conditions of approval.
Mandatory conditions prevail over discretionary conditions to the extent of any inconsistency
between them.71 The applicable conditions are specified in the relevant licences or
approvals. Common conditions on water access licences and works approvals for
groundwater extraction include:
 Installation, maintenance and operation of appropriately configured water meters.
 Installation, maintenance and operation of a data logger.
 Recording of pumping activities in a logbook.
 Provision of data, records and reports to the Minister and/or the Department covering
water quantity, water quality, application of water, etc.
 Permitting the Department access to the site to inspect and test the works.
 Duty to notify the Minister or Department of breaches.72

4.5.2 Conditions of development consents

Bottled water operators in the Northern Rivers region generally require a development
consent under the Environmental Planning and Assessment Act 1979 (EP&A Act) to
construct or expand the necessary facilities to extract water for bottling purposes. To obtain
the necessary consent, the operator submits a development application to the local council.
Historical development consents for water bottling firms operating in the region date back to
1993.
It is within council’s remit to apply conditions and performance measures to the consent,
which can be in relation to impacts and consequences of the development. These may
require the operator to develop a plan for monitoring and reporting their performance against
these conditions. For example, a recent consent (issued late 2018), imposed conditions
related to water take (specifically, the volume in ML/y of water permitted to be extracted), a
requirement to use a daily log book, and the transport of that water (specifically, maximum
truck movements and daily hours of operation permitted). Conditions provide a mechanism
for councils to set standards and outcomes for protecting certain environmental values and
to monitor and assure compliance. Conditions may be varied subject to an application by a
proponent to modify the consent, but only in relation to the subject area of that modification
application.

69
Water Management Act 2000, s 100(1)(a) and s 100(1AA)
70
Water Management Act 2000, s 100(1)
71
Water Management Act 2000, s 100(1)(a) and s 100(1AA)
72
A search was conducted of conditions of water access licences and water supply works approvals for bottled water extraction
operations in the Northern Rivers region.

85
The Review considered a number of development consents for bottled water operators in the
Northern Rivers region. While the scope of conditions contained in these consents was
broad, the Review did not identify conditions that related to local impacts on aquifers or
groundwater, or potential environmental consequences associated with those impacts.
Tweed LEP 2014 required the Council to be “satisfied that development will not have an
adverse impact on natural water systems or the potential agricultural use of the land”73 and
under the EP&A Act, to consider the likely environmental impacts of the development. 74 As a
result, Tweed Shire Council has informed the Review that it expects more detailed
hydrogeological information to be submitted during the assessment process for development
approvals for the bottled water industry, including seeking information on conceptual models,
testing and ongoing monitoring plans.
The Review also notes that these expectations around hydrogeological assessment also
need to be seen in the context of a regulatory framework that places significant responsibility
for hydrogeological assessment and licence or works approvals with the state government.
Tweed Shire Council has informed the Review that where the extraction of water has been
subject to the hydrogeological assessment and approval by the state government, the
Council historically focused on assessing and imposing conditions on other environmentally
relevant matters not covered by conditions of the water licence or works approval – for
example, noise, truck movements and hours of operation.
However given the LEP, there was a question as to the extent to which the Council needed
to or should undertake its own hydrogeological assessment.
Given the implications of the overlap between development consent conditions and water
licences or works approvals, further work should be undertaken to ensure consistency, to
avoid duplication of effort, and to address any gaps in the assessment and approvals
process.

4.5.3 ‘Water allocations’ or ‘available groundwater determinations’

WSPs provide a mechanism, a ‘water allocation’ or an ‘available water determination’
(AWD), to control water take for each licensed water user each water year. The AWD is
intended to ensure that water take is managed to the extraction limit, to prevent impacts on
the water source and other users or consequences to GDEs, and to provide certainty to
water users regarding the amount of water that can be taken and under what conditions.
The AWD process for groundwater sources determines the available water in the coming
water year by considering the LTAAEL,75 water entitlements under access licences and
basic landholder rights, and actual water take. The AWD assigns a portion of the available
water to each licensed water user based on their water entitlement. While the AWD is
conducted each water year, the focus of the AWD is to manage sustained growth in actual
water take to the LTAAEL, which is a long-term measure.76
On 26 June 2019, DPIE Water issued an Available Water Determination Order for the North
Coast Costal Sands and the North Coast Fracture and Porous Rock Groundwater Sources
2019 for the 2019-20 water year commencing 1 July 2019. The statement allocated local
water utility and aquifer licence holders covered by the WSP groundwater sources an
allocation of 100 percent of their entitlement, or 1 ML per share unit.

75
The LTAAEL represents the extraction limit of a particular groundwater source over the long term, expressed as an average.
76
For example, the standard water allocation for licensed water users is 1ML/unit share, but Clause 29(2)-(3) the Water
Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 specifies that where growth in water
take is assessed to have increased more than 5 percent above the LTAAEL extraction limit over a three-year period, the water
allocation may be reduced to less than 1ML/unit share.

86
4.5.4 Temporary water restrictions
Under section 324 of the Water Management Act 2000, temporary water restrictions can be
imposed in the public interest.77 These restrictions can prevent or impose restrictions on the
taking of water from a specified water source and area for a specified period:
 to maintain or protect water levels,
 to maintain, protect or improve water quality,
 to prevent land subsidence or compaction,
 to protect groundwater dependent ecosystems, or
 to maintain pressure or to ensure pressure recovery. 78
The majority of Temporary Water Restriction Orders that have been enacted are for
regulated river systems. Fewer Orders have been enacted for groundwater sources.79

4.5.5 Collaborative data and approach

The Natural Resources Commission’s 2005 Standard for Quality Natural Resource
Management established ‘Opportunities for collaborations’ as one of seven components of
the Standard, and the subsequent 2013 NRC review of the 2004 WSP was undertaken in
view of this Standard (NRC, 2005). The 2013 review identified a number of examples of
collaboration and further opportunities, including sharing spatial information between
agencies, with licence holders on operational matters; to ensure shared data is used; and
ensure NSW Government’s various regulatory frameworks, investments and interventions
complement each other (NRC, 2013).
This Review (by the NSW Chief Scientist & Engineer) has also identified opportunities for
enhancements in the way that agencies and council collaborate on data sharing and their
various roles in the water management process including with respect to the bottled water
industry. These include the need for relevant authorities to:
 Develop a shared understanding of the data, modelling and information needed by
each agency in decision making
 An agreed and documented set of standards for data capture, sharing, storage
(frequency, metadata requirements, computational and program requirements, data
sharing protocols including in relation to commercial data)
 When potentially new measurement and monitoring technologies or regimes are
introduced, an approach to discuss (and reach agreement on) the most effective
measures to roll out new monitoring to maximise the utility and outcomes for the
group of agencies
 Where new performance indicators are identified through the WSP or performance
measures conditioned on proponents, a shared understanding by agencies of which
legal instrument these measures should be attached to, and how data on these
measures is communicated with relevant agencies for the purpose of compliance, or
measuring cumulative impacts, or providing for research activities. Where new
performance measures are required of proponents, a forum to discuss between

77
Water Management Act 2000, s 324(1).
78
Water Management Act 2000, s 324(2)
79
On 21 February 2018, temporary water restrictions were imposed for the Botany Sands Groundwater Source in response to a
threat to public health and safety, see NSW Government Gazette No 23 of 23 February 2018, 816. On 1 September 2009,
temporary water restrictions were imposed for the Mid Murrumbidgee Groundwater Management Area 013 in response to a
threat to public health and safety, see NSW Government Gazette No 136 of 25 September 2009, 5229. On 17 August 2009,
temporary water restrictions were imposed for the Upper Namoi Zone 11 – Maules Creek Groundwater Source to protect water
levels in an aquifer and to protect groundwater dependent ecosystems, see NSW Government Gazette No 113 of 18 August
2009, 4815. On 25 June 2009, temporary water restrictions were imposed for the Lower Murrumbidgee Groundwater Sources
and Lower Murray Groundwater Source in response to a water shortage, See NSW Government Gazette No 95 of 26 June
2009, 3797-8. On 14 March 2009, temporary water restrictions were imposed for all aquifers or parts of aquifers underlying the
Blue Mountains City Council Local Government Area to protect groundwater dependent ecosystems, see NSW Government
Gazette No 52 of 20 March 2009, 1431.

87
 agencies where there are potential contradictory measures, and the most appropriate
instrument to reflect conditions in a way that prevents duplication.

4.6 OBSERVATIONS ON PAST HYDROGEOLOGICAL

REPORTS
Hydrogeological reports dating from 2007 to 2019 were made available to the Review by
seven operators (or proponents) of bottled water extractions in Tweed Shire and Ballina
Shire. These represent a large proportion of the industry hydrogeological reports that are
being used to support recent licence or development applications related to the bottled water
industry. Two reports that did not support license or development applications are discussed
in this section in general terms only. The Review notes that a small number of operators
have historical licences and documentation was limited.
This section summarises the reports’ purposes and contents, public and government
responses to the reports, and general observations regarding the comprehensiveness of the
assessment and alignment with currently available guidelines. The purpose of this analysis
is to identify general challenges that emerge in developing reports (versus a full technical
assessment) that are sufficient to inform license assessments or development applications.
The hydrogeological reports, in most cases, assess hydrogeological impacts of proposed
developments on surface flows, GDEs and surrounding groundwater bores. Some of the
hydrogeological reports have additional purposes including assessing the commercial
viability of the bores. One report focusses solely on the water quality risks to the bore. One
report is an update of a historical report, including supplying updated water quality results,
without supporting a current development or entitlement application.
The focus and style of the reports varies depending on the purpose and the identified or
perceived risks. Six of the seven reports are presented in conventional consultancy report
format; one is presented as a letter to the proponent.

4.6.1 Review of the available hydrogeological reports for the industry

The collective content of the hydrogeological reports is reviewed here against the
Groundwater Consultancy Requirements - Hydrogeological investigation report standardised
table of contents in the “Coastal groundwater: Test pumping groundwater assessment
guidelines for bore licence applications”. This guideline is specific to pump tests for
supporting license applications and does not provide a definitive guideline for the
hydrogeological reports in question; however, it provides a useful, established template for
this review (Table 15). Other relevant comments regarding the content of the hydrogeology
reports, for the purpose of supporting development applications, are listed below.

4.6.1.1 Level of detail in the hydrogeological reports commensurate with the

risks and requirements of applicable guidelines and policies
All reports refer to the WSP. Identification of specific relevant requirements of the WSP is
variable between the reports.
The main public policy document referred to in the hydrogeology reports is the WSP. This
policy stipulates minimum distances of extractions from high priority GDEs, high banks of a
river or stream, culturally significant sites and water supply works. There is provision for
changing these distances on a site specific basis if risks can be shown to be low, implying
additional investigation would be required. In most of the existing/proposed bottled water
sites, these minimum distances are reportedly met, and so there is no explicit requirement in
the WSP to conduct further hydrogeological investigation.
In cases, there are rivers/creeks within the stipulated minimum distance (40 m) of the
extraction sites, and in all cases there are rivers/creeks within a few hundred metres;
therefore there is an onus on the proponent to demonstrate there is no (or very low)

88
hydraulic connectivity between the target aquifer and the surface water. This is done with
reference to the regional-scale classifications of connectivity in the WSP, to bore logs that
show confining layers, to results of pump tests, and, in cases, to differences in water
chemistry between the aquifer and surface water and anecdotal evidence of stable surface
water responses.
During consultations in the Review, concern was expressed over the lack of objective, local-
scale evidence to demonstrate low connectivity between the target aquifer and surface
water. This is reflected in hydrogeology report conclusions by use of subjective terms such
as “unlikely to cause impacts”, which are considered by some stakeholders as lacking an
evidence base.
Beyond the WSP requirements, the purpose of the reports includes satisfying the council
requirements that the hydrogeological risks are acceptable. Meeting the minimum distance
requirements of the WSP and the largely qualitative analysis of connectivity undertaken in
the reports is not necessarily sufficient to meet this purpose. The absence of detailed
modelling or significant monitoring of assets at potential risk means impacts and risks are
unquantified. The cost-benefit of additional modelling and monitoring where this is not
explicitly required by the WSP or other published guidance is predominately based on
judgement and is disputable even among hydrogeology experts .
The hydrogeology reports all recognise that lack of data hinders understanding. In most
cases the reports include recommendations for further investigation, although they do not
explicitly recommend this is required prior to approval. Two reports recommend further pump
tests to confirm sustainable extraction rates, two recommend additional monitoring during
bore operations, and one recommends monitoring shallow groundwater with an associated
cease-to-pump trigger. The review of the hydrogeology reports also noted that in some
cases uncertainties and data limitations were identified in the report; however these were not
always well reflected in the report’s conclusions or executive summary.

89
Table 15: Contents of five bottled water industry hydrogeological reports from the Tweed and Ballina Shire compared against standardised table of contents in the
coastal guidelines
Hydrogeological investigation report standardised table of contents Included in report Extent of detail
Certification Groundwater consultant (qualified) Yes, in all reports
Introduction Property location, identification of the proposed Yes, in all reports All reports provide maps showing property location. All reports
development include a map or maps showing key features of the property (~1
km); all but one show the location of the property with respect to
locality features (~5-10 km); and most also show location on the
regional scale (~50-100 km scale). Quality of maps is variable.
Purpose for which the licence is being sought Yes, in all reports Sufficient
Geology Geological description of the property and surrounding Yes, in all reports All reports describe the site’s geological context. The level of
region detail in all reports is constrained by lack of available regional
and local scale geological data. All full reports contained one or
more bore logs that describe geology over the depth of the
bores. Only one report provides a regional geological map,
although the value of this is debatable, and other reports refer to
relevant published maps.
Stratigraphic boundaries or structural features that may No Available data does not permit this, although it is likely to be
influence groundwater availability relevant to at least some of the sites
Hydrogeology Setting: Description of the type of aquifer and a summary Yes, in all reports In some reports the information could be improved using existing
of typical water bearing zones encountered in test bores sources of data
in the vicinity of the property
Licensed: Details of licensed water supply bores within All reports identify The likelihood of impacts is addressed with support from pump
1km of the property including works purpose and nearby water supply test results, water quality data, bore logs, regional hydrogeology
likelihood of being impacted should the proposed bores, but not all data, the thresholds specified in the WSP and in some cases the
development proceed use a 1 km radius; AIP, and in two cases other (unsubstantiated) local hydrological
all assess likelihood observations. Due to absence of monitoring and modelling at the
of impacts potential receptor bores/GDEs and pathways from the extraction
to these receptors, the analysis of likelihood is subjective.
Environment Identification of ecosystems likely to be Yes, in all reports The identification of GDEs in most cases relies on the
groundwater dependent, surface water systems that High Priority Groundwater Dependent Ecosystem (GDE) Map in
could be affected by reductions in discharge with the WSP and the National GDE Atlas webtool supplemented by
prolonged pumping site visits.
Particular identification of sensitive ecosystems of special Yes, in all reports One report is vague about criteria used for determining presence
conservation value of GDEs.
Field work Test bore establishment: Details of the drilling and Information about The pump test bores in all cases are existing pumping and
construction of the subject bore, identifying the test bore the bores is monitoring bores (piezometers). I.e. they were not drilled for the
licence under which it was authorised. A statement of provided in all purpose of the reported pump tests. The available details on
compliance with the Minimum Construction Requirements cases although the bore establishment are variable. In most cases bore logs are
for Water Bores in Australia – Second Edition 2003 or level of detail is provided. A statement of compliance is made in only one case.

90
Hydrogeological investigation report standardised table of contents Included in report Extent of detail
subsequent equivalent guideline. variable. In all cases but one, other relevant observations of the local
hydrological and land use context made during field visits are
described.
Test pumping and recovery: Measurements and graphical Pump test data are As stated above, the observations make use of available bores
analysis documentation of drawdown and recovery data provided in all rather than being drilled specifically for the tests. In one case an
for pumping and observation bores. Calculated aquifer cases. In all cases observation bore was not used (it was recommended if the
transmissivity and storativity values, together with bore water quality approval was given). In all cases but that one, aquifer
efficiency estimates. Details of the water quality tests information is transmissivity and storativity values are estimated, together with
(compliant with requirements; National Uniform Drillers provided to show bore yields. Water quality assessment is extensive due to its
Licensing Committee, 2012) undertaken to demonstrate fitness for purpose commercial importance. Sufficiency of pump tests (including
the groundwater is suitable for the intended purpose. although location of monitoring bores (piezometers) and length of test) is
compliance is not questionable in cases, and further pump tests are recommended
stated. in some cases.
Impact Sustainability: Predictions of the impacts of pumping of All reports The level of quantitative analysis presented is variable,
assessment the subject bore on neighbouring licensed users and undertake this to depending on the availability and quality of pump test data for
potential groundwater dependent ecosystems based on some extent identifying a zone of influence. In most cases expert judgement
the required controlled test pumping, together with the is relied upon, including caveats about lack of data. The potential
predicted effects on groundwater levels for the region to affect discharge to surface systems is not assessed because
surrounding the subject property and the potential to aquifer discharge locations are unknown but are not thought to
affect discharge to surface water systems be local based on the regional-level connectivity descriptions in
the WSP. Connectivity of the target aquifers with shallow
groundwater and surface water is assessed using regional scale
knowledge supplemented by bore logs and water quality.
Trigger levels: Identification of the threshold drawdown One report Only one report recommends a trigger. In all cases, risks to
levels adopted to prevent impacts on neighbouring bores recommended a neighbouring bores and GDEs are concluded to be low, and it
or ecosystems, and estimations of the maximum shallow may be assumed that triggers were not considered appropriate
drawdown impact on neighbouring bores, monitoring groundwater trigger (cease-to-pump triggers are generally only used as part of
bores (piezometers) and ecosystems with and without level groundwater licensing in high-risk projects such as mining).
trigger levels being active
Management Responses: Actions to be taken if threshold
levels are reached or exceeded, including reporting to
regulatory authority, cease-to-pump conditions, and
provision of water to affected users
Operation Schedule Identification of the proposed operating regime All reports refer to All reports refer to licensed volumes; some also refer to
including discharge rate and hours of pumping pumping volumes. proposed annual volumes; and some to hours of pumping that
can be sustained at a given pumping rate. None refer to the
specific operating regime, which is likely to be unknown at the
time of the analysis.

91
Hydrogeological investigation report standardised table of contents Included in report Extent of detail
Monitoring Descriptions of the location of monitoring Monitoring Most reports recommend continual monitoring of drawdowns at
bores (piezometers), the frequency at which monitoring is piezometers and existing or new monitoring bore (piezometers).
to be undertaken and the type of data to be collected the monitoring used
during pump tests
are described
Reporting Details of the timing of reports, the type of This is not included The reports are not written in the context of an adaptive
information to be reported to the regulatory authority, the in the hydrogeology regulatory regime, since this does not typically apply to low risk
number and nature of exceedances and response times reports extractions. Most reports recommend continued monitoring to
between an occurrence and management actions being inform sustainable extractions.
implemented, and methodologies to be adopted to
mitigate impacts should they be ongoing
Constraints Identification of any consent conditions imposed by All reports describe The reports identify the status of license and development
council or other regulatory authority that would prevent current relevant applications and relevant constraints. In most cases licenses
the requested entitlement being realised in full for the licenses exist and the application is about a change of bore use and/or a
purpose for which the council development application.
licence is being sought. In particular, conditions limiting
the supply of water to other parties are to be identified
References Citations of all documentation referred to within the Yes, in all reports. There are cases where the referenced documentation must be
report. inferred, rather than being properly cited.
Figures All diagrams referred to within the report, including a All reports include The maps vary from excellent quality showing all relevant
locality map, a plan of the property identifying separation maps available information, to poor quality and of questionable
distances between the subject bore and site boundaries completeness.
or other features (especially suspected groundwater
dependent ecosystems, licensed works and surface
water bodies), geological map and sections, together with
a plan illustrating the extent of predicted drawdown during
the proposed pumping operation
Appendices Raw data and additional diagrams or text required to All reports (except All relevant raw data are not included in all reports, and in cases
provide background or support to the findings of the that in the format of where it was provided, there was no accompanying
investigation a letter) provide interpretation or description of methods used to obtain the data.
appendices with
further data
Source: (WaterNSW, 2017)

92
4.6.1.2 Accessibility of the hydrogeological reports to a non-technical audience
The hydrogeology reports are necessarily highly technical are not comprehensible to non-
experts. In most cases the Executive Summary and/or Conclusions sections of the reports
are more accessible; however these alone cannot be relied upon to make approval decisions
since they do not always reflect the full detail of the report including any particular
uncertainty in the local hydrogeological conditions and limitations of the tests and data
employed.
During consultation, some councils commented on their limited internal hydrogeological
capacity and reliance on guidance by WaterNSW/DPIE Water and/or external consultants
and hydrogeologists when necessary. Documents show that significant assessments are
made to support the Council by WaterNSW (using technical advice from DPIE Water). It is
important that the reports retain the highly technical content given the complexity of
groundwater assessments in general, but particularly in complex systems like those in this
region and the risk of losing vital information. Clarity around this could be improved through
the application of standardised table of contents like the one in the Coastal Groundwater
guidelines.

4.6.2 Public responses to the hydrogeology reports

In three cases, the Review team has seen responses by public interest groups and/or their
expert advisors, where concerns about the methods and conclusions of the hydrogeology
reports are raised. The concerns mainly relate to:
 The lack of measured data to support assessment of local impacts. In particular, the
responses from experts and interest groups argue the lack of characterisation of the
connectivity between the proposed extraction point and nearby creeks.
Recommendations are made by academics, commissioned to peer review the
reports, to: measure shallow and confined groundwater pressures to understand the
connectivity; conduct shallow groundwater monitoring over different depths between
the creek and target confined aquifer to measure responses to pumping; and to
measure creek flow responses to pumping including in low flow periods that are most
vulnerable to impacts. In one case, the expert presents calculations that demonstrate
the potential for large impacts on creek flows if a high degree of connectivity is
present. In another case, water quality data are re-interpreted by an expert as
signifying the potential for a connection between the targeted aquifer and the creek.
 The over-reliance on aquifer-scale generalisations to support local scale assessment.
This relates to assumptions about connectivity taken from regional generalisations in
the WSP. While the low levels of the proposed extractions compared to the extraction
limits in the WSP are recognised by the expert, this is considered to be an insufficient
indicator of local impact risk. Cumulative impacts due to future potential extractions
are also raised as a concern.
 Uncertainty. Following from the above points, the experts and interest groups argue
that some conclusions of the reports are unreasonable given the level of uncertainty
in the assessment.
 Perceptions of factual errors. In one case the interest group notes that there are
bores near the proposed site that are omitted from the report. In a later response, the
hydrogeology consultant noted that these bores could not be identified from publicly
available databases and welcomed further information about them.

4.7 FURTHER INFORMATION TO UNDERSTAND SYSTEM

IMPACTS FROM BORE WATER EXTRACTIONS
As discussed, there are considerable challenges in measuring and managing local impacts
from groundwater extractions due to numerous factors, in particular the spatial and temporal

93
variability of hydrogeological systems, and imperfect monitoring technologies and modelling
methodologies. However, further improvements could be made to assist in measuring,
assessing and managing the impacts of groundwater extraction within acceptable limits.
While some of these improvements could pose resourcing challenges for proponents and
government, a set of potential improvements is discussed below.

4.7.1 Investment in monitoring and technological advances

One of the challenges associated with assessing potential impacts from extractions is the
limited amount of field data available, particularly in the Tweed systems. At present there are
very few monitoring piezometers in the Tweed, although this is not the case with the state
monitoring network in the Alstonville region. Improving piezometric monitoring is needed, at
multiple complementary locations where there is perceived risk and/or lack of knowledge of
groundwater responses and flow. However, investment in additional monitoring should
balance the resources required to install and maintain equipment, and record and interpret
the data against the value of expected improvements in data availability and quality. This
should be informed by the characterisation of the risk level that the system is in – systems at
greater risk would be prioritised over locations and systems at lower risk.
Another important development is the increasing ease at which sensing technologies can be
deployed to the field at low cost. These technologies are variously commercially available or
under development, and include improved traditional sensing technologies (for example,
lower cost and more adaptive remote sensors), as well as advanced technologies offering
novel capabilities (for example, quantum gravity sensors able to penetrate depth with
significant accuracy).

4.7.2 Metering data

Water extraction metering supports the validation of actual groundwater extraction against
licenced allocations and is an important improvement for water policy and management at a
regional level (including the development of water sharing plans and available water
determinations), as well as enforcement actions against individual licence holders. Access to
extraction volumes would also help better determine impacts caused by extractions versus
climate variability.
The Review notes that four of the bottled water operators in the region were required in
December 2018 to install meters on their systems. The metering policy for qualifying
groundwater systems will take effect in this region in 2023 (DOI, 2018b).

4.7.3 Improving clarity on NSW government’s expectations of

hydrogeological assessment reports
The evolving regulatory context has resulted in a number of policy and guidance documents
that are potentially applicable.80 These documents have led to a complicated array of
guidance to proponents and consultants about what is required for a hydrogeological
assessment.
As discussed above, hydrogeological assessment reports are variable in content and the
level of contextual information, monitoring and data analysis included reflects the
characterisation of the risk (Figure 22). Some of the hydrogeology reports this Review
considered contain assessments of commercial viability of the bores as well as
environmental assessments. In cases, the monitoring or methods used were not considered
adequate by experts representing interest groups or by the agencies responsible for
evaluating the reports. To a large extent this can be addressed by comprehensive and

80
These include the WSP and its supporting documents, the AIP and its supporting documents, the 2018 Water resource Plans
– Fact sheet – Assessing groundwater applications, and the Coastal groundwater test pumping assessment guidelines, among
others.

94
consistent guidance on approaching the hydrogeological assessment as discussed above.81
To help facilitate consistent contents of the reports, the environmental component could be a
separate report or be clearly signposted.
These challenges have been addressed to some extent by the recent fact sheet (September
2018) that provides definitive criteria for acceptable levels of impacts (DOI, 2018a). The
Review is also aware that DPIE Water is currently updating some pump test requirements. It
is recommended that the relevant NSW government agencies should seek feedback from
consultants on this document and its interaction with other policies, potentially leading to
revised and integrated guidance. Similarly, feedback on applicable pump test guidelines
could be sought and considered. The NSW government should continue to strive towards
publicly available guidelines that reflect consistent internal assessment methods of agencies,
and continue to provide access to its technical experts to provide any clarification needed to
consultants.

4.7.4 Additional requirements concerning shallow groundwater

drawdowns
The assessment of impact risks in the approvals process is presently based on minimum
distances between extraction and assets, maximum permissible drawdowns in groundwater
pressure heads at stipulated distances from assets, as well as guidelines on minimum
requirements for pump tests and their reporting.
Currently there are no explicit requirements for the proponent to demonstrate that there are
no detectable impacts at assets surrounding the site or detectable hydraulic connection
between extractions from (assumed) confined aquifers and overlying shallow groundwater
and surface water. However, monitoring of shallow aquifers is seen by many experts as a
practicable method of measuring whether pressure drawdown at the extraction point has
propagated towards assets at potential risk. In principle, this may be implemented as part of
the pump tests, and/or as a requirement of the approval with an associated cease-to-pump
criterion. This would require careful selection of appropriate monitoring points, which
represent the potential hydraulic connection between the extraction and assets at potential
risk, as well as a criterion for assessing acceptability of any observed effect. The
requirement for an adequate period of baseline data would be an additional consideration.
In projects involving large groundwater extractions, it is common for shallow groundwater,
surface water flow and/or GDE health indicators to be used to define triggers, whereby a
defined level of impact would trigger a cessation of pumping. In practice, such monitoring is
expensive to install and operate, requires a process for identifying acceptable impacts and
triggers, a process for reporting and auditing, may require access to private land and
increases investment risk for the proponent, due to increased uncertainty over continuity of
operations. Therefore it is most appropriate in potential high risk cases. In the case of
managing local impacts of high volume extractions on high-risk assets, a viable option may
be to monitor easily measured parameters such as shallow groundwater levels on the
property in question. However, in many cases the link between shallow groundwater levels
and surrounding off-property assets such as GDEs will be arguable.
As well as providing an increased (although not comprehensive) safeguard against impacts,
provide insight into factors like how the shallow system responds to rainfalls and the layering
of the system at the local scale.
It is also important to consider the appropriate authority for any additional monitoring
requirements. For example, if this monitoring is to be required as part of a development
consent condition, local government should consider how risk tolerances should be balanced

81
For example, the AIP comes with an application guide for consultants that includes a checklist, which allows the consultant t o
specify where in the hydrogeology report the AIP requirement is addressed. This might be considered for other applicable
guidance.

95
against imposing further obstacles to development. Further, if the monitoring requirements
for development consents are different from requirements of license dealings and works
approvals, this would require a separate guidance documents for proponents and their
consultants. Since the requirements regarding number, location and depth of shallow
groundwater monitoring will be project-specific, the guidance is likely to be about the
process, principles and general criteria rather than monitoring design.

4.7.5 Field verification

To address hydrogeological complexity, site investigations may be undertaken or required
prior to designing an impacts assessment. This may include exploratory drilling, and/or the
examination of drill logs from existing or new site bores before designing a pump test. This
preliminary work may result, for example, in a recommendation for multiple monitoring bores
(piezometers) in non-linear locations to understand better how depressurisation propagates
in three dimensions. Intermittent review and iterative improvements to monitoring may be
appropriate during this process.
Furthermore, temporal variability of groundwater pressures and surface flows means that
adequate baseline data are essential to isolate effects of new extractions. Baseline data
require foresight from project proponents, and also from asset owners/managers or the
government if they wish to ensure that baseline data exists for valued assets. These
baseline data requirements must balance the risks of the proposed extraction against the
value of the extended baseline data required as collecting this data can require significant
time and cost.
Methods to identify and classify vegetation can be designed to recognise GDEs and
estimate the extent of impact. Vegetation assessment methodologies, such as the
Bioregional Assessment or Biodiversity Assessment Methodology are useful approaches.
However, engagement of an ecohydrologist consultant to undertake field verification on
environmental assets would complement the assessment of potential impacts.

4.7.6 Improving data collection, accessibility and management

Data from the monitoring network at Alstonville is publicly available on the WaterNSW
website (https://www.waternsw.com.au/waterinsights/real-time-data). This website provides
real-time data on NSW river heights, streamflow, groundwater bores, meteorology and
rainfall, as well as dam and reservoir levels and volumes.
It was noted above that extraction data is not readily available, but would greatly assist to
better understand whether observed impacts on water levels and pressures are due to water
extraction or climate variability.
For data that can be made publicly available, there are state managed environmental
databases (e.g. SEED) that could be utilised.
The accessibility of any data is central and the preponderance of manual collection is an
impediment in this regard. Advances in technology to provide robust and tamper-proof
telemetering options that are commercially cost competitive would have a significant impact.
Current technology is also available to enable standardised templates and reports to be
managed electronically. This would improve the flow of information to relevant agencies and
other parties.

4.7.7 The potential role of local research studies

Knowledge of the groundwater requirements of GDEs is limited and research tends to be
conducted on unconfined aquifers or systems that are simpler than fractured rock systems.
Identification and requirements of subterranean GDEs and the contribution of groundwater to
surface water baseflow is even more limited. Local impact studies that monitor groundwater
drawdown on nearby GDEs may be considered where there is concern about the impacts

96
(e.g. near Tweed World Heritage site or areas of high extraction) of drawdown on the local
environment, surface water or other users.
Monitoring as part of a development consent or as a requirement of an approval can go
some way to detecting local impacts or increasing confidence that they are acceptable or
not, and to detecting local connectivity between deep and shallow groundwater at monitored
locations. However, it is likely to leave gaps in knowledge about local (~0.01-0.1km2) scale
processes and conceptual models, and practical and cost constraints mean it is unlikely to
provide new knowledge about catchment/regional-scale impacts (0.1-100km2). For these,
multi-year, multi-scale research projects would be needed.
Research should be encouraged, which monitors continuously at both the local scale and in
the surrounding catchment to understand sources of groundwater water, the transmission of
depressurisation due to pumping, and impacts on groundwater discharges. This would
ideally require voluntary participation of a bottled water operator and surrounding
landowners, including installation of continuous pressure transducers in pumped bores,
monitoring piezometers and shallow water bores, and availability of metered pumping rates.
It should also include tracer studies and surface flow monitoring, and potentially could
include ecological indicators.

4.8 CONCLUSIONS
Impacts
 Based on the assessment of available information and analysis undertaken by the
Review, there is no measured evidence that current bottled water extractions have
impacts on other properties’ bores, surface water or GDEs in the Northern Rivers
region. This is at least partly due to the relatively low current levels of extractions,
hydrogeological conditions and absence of monitoring capable of detecting these
impacts.
 Alstonville is the location that has the greatest level of extraction and has monitoring
that has been assessed, which provides confidence on the health of the groundwater
source. In the case of the Tweed area, while this has minimal monitoring, it also has
very low extraction levels for the water source overall – far below the allowable
extraction limits.
 While all groundwater extractions have impacts, the magnitude of those impacts and
potential consequences will vary. Whether these impacts are measureable, or are of
a magnitude to have detrimental consequences on an ecosystem or environmental
asset is the focus of monitoring and measurement that occurs both during the
assessment phase, and also during the operational phase for approved operations.
 There are significant complexities in measuring local impacts from water extraction
due to the spatial and temporal variability of the hydrogeology of fractured and
porous rock systems
 While there are existing approaches to measuring and modelling local impacts, these
have challenges in terms of accuracy, practicability and cost. Decisions about these
investments are also typically done in light of the risk that is being addressed – risk
likelihood and consequence.
 Bore water extraction can potentially impact water within the same aquifer, within a
connected aquifer, or within a connected surface water body, leading to possible
changes in water quantity and quality. The pump test is a common field technique,
used in hydrogeological assessments, to derive local scale aquifer properties and to
indicate proposed impacts of the extraction. In fractured rock systems, the fracture
network that intersects the point of extraction will determine the response to
pumping, which is complex and requires hydrogeological investigations and

97
 interpretation of results in order to design the pump test. Impacts may be proximate
to or at distance from the point of extraction, and occur vertically as well as
horizontally.
 Noting the low level of current groundwater monitoring in three of the four relevant
groundwater sources, there would be merit in reviewing the need for additional
monitoring that will provide the baseline data, conceptual hydrogeological models
and recharge estimates commensurate with potential future risk levels.
 At a regional scale, the cost of traditional monitoring bore infrastructure is likely to be
an ongoing challenge. This is particularly the case in fractured rock systems subject
to high hydrogeological variability. Emerging sensing technologies able to gather
data over large areas and at depth may provide a step-change to the field, subject to
cost and commercial availability. Whether at the local or regional scale the choice of
monitoring will be informed by the level of risk and the cost-effectiveness of the
monitoring. Local research studies may prove a useful adjunct.
 The assessment process for proposed extractions takes into account the risks of
local impacts through a risk assessment process, requirements for some applicants
for proposed medium and high risk extractions to submit a hydrogeological report to
support their application, and criteria for acceptable levels of local impacts.
 Local scale monitoring during extraction operations can assist with better
understanding of local hydrology and extractive impacts and consequences. This
may include piezometric monitoring of the pathway between the point of extraction
and locations where there is perceived risk. The cost of this monitoring is likely to be
a challenge and its requirement should be justified by the risks as identified by an
expert following analysis of pumping test data.
 Local scale monitoring during extraction operations could potentially support adaptive
management, for example, through additional reporting and cease-to-pump rules
related to observed groundwater pressures.
 The Review considered a number of past hydrogeological reports submitted to
support proposed extractions by the bottled water industry in the Northern Rivers
area as components of development applications. The hydrogeological reports, in
most cases, assess hydrogeological impacts of proposed developments on surface
flows, GDEs and surrounding groundwater bores. The focus and style of the reports
varies depending on the purpose and the identified or perceived risks.
 Both industry and decision makers would substantially benefit from greater clarity,
specificity and standardisation of requirements for hydrological reports. Current
technology is available to enable standardised templates and reports to be managed
electronically.
 Robust local assessment of potential connectivity between aquifer and overlying
shallow groundwater and surface water should form part of pump tests and feature in
hydrogeological reports. This is important, as observed in Alstonville, where deeper
aquifers are not necessarily confined and may have connections to surface systems
or shallower aquifers. It is important to increase understanding of how confined the
aquifer is, as assessment criteria of allowable drawdown differs between confined
and unconfined systems. In addition, field verification is an important part of the
process.
 The Review received consistent reports from the community and sometimes
neighbours of bottled water extractors about observed changes including
environmental effects of drying watercourses and loss of water from previously
productive bores. The Review has not identified scientific studies or other evidence
establishing a causal link between these observed effects and extraction specifically

98
 undertaken by the bottled water industry. Going forward, data from extraction bores,
together with monitoring bores (piezometers), local studies and other sources of
information should help improve knowledge of impacts from a range of sources.
Data
 Lack of extraction data is an impediment to establishing appropriate extraction limits
for individual bores, measuring impacts, and at a regional scale, development of
WSP and making determinations of available water. A state-wide metering policy for
qualifying groundwater works with bore diameters of 200mm and above will take
effect in the Northern Rivers region from 2023. Four of the bottled water extractors in
the region are currently required by the regulator to have meters installed.
 The accessibility of any data is central and manual collection can be an impediment
in this regard. Advances in technology to provide robust and tamper-proof
telemetering options that are commercially cost competitive would have a significant
impact.
 Making water extraction and monitoring data available in standardised formats
through open databases would benefit decision-makers, researchers and the general
public to better understand activities and impacts, including cumulative impacts at
local and regional scale. Approvals by relevant state and local government
authorities could include requirements that all hydrogeological data are published.
There are state managed environmental databases (e.g. SEED) that could be
utilised.
Decision-making
 As with any environmental, engineering, resource activity the proponents and
decision makers and regulators operate in a realm of imperfect information. This
leads to levels of uncertainty around data and information, however uncertainty need
not prevent decisions being made.
 There are a number of approaches and tools employed to reduce uncertainty with
regard to the assumptions, hydrological domain, impacts, consequences of water
extraction. These include taking conservative estimates, using multiple lines of
analysis, being judicious in decisions around the type and location of monitoring,
employing adaptive management approaches.
 There is a lack of clarity around water planning, management and decision-making
roles and processes at state and local government level and between relevant
authorities.
 Given the implications of the overlap between development consent and water
licences or works approvals, work should be undertaken to clarify roles to ensure
consistency, avoid duplication and address any gaps in the assessment and
approvals process.
 If Local government is to undertake hydrogeological assessment as part of the
development application process, then it needs access to relevant expertise to
interpret modelling and technical reports to inform its decision-making, including
requirements for development applications.
 Access to government and industry water data through a common open platform
housing standardised, well-curated and long-term data sets that can be expanded
would assist assessment and decision-making of applications.

99
5 TECHNICAL APPROACHES TO SOCIO-ECONOMIC FACTORS

Term of Reference 3a requests advice on the scientific and technical approaches to

examining socio-economic factors and impacts and possible solutions using locally relevant
examples. These related primarily to concerns about the potential growth of the industry,
issues associated with truck movements and the use of plastics. The first is dealt with in
Chapter 2, the latter two here.

5.1 TRANSPORT (TRUCKS AND ROADS)

Heavy vehicles are used to transport water from the extraction site. These vehicles comprise
either bulk water tank trucks or general-purpose freight trucks. The Initial Report also found
a total of 128 return truck journeys per week for the bottled water industry in the Northern
Rivers region.
Issues raised include: truck movements occurring outside of approved hours or number of
trips; dangers from the presence of large trucks on small roads, including in school zones;
potential for more significant harm from larger vehicles in the event of an accident; loss of
visual amenity and stress from truck-associated noise; and the scale of road damage and
maintenance requirements associated with large truck movements shouldered by the
broader community.

5.1.1 Frequency and hours of operation

The Heavy Vehicle National Law (HVNL) prescribes a national regulatory framework for a
nationally-consistent approach to heavy vehicle legislation (over 4.5 tonnes gross mass)
(NTC, 2019). The HVNL is the result of a collaborative process between industry and
government and is led by the National Transport Commission (NTC). In most states of
Australia, including NSW, the HVNL establishes the Heavy Vehicle National Regulator
(HVNR) as the regulator responsible for truck standards and on-road enforcement.
The HVNL regulates driver fatigue management by specifying heavy vehicle operations that
require the maintenance of a work diary with driving and rest times. This requirement is
linked to a heavy vehicle licence that is regulated in NSW through Roads and Maritime
Services (RMS) and NSW Police. Outside of driver fatigue management, the HVNL and
Regulations do not prescribe specific conditions that prohibit trucks moving within certain
times of the day or how many times they can traverse a region.
Development consent is required for the extraction of water from groundwater systems for
commercial bottling. In approving a development application, it is within the remit of local
government to apply conditions related to trucks and truck movements, including permissible
numbers of movements to and from properties and hours of operations, as well as to enforce
compliance. For all new or modified development applications for water bottling, a traffic
assessment may be required by council. Councils can only revoke or modify development
consents on a limited number of grounds. Where applicants seek to modify conditions of
approval, the conditions can be modified only on the issue the application seeks to amend.
A summary of conditions of approval for truck movements is provided in Table 16. Generally,
approved truck movements occur between 7:00 – 18:00 on weekdays, with shorter hours on
weekends. The number of truck movements per day is highly variable, ranging from two to
twelve trips per day. Local Government authorities are responsible for compliance with
conditions of development consents.

100
Table 16: Summary of conditions of approval for truck movements
Issue Current measures Responsibility of current measures

Timing Range: 7:00-18:00 (Monday to Friday) and Council (through the development
8:00-12:00 (weekends) consent)

Frequency Range: 2-12 trips (Monday to Friday) and 4-8 Council (through the development
trips (weekends) consent)

If operational approval under the relevant development consent includes school zone times,
trucks are legally entitled to operate during those hours. Trucks are held to the same speed
conditions in school safety zones as other road vehicles, and these conditions are enforced
by NSW Police.
Reviewing development applications currently pending, the Review identified conditions
requiring traffic report assessments for the primary road that the water trucks enter and exit
from the property, at major intersections and in swept paths within the immediate route of the
extraction property. However, the Review has not identified traffic reports being required for
historical water extraction licences at the time of assessment.

5.1.1.1 Potential solutions and technologies

Potential strategies available to councils to assist with ongoing monitoring of compliance with
conditions of approval include use of written or digital logbooks and electronic tracking.
Logbooks would enable collection of data such as odometer reading, location of operation
and timing. Electronic products linking data to a database or a phone application could also
facilitate accurate record keeping and ease of analysing results. Currently, the majority of
development consents for bottled water extractors currently active do not require the use of
logbooks to track specific truck movements in and out of the properties. However:
 An approval dating from 2016 stipulates “an annual statement of truck movements to
and from the subject site is to be supplied to council at the end of each financial year
to the satisfaction of the General Manager or his delegate”.82
 A development application currently under consideration includes a draft condition
that states “The movement of trucks off the site in accordance with this development
consent is to be maintained in a daily log which records the date and time of all
inbound and outbound trucks from the subject site. At any time, Tweed Shire Council
officers may request a copy of the log to be provided for audit of compliance with
conditions of this development consent in regard to the times and frequency of truck
movements in and out of the subject site…and is to be accompanied by a Statutory
Declaration…declaring that the information contained in the log is true and correct.”83
 Tweed Shire Council advised that any new or modified applications for water
extraction could have conditions requiring logbooks and the installation of security
cameras (CCTV) to assist in the monitoring of truck movements (TSC 2019, pers
comm., 20 September).
Alternative technologies for monitoring truck movements include active or passive vehicle
tracking systems utilising GPS navigation devices and software to collect data for a
comprehensive picture of vehicle movements.84 These systems would be useful in collecting
haulage logistics and transport data as they can provide precise and constant data on
average speed, distance, fuel consumption, driver time and location.

82
Development application DA06/1023 (condition 4.2; 10-20 Edwards Lane, Kynnumboon)
83
Development application DA16/0936 (Rowlands Creek Road)
84
Passive systems store data and are required to return to a predetermined point where the information from the device can be
downloaded and analysed, whereas active systems transmit the date in near-real time via cellular or satellite networks to a
computer or data centre for evaluation. Many modern devices will combine both active and passive tracking abilities so that if
the cellular network becomes unavailable, the device will store the data to the devices internal memory.

101
Most devices are installed in the vehicle; however, new technologies also enable mobile
phones to be used for tracking multiple variables. These tracking devices are readily
available and relatively inexpensive, ranging from approximately $100-$400 (depending on
functionality) plus monthly ongoing costs of $20-$30 per month for real-time tracking, history,
odometer readings etc.85
Traditional approaches, such as traffic surveys, could be used to gather a broader picture
about traffic, particularly on roads of concern. Traffic surveys have the capacity to collect
data on traffic volumes, vehicle passing probability, road widths and swept paths and can be
conducted as observational data during peak traffic times or conducted using Automatic
Tube Counts (ATC).86 Using ATC, data can be collected on the number of vehicles, speed,
vehicle types and platoon data by time intervals.87

5.1.2 Truck noise

Noise standards for heavy vehicles are prescribed at a national level under the Australian
Design Rules (ADRs). All Australian road vehicles must comply with the relevant ADR in
place at the time of manufacture and supply to the Australian market. New noise standards
for heavy vehicles took effect in 2005 and provide for a national standard for vehicle safety,
anti-theft, lighting, noise, engine exhaust emissions and braking.
The National Transport Commission (Model Law on Engine Brake Noise Limits) Regulations
2009 also impose limits on the level of noise emitted by engine brake devices. RMS enforce
these regulations through periodic inspections of heavy vehicles at testing stations to ensure
that silencers are fitted and maintained and ensure they meet all other noise requirements
as specified in the ADR and the HVNL. Current measures and responsible bodies
surrounding truck noise is summarised in Table 17.
Despite the regulations, the NSW Environmental Protection Authority (EPA) reports a high
instance of complaints from the broader NSW community regarding noise from engine or
compression brakes from all types of heavy vehicles (EPA, 2013), and RMS have advocated
for tighter vehicle noise stands. RMS and EPA report that noise from trucks braking can be
intrusive and it is advisable that heavy vehicle drivers should avoid using exhaust brakes,
engine compression or ‘Jake’ brakes near residential and noise-sensitive areas to help
reduce the stress associated with excessive truck noise (EPA, 2013).
Currently, a national scheme is being implemented to impose noise limits from engine
compression brakes using roadside noise ‘cameras’ as an aid to enforcement. There is
currently no legislative basis to issue fines for noise from engine compression brake use in
any jurisdiction in Australia. However these noise ‘cameras’ can be used to issue warning
notices to truck owners whose vehicles exceed the national engine brake noise in-service
noise standard (Parliament of NSW, 2012).

5.1.2.1 Potential solutions and technologies

Council can assist in controlling noise levels generated by trucks through the assessment
process and conditions of approval. For all pending or future applications to modify
development consents associated with the water extraction industry, a statement of
environment effects is required, possibly including an Environmental Noise Impact
Assessment. As vehicle noise regulations are set at a federal level, state and local
authorities have limited powers to regulate low noise technologies on the vehicles
themselves. However, other infrastructure technologies, for example sounds walls designed

85
These devices are commonly used across haulage companies. In Australia, Linfox, one of the c ountry’s largest logistics and
supply chain company deploys a GPS system to record a range of real-time data including road speed, engine RPM, fuel
efficiency, vehicle location, kilometres travelled, driver identification and engine fault codes and warnings. The data is captured
on a small digital recorder mounted in the front of vehicles, with data uploaded to a control room for analysis (Linfox, 2019)
86
Automatic Tube Counts (ATCs) detect the axles of vehicles using a rubber pneumatic tube to measure vehicle movements.
87
Platoon data measures the speed of groups of vehicles.

102
to reduce vehicle traffic noise in specific locations, are a technical option that could be
considered.

Table 17: Summary of regulation relating to truck noise

Current measures Responsibility of current measures
Restrictions on truck movement frequency, route Council (through the development consent)
and hours of operation
Engine noise standards during manufacture Australian Design Rules
Heavy vehicle noise level limits Heavy Vehicle National Law (National Transport
Commission)
Heavy vehicle inspections NSW Police, NSW Roads and Maritime Services

5.1.3 Truck size and road damage

A summary of development consent conditions related to truck size is provided in Table 18.
The allowable truck size for water extraction under development consents varies across
extractors, but generally ranges from 6 to 9 metre long B-double trucks and infringements
can be issued if conditions are breached.
The NHVR classifies vehicles of this size as ‘General Access Heavy Vehicles’.
Table 18: Summary of conditions of approval for truck size
Issue Current measures Responsibility of current measures
Safety Range: 6m to 19m B-double truck Council (through the development consent)
Heavy vehicle standards on horns, mirrors, lights Heavy Vehicle National Law (National
and reflectors Transport Commission)
Damage Registration cost for operating a heavy vehicle Proponent/ NSW Roads and Maritime
Services
Road contribution plan Council (through the development consent)
Range: $1200 - $ 17536
Road maintenance (state and council roads) NSW Roads and Maritime Services and
Council
Heavy Vehicle Safety Stations NSW Roads and Maritime Services

The NHVR provides specifications and standards for all heavy vehicles registered on NSW
roads to ensure the safety of the trucks and other road users within the community. In terms
of safety, these include specific standards on horns, vision mirrors and lights and reflectors.
Water trucks are subject to the same regulation.88 Under national mass and loading
arrangements, General Access Heavy Vehicles have unrestricted access to the road
system, except where a road or bridge is sign-posted otherwise. Provided these vehicles
have current registration appropriate to the vehicle configuration, no specific access
restrictions or additional safety precautions apply and no additional permits are required
(RMS, 2019).
RMS also has a compliance program to inspect heavy vehicles that may be operating in an
unsafe manner on NSW roads. Heavy Vehicle Safety Stations (HVSS) inspect the mass,
dimension and loading of a heavy vehicle and ensure it is compliant with the vehicles
registration.
The cost associated with road damage caused by heavy vehicles falls to State and Local
Governments and the proponent of a development consent in accordance with the

88
Note that although additional safety precautions are required for vehicles that are considered ‘Oversize Over Mass (OSOM)
Vehicles’, the size of all approved water trucks across the LGAs within the Northern Rivers region meet the OSOM vehicle
criteria (TSC, 2019a)

103
conditions. State registration charges for heavy vehicles aim to recover expenditure on roads
from trucks and ensure safe roads for all road users. This registration fee, which includes a
regulatory component, is collected through the RMS. The quantum of the fee depends on
several factors including the number of axles, gross vehicle mass and what category of
vehicle it falls into. These charges are contained in the Heavy Vehicle Charges Model Law. 89
Although passenger cars account for a high proportion of vehicle-kilometres travelled in
Australia, trucks make a greater contribution to pavement damage (Bureau of Transport and
Communications Economics, 1997). Pavement damage attributable to a specific vehicle
depends on a number of factors including roadway design as well as weight and axle
configuration. A commonly agreed method to approximate the relative impact of different
categories of vehicles on roads is through the ‘Generalized Fourth Power Law’, which
predicts that the change in pavement damage is proportional to the difference in the vehicles
axle weight to the fourth power (Freight on Rail, 2019). However, determining the number
and types of wheel/axle loads that a particular pavement is subject to in any given time is
more complex.
Road ownership generally determines the authority responsible for road maintenance
charges. In the Northern Rivers region most roads are owned by local government, with only
a few large inter-passes that are state owned. Some council roads that are strategic to
traversing the region and have high use may receive a state contribution for their
maintenance.
Councils can adopt Road Contribution Plans (RCP), as a mechanism to collect contributions
from developers to support public road infrastructure. 90 The RCPs include an additional
component relating to vehicle weight.91 The Review found that development applications for
bottled water extraction approved over the last decade include a contribution under the RCP.
Tweed Road Contribution Plan modelling found that the Tweed Shire road network will
experience considerable traffic growth, especially on the Tweed coast and in urban areas, as
a result of the anticipated urban development and that most major urban road corridors will
be required to carry considerably more traffic (TSC, 2016).

5.1.3.1 Potential solutions and technologies

Where feasible, increased council or police presence or random spot checks at extractors
during operational hours may assist to regulate truck size compliance and alleviate
community concern that larger than authorised trucks are being used. Technologies
described in previous sections would also be relevant.
Councils can continue to exercise their powers in relation to truck size when approving
development applications.

5.1.4 Conclusions
 There are technologies available that can provide accurate, consistent and real-time
data on truck movements, which could be included as a condition of the development
consent.
 Responsibility for governing truck safety, movements and size spans Federal, State
and Local Government authorities. Each of the responsible bodies has measures to

89
As implemented by each jurisdiction .The heavy vehicle registration fee is based on the pricing principles set by the Transport
and Infrastructure Council and the Council of Australian Government (COAG) and undergoes annual adjustments (Transport
and Infrastructure Council, 2017).
90
Section 7.11 of the Environmental Planning and Assessment Act 1979 is the principal legislation enabling Councils to levy
development contributions for public amenities and services. A monetary contribution can be imposed by a way of a condition
of development consent and can be in the form of a Road Contribution Plan.
91
For example, the Tweed Road Contribution Plan (s 6.5) includes a heavy haulage fee based on a formula comprising the
value and life of pavement.

104
 regulate and monitor heavy vehicles through existing legalisation, approval of
applications and technologies.
 Technologies and strategies are available to measure traffic volumes and impacts.
Local government can levy heavy vehicle road users to contribute to the cost of road
maintenance and repair.

5.2 PLASTIC BOTTLES

The Initial Report included concerns expressed during consultations about the environmental
impacts of plastics used in the bottled water industry. These views were reiterated in
submissions received by the Review.
The issue is international in scope and management of the impacts and solutions will be
influenced significantly by factors and developments beyond the Northern Rivers region.

5.2.1 Extent and management of plastic

The main polymers that plastic bottles for drinking water are produced from are Polyethylene
terephthalate (PET) and High density polyethylene (HDPE) (PricewaterhouseCoopers, 2008;
Locock et al., 2017). PET and HDPE are predominately made of non-renewable sources
such as oil or gas (CIEL, 2017) although recycling can collect waste PET and HDPE for
reprocessing (recyclate). New plastic water bottles can contain PET recyclate (Locock et al.,
2017).
Across Australia, 3.4 million tonnes of plastics were consumed (Envisage Works, 2019) in
2017-18 and approximately 58% of total plastic packaging generation was disposed to
landfill from collection (Madden & Florin, 2019).92 The recycling rate of PET in Australia has
increased from 16% in 2016-17 to 21% in 2017-18 (Envisage Works, 2018, 2019).93 In the
same period, consumption increased from 345,600 tonnes to 355,300 tonnes, 32% of which
was from NSW (Envisage Works, 2018, 2019).
NSW, together with the Commonwealth and other jurisdictions has committed to establishing
a timetable and strategy to ban the export of key waste materials, including plastic; reduce
plastics waste and diversion to landfill; and build capacity to recover value from waste and
generate high-value recycled products (COAG, 2019). This follows a 2018 commitment for
100% of Australian packaging to be recyclable, compostable or reusable by 2025 and 70%
of plastics to be recycled or composted (Waste Management Review, 2019). The Australian
Packaging Covenant Organisation (APCO) manages the national product stewardship
scheme for the delivery of the sustainable packaging pathways in Australia, and is leading
the delivery of the 2025 National Packaging Targets.94
A nationally harmonised approach will facilitate implementation of NSW policies and
initiatives. The NSW Circular Economy Policy Statement ‘Too Good To Waste’ was released
in February 2019 (EPA, 2018), building on the NSW Waste Avoidance and Resource
Recovery Strategy 2014–21 (EPA, 2014) and associated programs. These strategies are
driven by the most efficient approaches for resource use, guided by the waste hierarchy to
avoid and reduce waste in the first instance, followed by options to reuse and recycle waste,
to recover energy and treat waste and finally to dispose of waste (EPA, 2014).

92
Consumption is defined as the Total use of product by Australian industry and consumers. It includes locall y made and used
product, imported product and locally utilised recyclate. Does not include locally made product that is exported for sale.
93
The Recycling rate is determined from the extent of recyclate sent to plants domestically and internationally for r e-processing
94
APCO is a co-regulatory not-for-profit organisation administering the Australian Packaging Covenant, an agreement between
Federal, state and territory governments and the packaging industry to reduce the harmful impact of packaging on the
environment. Plastics in packaging accounts for 60% of the plastics waste stream. National Packaging Targets to be achieved
by 2025 are: 100% reusable, recyclable or compostable packaging; 70% of plastic packaging being recycled or composted;
30% of average recycled content included in packaging and the phase out of problematic and unnecessary single-use plastics
packaging (APCO, 2019d).

105
The NSW Government is currently developing a NSW Plastics Plan as well as a 20 Year
Waste Strategy. In 2017, the NSW Government Return and Earn Container Deposit Scheme
was introduced, resulting in the return of 750 million drink containers and a 33% reduction of
eligible containers in the litter stream in one year (Blue Environment, 2018). Over January to
June 2019, sorted PET plastic containers made up 28.1% of the volume collected across
NSW, while HDPE made up 1.4%. A number of collection points are located in the Northern
Rivers and there has been an increase in the return of containers since 2017 (Figure 23).

Figure 23: Return and Earn collection volumes of sorted PET plastics in select areas
Note: data includes Reverse Vending Machine, Automated Depot return volumes and disaggregated Over The Counter weekly
return volumes collected through the Network Operator Exchange for Change

5.2.2 Bottled water industry – current practices and commitments

In the Northern Rivers region, proponents extracting water for the bottled water industry
either sell the water on to a bottler and/or bottle the water themselves. Large companies,
CCA, Asahi Holdings (Australia) Pty Ltd and Nu-Pure Pty Ltd account for over 70% of the
market share of Australian bottled water products (IBISWorld, 2019). Both CCA and Asahi
are signatories to the Australian Packaging Covenant, and were graded as level four
(leading) out of five against the 13 criteria of the Sustainable Packaging Framework (APCO,
2019a, 2019b).
Coca-Cola Amatil has committed to reducing the amount of waste sent to landfill by making
100% of all packaging recyclable by 2030 (APCO, 2019b); has targets for increasing the
recyclability of on-pack labelling, light-weighting bottles (i.e. reducing the amount of plastic
used in containers) (APCO, 2019b); and increasing the amount of recycled plastic in PET
bottles (Coca-Cola Amatil, 2018b). In 2019, the company announced seven out of 10 of its
bottles manufactured for the Australian market would be 100% recycled plastic by the end of
the year.95 Asahi’s Action Plan to meet the 2025 Targets include more than 50% of primary
packaging to be recoverable, and improving material efficiency in up to 20% of products
(APCO, 2019a). Asahi have commenced blowfill manufacturing to produce lightweight PET
CSD bottles (Asahi Beverages, 2016). The Nu-pure spring water range uses 50% recycled
PET (Nu-Pure, 2019).
There are smaller operators in the Northern Rivers who both extract and bottle, who use a
range of plastic bottle products. While some use small plastic PET bottles (350-650ml); other
operators are using options that may encourage minimisation of plastic production and
waste. This includes use of large plastic containers designed for water coolers (up to 19L),

95
This includes the Mount Franklin and Pump water brands. See https://www.marketingmag.com.au/news-c/news-coca-cola-
recycled-bottles-2019/

106
that are returnable or with a refundable deposit fee per bottle; use of plastic bottles with the
addition of a compound (reverte) to encourage oxo-biodegradation and substitution of glass
for plastic.

5.2.3 Opportunities and challenges

To achieve the 2025 National Packaging Targets, a whole of supply chain approach is
required to focus on design, collection and recycling systems, reuse, materials circularity and
consumer engagement (APCO, 2019f). It is anticipated that this period of transition for
industry and consumers will see waste management initiatives implemented in a series of
stages that are in line with a global shift to reduce waste entering landfill and improve
recycling.
Globally, identified opportunities and challenges include improving design to avoid pollutant
materials and minimise reprocessing challenges; improving recycling capability; increasing
the quality and value of processed material; substitution and decoupling plastics from non-
renewable resources (Ellen MacArthur Foundation, 2014; DEE, 2018).
APCO is leading a number of pieces of work on Australian used packaging volumes and
resource recovery capabilities. Due for release later this year, these include reports on
recycling infrastructure and emerging and at-market technologies to manage a range of
waste streams, including plastics. These reports will provide a more granular picture than
currently available at both national and state levels and will be a key input to strategies going
forward.
Better characterisation of waste streams, understanding of material stream flows and
establishing mechanisms to support market development will also underpin uptake of
technologies.96 The Institute of Sustainable Futures at the University of Technology Sydney
was commissioned by APCO to provide a packaging material flow analysis (Madden &
Florin, 2019) and developed a packaging sustainability framework that has been
implemented as an online self-assessment tool for companies (Kelly et al., 2017).
Design choices at the manufacturing stage would help reduce contamination as recyclable
containers that use different polymers for caps, adhesives and labels can present sorting
challenges. Eliminating polymers and certain colours that are problematic for recycling would
also improve recycling efficiencies as well as improving collection and sorting into clean
single-polymer streams.
In Australia the majority of used plastic packaging that is disposed to landfill is not collected
for recovery (Madden & Florin, 2019) and contamination also results in large volumes of
recyclable plastics going to landfill (DEE, 2018). A large source of contamination is due to
the handling and compaction of glass in co-mingled collections, causing the glass to break
and producing glass fragments that are difficult to separate (Madden & Florin, 2019). Plastic
bottles placed in co-mingled recycling bins for municipal kerbside collection are generally
sent to Materials Recycling Facilities (MRFs) and are sorted into either a mixed-plastics
grade, or more commonly three grades: PET, HDPE and other residual mixed plastics (DEE,
2018).
MRFs vary in technological sophistication, most using a mix of human, mechanical and
electronic processes to sort co-mingled recycling, and lack capacity to sort highly
contaminated waste (DEE, 2018). Of 12 reprocessing plants Australia-wide, three in NSW
are reprocessing PET (DEE, 2018). There are opportunities to improve collection and
processing infrastructure to reduce contamination and improve the quality and value of
recycled materials, including through the use of automated optical sorting equipment (DEE,
2018).

96
In May 2019, the Federal Government committed $1.6 million to develop an online platform and marketplace and a further
$9.2 million in August 2019 to advance innovation in plastics recycling under Round 8 of the Cooperative Research Centres
Projects program (CRC-P).

107
PET recycling is mature technology and driven by markets for recycled PET overseas and
locally. There is significant local capacity capable to recycle PET to food grade and there are
projects in train that will increase this capacity. The capacity for recycling HDPE back into
food grade remains limited compared to the volumes consumed. 97
The market for recyclate is also influential, fluctuations depending on manufacturing
capacity as well as oil and gas prices for the production of new plastic (Locock et al., 2017).
Many of the NSW Government 2018 Waste Less Recycle More grants will address the
issues of increased restrictions on export of recyclate, cross contamination of recyclables;
and will initiate opportunities to process recyclate for new purposes and update MRFs. 98

5.2.3.1 Alternatives and additives

Alternatives to plastic bottles being explored by the global bottled water industry include
glass and paperboard (Carton & Co Water, 2019). Substitution with glass may be influenced
by market forces e.g. due to the lower cost of importing bottles (Meldrum-Hanna, Davies, &
Richards, 2017; DEE, 2018), the quality of glass collected by municipal kerbside recycling
(DEE, 2018) and higher transport costs (freight and energy) associated with glass (Meldrum-
Hanna et al., 2017).
Bioplastics are produced from renewable sources, usually from vegetable fats or corn
starch.99 Currently, over 75% of bioplastics on the market are non-biodegradable
(Australasian Bioplastics Association, 2019). Bioplastics can be designed to be either
recyclable or compostable, but not both. Developing a consistent labelling system and
avoiding compostable packaging contaminating the recyclables stream (and vice versa) has
been identified as a priority (APCO, 2019c; Australasian Bioplastics Association, 2019). Most
compostable bioplastics require specific conditions; as a result, the speed of biodegradation
is different in industrial composting plants compared to home composts.
Plastic bottles may also be made with conventional polymers that have an oxidising additive
to assist with degradation. These are referred to as oxo-degradable, hydro-biodegradable or
oxo-biodegradable plastics (PACIA, 2007). Issues include establishing the timeframe in
which complete degradation occurs in order for a biodegradable product to have an
environmental benefit (European Commission, 2018) and concerns about oxo-degradable
products (European Commission, 2015; Selke et al., 2015; European Commission, 2018). In
Australia, APCO has identified working towards a ban on oxo-degradable plastics as a
priority project as these plastics are included in problematic and unnecessary packaging to
be phased out by 2025 (APCO, 2019e).

5.2.3.2 Emerging technologies

There are increasing examples of, and opportunities for, end of use applications for plastic
where strong, light-weight material is required. Science and engineering advances in
materials, chemistry and synthetic biology also have a role in improving design, substitution
and reuse. While some developments are at early stages, many are in scale up to
commercialisation stage. Research and development efforts to replace, repurpose and
recycle plastics is a fast-moving and evolving space that is predicted to show significant
growth within the next 5-10 years.

97
Australian manufacturers are required to comply with the Food Standards Code that outlines standards for plastic materials in
contact with food. The Code refers to a voluntary Standard (AS 2070:1999) which states that post-consumer recycled material
is not to be used in direct contact with food. However, it is understood the Code is currently under revision, and the reference is
likely to be removed.
98
Through this program, Lismore City Council will undertake technological updates to reduce cross contamination of
recyclables at an MRF that services four council areas in the Northern Rivers region (DPIE, 2019b).
99
Bioplastics can also be made from agricultural by-products, used plastic bottles and other containers using microorganisms.

108
This capacity for growth was highlighted in the ‘Innovation in the NSW environmental goods
and services’ report. The report found that the environmental goods and services sector
valued at $43.9 billion in 2017-2018, is growing at a rate of 7.1% per annum, faster than the
wider economy, with exports worth $3 billion (NSW Innovation and Productivity Council,
2019).
Emerging research in materials engineering using Green Steel technology was invented at
the Centre for Sustainable Materials Research and Technology (SMaRT) at UNSW. The
Polymer Injection Technology substitutes coke with waste, using old tyres and plastics to
provide a source of carbon to replace a significant proportion of the non-renewable coke
used to make steel in electric arc furnaces. The SMaRT facility is also home to Green
Microfactories™ designed to reuse and repurpose materials including transformation of
plastic waste into high-quality 3D printing filaments. The recently announced ARC Research
Industrial Transformation Research Hub for Microrecycling of Battery and Consumer Wastes
opens up numerous new pathways to leverage high temperature process to access the
wealth of resources embedded within complex wastes such as metals, plastics and glass.
Instead of becoming landfill, they will be transformed into valuable materials and products,
including metallic alloys, oxides and carbon.
In the field of Chemical engineering, the CSIRO Chemistry and Polymer Research Group is
scoping projects to substitute materials for plastics, product development and the
development of new materials (additives) to aid polymer reprocessing.
Licella and the University of Sydney have co-developed a technology called the Catalytic
Hydrothermal Reactor, or “Cat-HTR”, to chemically recycle End-of-Life Plastics. Cat-HTR
breaks plastics down into smaller hydrocarbon components, using water at high temperature
and pressure and a mix of catalysts to stabilize the break-down products, preventing the
intermediate radicals from reacting with each other. The resulting liquid and gas products
can be readily upgraded using existing hydrocarbon refining and blending infrastructure into
useable products such as high-value waxes, lubrication oils, fuels, chemicals and
gases. Since the hydrocarbon products closely resemble the crude oil from which the
plastics were made, the Cat-HTR products can also be further cracked and refined to
monomers from which new plastics can be made, providing a circular economy advantage to
the technology. A large pilot plant with commercial scale reactor modules is established on
the NSW Central Coast. The first commercial Cat-HTR plant is currently under development
in Wilton (North East UK) and will convert 20,000 tonnes of End-of-Life Plastic annually.
Similar commercial plants are under development in Australia.
Synthetic Biology is an emerging field where complex artificial biological systems are
engineered. This research can be applied to the generation of plastics using genetically
engineered microbes to replace polymers from petrochemical sources. A recently approved
Australian Research Council Centre of Excellence in Synthetic Biology headquartered at
Macquarie University is focusing on converting biomass from agriculture or waste streams to
a range of products including bioplastics, building on work previously undertaken through the
international Yeast 2.0 consortium.100 While developments in this space may not be
commercially viable in the short term, the university has industrial partners undertaking pilot
projects to develop cost-effective manufacturing solutions at market scales.

5.2.4 Conclusions
 The presence and management of plastics is international in scope and management
of the impacts and solutions will be influenced significantly by factors and

100
The Yeast 2.0 project is a global partnership focused on utilising synthetic biology tools to build the world’s first synthetic
eukaryotic genome”. See www.mq.edu.au/research/research-centres-groups-and-facilities/centres/synthetic-biology-
consortium/our-projects.

109
 developments beyond those extracting water for bottling purposes in the Northern
Rivers region.
 The NSW Government is developing a 20 year waste strategy and plastics plan in
the context of broader Federal Government and inter-jurisdictional commitments to
address waste and transition from linear to circular economies.
 There is a NSW Government container deposit scheme that has resulted in a one-
third reduction across the state of eligible containers, including bottles entering the
litter stream.
 Research and development efforts to replace, repurpose and recycle plastics is a
fast-moving and evolving space that is predicted to show significant growth within the
next five to ten years.

110
6 FINDINGS AND RECOMMENDATIONS

This Review has occurred during a period of widespread concern and public debate about
extended drought and long-term water futures. The Review recognises the community
concern about water allocations and use, and the desire for greater certainty and more
definitive information to inform decision making.
Yet all decisions are made in the context of imperfect knowledge. In groundwater studies
and management, there will always be a level of uncertainty associated with predictions and
a precise value may not be achieved due to the complex and heterogeneous nature of
groundwater movement. This is particularly evident in fractured rock systems that are difficult
to characterise fully.
The question of how to manage risk and uncertainty optimally in relation to water resources
is long-standing. Managing risk relies on efforts of the proponent, the regulator, state
agencies and local government and other stakeholders. Policy and regulatory instruments
provide a framework and strategies to help manage risk and reduce impacts. These include
adaptive management, risk assessment of proposed developments and approval conditions
for licences and development applications.
However, within these frameworks, judgements still need to be made. Relevant are risk
appetite, context, available information, potential consequences and the degree of
confidence in the assumptions made. Also important is access to tools that can be drawn on
to reduce uncertainty and manage risk in a way that is cost effective and proportional to the
level of risk.

6.1 FINDINGS
6.1.1 The bottled water industry
 Available industry data indicates that across Australia, over three-quarters of bottled
water is sourced from underground wells, and the remainder from standard
reticulated water supplies. Approximately 8% of Australian bottled water production is
exported.
 The Review identified seven operators in the Northern Rivers region with allocations
of 240.5 ML/y who are actively extracting for water bottling purposes, representing
0.55% of water licences and basic landholder rights (together defined in the WSP as
‘total water requirements’) and 0.008% of estimated total annual aquifer recharge in
the four groundwater sources.
 Four further proposals, if approved, would amount to an additional 168 ML/y, being
an additional 0.38% of estimated total water requirements and 0.006% of total annual
aquifer recharge.
 Changing consumer preferences, trade imbalances, the availability of tap water and
private (‘no name’) brands and population growth are expected to impact future
bottled water production and consumption volumes.
 Scenario analyses conducted by the Review suggest the Australian bottled water
industry is most likely to grow at a rate of less than 2% per annum to 2024 and that
growth in the Northern Rivers region is likely to be consistent with this trend. Under
most scenarios to 2024 considered, the 168 ML/y of additional proposed bottled
water operations would be sufficient to meet fully projected growth in demand.

111
  The Review also considered ‘highly unlikely’ and ‘extremely unlikely’ scenarios to
2034, being growth continuing at the current rate of 10% per annum and
establishment of a major premium bottled water exporter in the Northern Rivers,
respectively.
o If the ‘highly unlikely’ scenario occurred, the bottled water industry would
represent less than 2.3% of ‘total water requirements’ and 0.034% of
estimated total annual aquifer recharge.
o If the ‘extremely unlikely’ scenario occurred, the bottled water industry would
represent less than 4.6% of ‘total water requirements’ and 0.069% of
estimated total annual aquifer recharge.
 As the scenario analyses considered an unchanged regulatory and policy
environment, these forecasts may be affected by regulatory intervention which
directly or indirectly impacts the bottled water industry in this region.
 For the purposes of water extraction licensing, the bottled water industry is treated
the same as other prospective commercial users. However, development consent
under the Environmental Planning and Assessment Act 1979 is required for water
bottling activities. Approvals identified by the Review for bottled water extraction in
the Northern Rivers region date from 1993.

6.1.2 Allocations
 The WSP determines the allowable extraction limit, set from the recharge value of
each aquifer, with an amount of the recharge reserved for the environment and the
reminder determining the Upper Extraction Limit or the LTAAEL
 Under the WSP, environmental water and basic landholder rights are given priority
over licensed water extraction. Among licensees, priority is given to water utilities and
licensed stock and domestic over commercial licensed purposes.
 At the commencement of the WSP for the four groundwater sources, 100% of
storage is reserved for the conservation of the groundwater system.
 Water available for extraction is a portion of the estimated recharge value for each
groundwater source. This is determined by the WSP. An amount of the recharge is
reserved for the environment. The amount reserved for the environment equates to
97% of the estimated recharge value for New England Fold Belt Coast, 96% for
North Coast Volcanics, 82% for Alstonville Basalt Plateau and 48% for Clarence
Moreton Basin.
 The remaining amounts can be allocated for licensed purposes. Of these amounts,
38.0% of the New England Fold Belt Coast is allocated, 51.3% of the North Coast
Volcanics and 1.7% in the Clarence Moreton Basin. Alstonville is fully allocated.
 These are average values over the groundwater source areas; which means that the
environment is not protected to these levels in locally impacted areas.

6.1.3 Water Sharing Plan assumptions and uncertainty

 In groundwater studies and management, there will always be a level of uncertainty
associated with predictions (e.g. recharge rates) and a precise value may not be
achieved due to the complex and heterogeneous nature of groundwater movement.
This is particularly evident in fractured rock systems that are difficult to characterise
fully.

112
 The WSP plan was developed based on the best available data at hand and followed
a standard procedure. The assumptions made in the WSP are practical, reasonable
and in agreement with standard practice. In general, the WSP incorporates a
reasonable level of conservatism for extraction limits based on the risks identified. .
 The rainfall data used and the methodologies are sound and apply limited uncertainty
to the extraction rates.
 The portion of the estimated recharge value available for extraction is a function of
rainfall recharge over low environmental value areas together with an assessment of
environmental and socio-economic risk.
 Calculating recharge is complex due in part to the variability and complexity of the
hydrogeology and limited knowledge of the systems. Based on the analysis, the
Review considers the recharge rates used in the WSP are reasonable and
conservative. This statement is made with a relatively low level of confidence due to
lack of data for the groundwater sources of interest.
 In practical terms the groundwater sources are treated as geologically homogenous
which adds uncertainty and would benefit from further work. The Review recognises
that the complexity of the geology makes it difficult to incorporate heterogeneity into
the WSP recharge calculations. Particular attention should be given to the effects of
geological variability within groundwater sources, and soils and vegetation overlying
aquifer outcrops. The Review acknowledges the conservatism incorporated into the
current WSP through the allowable allocation figures.
 There is evidence to suggest that for the WSP recharge variable, there is a wide
range of values that can be applied as well as a number of different approaches to
calculate it. Limited field data is available to support a single estimate.
 Recharge rates applied to the four groundwater sources in scope in the WSP ranged
from 4% - 8% with studies and alternative methods indicating, with considerable
uncertainty, levels between 1% and 31%. The calculations by the Review using CBM
and baseflow filtering for recharge rates had results mostly above the values used in
the WSP. The Review noted the important contribution that surface conditions and
soil could make to the recharge of the underlying geology.
 The Review tested a scenario in which the recharge rates were reduced by 80%. It
found the recharge reserved for the environment for the New England Fold Belt
Coast and North Coast Volcanics would remain at around 80% of recharge. For
Alstonville, it would be reduced, but the network of monitoring piezometers provides
the ability to monitor long-term changes in levels. For the Clarence-Moreton Basin,
with only around 1.5% of available water allocated, the LTAAEL could be reduced
with no impact on licences.
 Based on the analysis, the Review considers the recharge rates used in the WSP are
reasonable and conservative. This statement is made with relatively low level of
confidence due to lack of data for the groundwater sources of interest.
 The application of the sustainability index appears to be a cost and time effective risk
tool that is applied as an additional means to protect resources where limited
information is available.
 The WSP incorporates a reasonable level of conservatism for the extraction limits
when the groundwater sources are not fully allocated and where they are fully
allocated at Alstonville, monitoring is applied.
 Additional monitoring in strategic locations in the Tweed would help inform gaps in
knowledge on a regional scale and provide a path towards better conceptual
understanding of aquifer flows.

113
  The overall system is managed with some level of adaptive management, including
an annual determination of the volume of water per licence share and WSP are
subject to an interim review at five years with a full review at ten years.
 Impacts of climate change should be considered in future WSP methodologies. A
warming climate can lead to increases or decreases in rainfall, variations in the
timing and frequency and strength of rainfall events, and increases or decreases in
evapotranspiration. The development by the NSW Government of Regional Water
Strategies will provide further insights into the impact that climate change could have
on the region and catchments over the coming decades, which can further inform
management approaches for the region’s water resources.

6.1.4 Sustainability of Water Sharing Plan extraction limits

 Due to limited extraction levels (where known allocations in the Tweed region are
much lower than the extraction limits contemplated in the WSP), limited data and
uncertainties described regarding the WSP parameters, it is not possible to conclude
whether the extraction limits are currently sustainable. However, the Review found no
evidence at this point in time that current WSP extraction limits are not sustainable.
 For the Alstonville Basalt Plateau Groundwater Source, which is fully allocated, and
there is a network of monitoring piezometers, data from 2006 onwards was analysed
by the Review, which concluded:
o The deeper piezometers (depths greater than around 25 m) showed a greater
stability and a steady upward trend over time of groundwater levels and/or
pressures. In contrast, the shallower piezometers showed greater variability
and appear to be recharged regularly with rainfall.
o Lagged rainfall is an important variable for understanding piezometer water
levels in the Alstonville Plateau. This was observed in shallow-sited
piezometers and in deeper piezometers sited in systems that are well
connected to surface waters and upper aquifers.
o There is limited amount of information available on current actual extraction
volumes. The Review notes that enhanced metering requirements will come
into force in the region in 2023 for eligible groundwater extractors. Given this
lack of data on extraction volumes, it is difficult to separate the effects of
environmental variables (such as rainfall) from the impacts of human
extraction (which tends to increase during dry periods).

6.1.5 Methodological improvements

 The Review considers there is room for improvement in the future assessment of the
variables underlying the extraction limits.
 Impacts of rainfall patterns in the region on recharge should be considered in future
WSP methodologies, including changing patterns associated with climate change.
 Particular attention should be given to assessment of groundwater recharge rates
across broad spatial areas and the associated need to distinguish between confined
versus unconfined aquifers.
 Developing a better conceptual understanding of the geological strata in the WSP to
reduce the level of uncertainty in the estimated recharge values. This could possibly
be undertaken via a 3D geological modelling tool (e.g. Leapfrog Geo) where there is
sufficient data and should include some soil mapping. This would require a large
scale detailed geological mapping survey or the collation of the existing core log data
and geophysical measurements, where available. The Review notes this level of

114
 detail has not been typically applied in similar WSP for easterly flowing rivers and
would require allocation of time and resources.
 Sensitivity testing could be undertaken to see whether a change in the recharge or
sustainability index might result in the aquifers being over allocated or stressed.
 Further work could be undertaken to assess whether the risk ratings given to specific
groundwater sources are appropriate.
 Water levels in the Department’s piezometers should be regularly assessed to
ensure periods of sustained water level decline are identified early. With further
analysis, it may be appropriate to set trigger values for water levels in key deep
aquifers that allow for adaptive management of groundwater extractions (e.g. once
water levels fall below a certain level, restrictions may be placed on extractions in
that area).
 Where the system is fully (or near fully) allocated, additional monitoring/sampling and
routine data analyses could be applied, as was undertaken at Alstonville, within an
adaptive management framework

6.1.6 Assessment and management of potential impacts from water

extraction
 Based on the review of available information, there is no measured evidence that
current bottled water extractions have impacts on other properties’ bores, surface
water or GDEs in the Northern Rivers region. This is at least partly due to the
relatively low current levels of extractions, hydrogeological conditions and lack of
monitoring detecting these impacts.
 While all groundwater extractions have impacts, the magnitude of those impacts and
potential consequences will vary. Whether these impacts are measureable, or are of
a magnitude to have detrimental consequences on an ecosystem or environmental
asset is the focus of monitoring and measurement that occurs both during the
assessment phase, and also during the operational phase for approved operations.
 There are significant complexities in measuring local impacts from water extraction
due to the spatial and temporal variability of the hydrogeology of fractured and
porous rock systems
 While there are existing approaches to measuring and modelling local impacts, these
have challenges in terms of accuracy, practicability and cost. Decisions about these
investments are also typically done in light of the risk that is being addressed – risk
likelihood and consequence.
 Bore water extraction can potentially impact water within the same aquifer, within a
connected aquifer, or within a connected surface water body, leading to possible
changes in water quantity and quality. The pump test is a common field technique,
used in hydrogeological assessments, to derive local scale aquifer properties and to
indicate proposed impacts of the extraction. In fractured rock systems, the fracture
network that intersects the point of extraction will determine the response to
pumping, which is complex and requires hydrogeological investigations and
interpretation of results in order to design the pump test. Impacts may be proximate
to or at distance from the point of extraction, and occur vertically as well as
horizontally.
 Noting the low level of current groundwater monitoring in three of the four relevant
groundwater sources, there would be merit in reviewing the need for additional
monitoring that will provide the baseline data, conceptual hydrogeological models
and recharge estimates commensurate with potential future risk levels.

115
 At a regional scale, the cost of traditional monitoring bore infrastructure is likely to be
an ongoing challenge. This is particularly the case in fractured rock systems subject
to high hydrogeological variability. Emerging sensing technologies able to gather
data over large areas and at depth may provide a step-change to the field, subject to
cost and commercial availability. Whether at the local or regional scale the choice of
monitoring will be informed by the level of risk and the cost-effectiveness of the
monitoring. Local research studies may prove a useful adjunct.
 The assessment process for proposed extractions takes into account the risks of
local impacts through a risk assessment process, requirements for some applicants
for proposed medium and high risk extractions to submit a hydrogeological report to
support their application, and criteria for acceptable levels of local impacts.
 Local scale monitoring during extraction operations can assist with better
understanding of local hydrology and extractive impacts and consequences. This
may include piezometric monitoring of the pathway between the point of extraction
and locations where there is perceived risk. The cost of this monitoring is likely to be
a challenge and its requirement should be justified by the risks as identified by an
expert following analysis of pump test data.
 Local scale monitoring during extraction operations could potentially support adaptive
management, for example, through additional reporting and cease-to-pump rules
related to observed groundwater pressures.
 The Review considered a number of past hydrogeological reports submitted to
support proposed extractions by the bottled water industry in the Northern Rivers
area as components of development applications. The hydrogeological reports, in
most cases, assess hydrogeological impacts of proposed developments on surface
flows, GDEs and surrounding groundwater bores. The focus and style of the reports
varies depending on the purpose and the identified or perceived risks.
 Both industry and decision makers would substantially benefit from greater clarity,
specificity and standardisation of requirements for hydrological reports. Current
technology is available to enable standardised templates and reports to be managed
electronically.
 Robust local assessment of potential connectivity between aquifer and overlying
shallow groundwater and surface water should form part of pump tests and feature in
hydrogeological reports. This is important, as observed in Alstonville, where deeper
aquifers are not necessarily confined and may have connections to surface systems
or shallower aquifers. It is important to increase understanding of how confined the
aquifer is, as assessment criteria of allowable drawdown differs between confined
and unconfined systems. In addition, field verification is an important part of the
process.
 The Review received anecdotal information suggesting bottled water extractors were
generally extracting water at an approximately evenly spaced production rate year-
round compared with other commercial users who extract on a more periodic basis.
The Review was not able to verify these observations. Further, all groundwater users
are subject to future changing environmental conditions, which may influence their
future patterns of use. The implementation of the NSW Non-Urban Water Metering
Policy will provide information about use patterns in the bottled water industry and
enable analyses of interactions and impacts.
 The Review received consistent reports from the community and sometimes
neighbours of bottled water extractors about observed changes including
environmental effects of drying watercourses and loss of water from previously
productive bores. The Review has not identified scientific studies or other evidence

116
 establishing a causal link between these observed effects and extraction specifically
undertaken by the bottled water industry. Going forward, data from extraction bores,
together with monitoring bores (piezometers), local studies and other sources of
information should help improve knowledge of impacts from a range of sources.

6.1.7 Data
 Lack of extraction data is an impediment to establishing appropriate extraction limits
for individual bores, measuring impacts, and at a regional scale, development of
WSP and making determinations of available water. A state-wide metering policy for
qualifying groundwater works with bore diameters of 200mm and above will take
effect in the Northern Rivers region from 2023. Four of the bottled water extractors in
the region are currently required by the regulator to have meters installed.
 The accessibility of any data is central and manual collection can be an impediment
in this regard. Advances in technology to provide robust and tamper-proof
telemetering options that are commercially cost competitive would have a significant
impact.
 Making water extraction and monitoring data available in standardised formats
through open databases would benefit decision-makers, researchers and the general
public to understand better the activities and impacts, including cumulative impacts at
local and regional scale. Approvals by relevant state and local government
authorities could include requirements that all hydrogeological data are published.
There are state managed environmental databases (e.g. SEED) that could be
utilised.

6.1.8 Decision-making
 As with any environmental, engineering, resource activity the proponents and
decision makers and regulators operate in a realm of imperfect information. This
leads to levels of uncertainty around data and information; however, uncertainty need
not prevent decisions being made.
 There are a number of approaches and tools employed to reduce uncertainty with
regard to the assumptions, hydrological domain, impacts, and consequences of
water extraction. These include taking conservative estimates, using multiple lines of
analysis, being judicious in decisions around the type and location of monitoring,
employing adaptive management approaches.
 There is a lack of clarity around water planning, management and decision-making
roles and processes at state and local government level and between relevant
authorities.
 State government agencies and local government should work to clarify roles and
responsibilities to streamline assessment and approval processes, to avoid
duplication of effort, and to address any gaps in the assessment and approvals
process. The first step for this would be by February 2020 relevant officers from
Water NSW, DPIE Water, NRAR and Tweed Council convene a workshop for
Northern Rivers region bottled water to discuss and develop an approach between
them. to:
o Develop a shared understanding of the data, modelling and information
needed by each agency in decision making
o Documenting a set of standards for data capture, sharing, storage between
agencies

117
 o Maximise the utility and outcomes for the group of agencies when new
monitoring is being rolled out
o Discuss gaps and overlaps in conditions between regulatory and compliance
instruments of different agencies. Access to government and industry water
data through a common open platform housing standardised, well-curated
and long-term data sets that can be expanded would assist assessment and
decision-making of applications.
 If Local government is to undertake hydrogeological assessment as part of the
development application process, then it needs access to relevant expertise to
interpret modelling and technical reports to inform its decision-making, including
requirements for development applications.
 Regional Water Strategies will be developed over the coming months for the 12
catchment regions across the state and will assist to manage the regions’ water
resources. The Greater Hunter Regional Water Strategy is already in place. These
will improve water security within each region and influence decisions about
infrastructure, water reuse, water sharing including during droughts, protect the
regions’ environmental assets as well as addressing community and industry needs.

6.1.9 Truck movements and road impacts

 There are technologies available that can provide accurate, consistent and real-time
data on truck movements, which could be included as a condition of the development
consent.
 Responsibility for governing truck safety, movements and size spans Federal, State
and Local Government authorities. Each of the responsible bodies has measures to
regulate and monitor heavy vehicles through existing legalisation, approval of
applications and technologies.
 Technologies and strategies are available to measure traffic volumes and impacts.
Local government can levy heavy vehicle road users to contribute to the cost of road
maintenance and repair.

6.1.10 Plastics
 The presence and management of plastics is international in scope and management
of the impacts and solutions will be influenced significantly by factors and
developments beyond those extracting water for bottling purposes in the Northern
Rivers region.
 The NSW Government is developing a 20-year waste strategy and plastics plan in
the context of broader Federal Government and inter-jurisdictional commitments to
address waste and transition from linear to circular economies.
 There is a NSW Government container deposit scheme, which has resulted in a one-
third reduction across the state of eligible containers, including bottles entering the
litter stream.
 Research and development efforts to replace, repurpose and recycle plastics is a
fast-moving and evolving space that is predicted to show significant growth within the
next five to ten years.

118
6.2 RECOMMENDATIONS
1. Further work is undertaken to incorporate climate change into the development of
recharge estimates for the Water Sharing Plan.
2. Consideration should be given to incorporate geological heterogeneity and soil and
vegetation types into recharge estimates where practicable. This may be dependent
in part on technological advances, including remote sensing, to characterise
systems.
3. Improved monitoring of piezometric water levels is needed in locations with a
perceived risk and/or lack of knowledge of groundwater responses and flow
directions. This could provide baseline data, conceptual hydrogeological models and
recharge estimates commensurate with potential future risk levels. Additional
investments in monitoring should balance the value of expected improvements in
data availability and data quality against the resources required.
4. Robust local hydrogeological assessments of aquifer connectivity with overlying
shallow groundwater and surface water should be investigated via well-designed
pump tests. This information should feature in hydrogeological reports.
5. Work should continue towards developing practical and comprehensive guidance on
the contents of hydrogeology reports to be submitted by proponents, including
specificity and standardisation of information provided and reporting requirements.
Ideally, these would be able to be lodged electronically and made publically
available.
6. State government agencies and local government should work to clarify roles and
responsibilities to streamline assessment and approval processes, to avoid
duplication of effort, and to address any gaps in the assessment and approvals
process. The first step for this would be by February 2020, relevant officers from
Water NSW, DPIE Water, NRAR and Tweed Council convene a workshop for
Northern Rivers region bottled water to discuss and develop an approach between
them.
7. Water extraction and monitoring data should be made available in standardised
formats through open and accessible portals. State managed databases and portals
(e.g. SEED) should be utilised where relevant.

119
REFERENCES

ABC. (2019). Refreshing Our Economy. Retrieved 31 January 2019, from

http://www.australianbeverages.org/for-consumers/refreshing-economy/
ABS, Australian Bureau of Statistics. (2018). 3222.0 - Population Projections, Australia, 2017
(base) - 2066. Retrieved 22 October 2019, from
https://www.abs.gov.au/AUSSTATS/abs@.nsf/mf/3222.0
ABWI, Australasian Bottled Water Institute. (2018). Model Code for Production of Bottled
Water. from http://www.australianbeverages.org/wp-content/uploads/2019/01/ABWI-
Code-Revised-2018.pdf
Acworth, I. (2019). Investigating Groundwater. International Contributions to Hydrogeology.
1st Edition. CRC Press.
Anderson, V.G. (1945). Some effects of atmospheric evaporation and transpiration on the
composition of natural water in Australia (continued). 4. Underground waters in
riverless areas. Journal of Proceedings of the Australian Chemical Institute, 12, 83-
98.
Antea Group. (2015). Water and Energy Use Benchmarking Study: Executive Summary.
Prepared for International Bottled Water Association. Retrieved from
https://www.bottledwater.org/public/IBWA%20Water%20and%20Energy%20Use%20
Benchmarking%20Report%20-
%20Exec%20Summary%20%28Revised%20May%2018%202016%29.pdf
APCO, Australian Packaging Covenant Organisation. (2019a). 2019 Annual Report & Action
Plan. Asahi Beverages. Retrieved 15 October 2019, from
https://www.packagingcovenant.org.au/documents/item/2653
APCO, Australian Packaging Covenant Organisation. (2019b). 2019 Annual Report & Action
Plan. Coca-Cola Amatil Limited. Retrieved 15 October 2019, from
https://www.packagingcovenant.org.au/documents/item/2897
APCO, Australian Packaging Covenant Organisation. (2019c). APCO Priority Projects 2019.
Version 3.
APCO, Australian Packaging Covenant Organisation. (2019d). Australia's 2025 National
Packaging Targets. Retrieved 15 October 2019, from
https://www.packagingcovenant.org.au/who-we-are/australias-2025-national-
packaging-targets
APCO, Australian Packaging Covenant Organisation. (2019e). Biodegradable Compostable
Packaging Working Group 2018.
APCO, Australian Packaging Covenant Organisation. (2019f). Delivering the 2025 Targets.
Retrieved 15 October 2019, from https://www.packagingcovenant.org.au/who-we-
are/delivering-the-2025-targets
Arnold, J.G., & Allen, P.M. (1999). Automated Methods for Estimating Baseflow and Ground
Water Recharge from Streamflow Records. Journal of the American Water
Resources Association, 35(2), 411-424. doi: https://doi.org/10.1111/j.1752-
1688.1999.tb03599.x
Arnold, J.G., Allen, P.M., Muttiah, R., & Bernhardt, G. (1995). Automated Base Flow
Separation Recession Analysis Techniques. Ground Water, 33(6), 1010-1018. doi:
10.1111/j.1745-6584.1995.tb00046.x
Asahi Beverages. (2016). Australian Packaging Covenant Action Plan 2016 - 2020.
Retrieved from
https://www.asahi.com.au/images/responsibility/Asahi_Beverages_APC_Action_Plan
_2016-2020.pdf
Asahi Group Holdings. (2019). Factbook 2019 (Updated on August 2, 2019). Retrieved from
https://www.asahigroup-holdings.com/en/ir/pdf/2019_2q_factbook.pdf

120
Asher, M., Croke, BFW., Jakeman, AJ. , & Peeters, LJM. (2015). A review of surrogate
models and their application to groundwater modeling. Water Resources Research,
51(8), 5957-5973.
Australasian Bioplastics Association. (2019). Frequently Asked Questions. Retrieved 15
October 2019, from https://www.bioplastics.org.au/resources/faq/#toggle-id-23
Beverage Marketing Corporation. (2018). Bottled Water 2017. Staying Strong. U.S. and
International Developments & Statistics. Retrieved from
https://www.bottledwater.org/public/BMC2017_BWR_StatsArticle.pdf
Bilge, H. . (2003). Alstonville Plateau Groundwater Model. Model Development and
Calibration. Prepared for NSW Department of Sustainable Natural Resources
Blue Environment. (2018). National Waste Report 2018. Prepared for Commonwealth
Department of the Environment and Energy
BOM, Bureau of Meteorology. (2015). Gridded Average Rainfall Metadata. Retrieved from
http://www.bom.gov.au/climate/map/rainfall/average-rainfall-metadata.pdf
BOM, Bureau of Meteorology. (2019). Australian Landscape Water Balance. Retrieved 24
October 2019, from http://www.bom.gov.au/water/landscape
Bureau of Transport and Communications Economics. (1997). Taxes and Charges in
Australian Transport: A Transmodal Overview; Working Paper 34. Retrieved from
https://www.bitre.gov.au/publications/1997/files/wp_034.pdf
Canty, A., & Ripley, B. (2019). boot: Bootstrap R (S-Plus) Functions. R package version, 1,
3-23.
Carton & Co Water. (2019). FAQs. Retrieved 15 October 2019, from
https://cartonandco.com.au/pages/faqs
CIEL, Centre for International Environmental Law. (2017). Fueling Plastics. Fossils, Plastics
& Petrochemical Feedstocks. Retrieved 15 October 2019, from
https://www.ciel.org/wp-content/uploads/2017/09/Fueling-Plastics-Fossils-Plastics-
Petrochemical-Feedstocks.pdf
COAG, Council of Australian Governments. (2019). Meeting Outcomes. Retrieved 15
October 2019, from https://www.coag.gov.au/meeting-outcomes/coag-meeting-
communique-09-august-
2019.%20Leaders%20committed%20to%20drawing%20on%20the%20best%20scie
nce,%20research%20and%20commercial%20experience%20to%20inform%20the%
20strategy.
Coca-Cola Amatil. (2016). 2015 Full year result. Retrieved 30 September 2019, from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/Results/2015/FY15-Results-
v10-FINAL.ashx
Coca-Cola Amatil. (2017). 2016 Full Year Result. Retrieved 30 September 2019, from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/Results/2016/170222-Full-
Year-Results-Presentation.ashx
Coca-Cola Amatil. (2018a). 2017 Full Year Result. Retrieved 30 September 2019, from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/ASX-
Announcements/2018/2017-Full-Year-Presentation.ashx
Coca-Cola Amatil. (2018b). Sustainability Report 2018. Retrieved from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/Annual-Reports/2019/2018-
Sustainability-Report.ashx?la=en
Coca-Cola Amatil. (2019a). 2018 Full Year Results. Retrieved 30 September 2019, from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/Presentations/2019/Coca-
Cola-Amatil-2018-Full-Year-Results-Presentation.ashx
Coca-Cola Amatil. (2019b). Annual Report 2018. Retrieved 30 September 2019, from
https://www.ccamatil.com/-/media/Cca/Corporate/Files/Annual-Reports/2019/2018-
Annual-Report.ashx
Crosbie, R. , Jolly, I., Leaney, F., Petheram, C., & Wohling, D. (2010). Review of Australian
Groundwater Recharge Studies. CSIRO: Water for a Healthy Country National
Research Flagship. Retrieved from
https://publications.csiro.au/rpr/download?pid=csiro:EP101470&dsid=DS2

121
Danone. (2000). Groupe Danone. Rapport Annuel 1999. Retrieved from
http://media.corporate-
ir.net/media_files/irol/13/131801/reports/annual/Rapport_Annuel_1999.pdf
Danone. (2019). Evian. Retrieved 30 September 2019, from
https://www.danone.com/brands/waters/evian.html
DECCW Water, NSW Department of Environment, Climate Change and Water - Office of
Water,. (2011). Environmental flow response and socio-economic monitoring. North
Coast - progress report 2009. Report No. NOW 10_392
DEE, Commonwealth Department of the Environment and Energy. (2018). Analysis of
Australia's municipal recycling infrastructure capacity.
DNRME, Queensland Department of Natural Resources, Mines and Energy. (2016).
Hydrogeological Conceptualisation Report for the Surat Cumulative Management
Area. Prepared by the Office of Groundwater Impact Assessment, Brisbane
DOI, NSW Department of Industry. (2018a). Assessing groundwater applications. Retrieved
8 October 2019, from
https://www.industry.nsw.gov.au/__data/assets/pdf_file/0008/175931/Assessing-
groundwater-applications-fact-sheet.pdf
DOI, NSW Department of Industry. (2018b). NSW Non-Urban Water Metering Policy.
Retrieved from
https://www.industry.nsw.gov.au/__data/assets/pdf_file/0017/205442/NSW-non-
urban-water-metering-policy.pdf
DOI, NSW Department of Industry. (2018c). Water management Fact sheet.
DOI, NSW Department of Industry. (2019). Water Sharing Plans status. Retrieved 31
January 2019, from https://www.industry.nsw.gov.au/water/plans-programs/water-
sharing-plans/status
Dowdy, A., Abbs, D., Bhend, J., Chiew, F., Church, J., Ekstrom, M., Kirono, D., Lenton, A.,
Lucas, C. , McInnes, K., Moise, A., et al. (2015). East Coast Cluster Report. Climate
Change in Australia Projection for Australia's Natural Resource Management
Regions: Cluster Reports. Eds. Ekstrom, M. et al., CSIRO and Bureau of
Meteorology, Australia. Retrieved from
https://www.climatechangeinaustralia.gov.au/media/ccia/2.1.6/cms_page_media/172/
EAST_COAST_CLUSTER_REPORT_1.pdf
DPE, NSW Department of Planning & Environment. (2019). Gateway Determination.
DPI Water, NSW Department of Primary Industries - Office of Water. (2015). Macro water
sharing plans - the approach for groundwater. A report to assist community
consultation.
DPI Water, NSW Department of Primary Industries - Office of Water. (2016a). Report card
for the Alstonville Basalt Plateau Groundwater Source.
DPI Water, NSW Department of Primary Industries - Office of Water. (2016b). Report card
for the Clarence Moreton Basin Groundwater Source.
DPI Water, NSW Department of Primary Industries - Office of Water. (2016c). Report card
for the New England Fold Belt Coast Groundwater Source.
DPI Water, NSW Department of Primary Industries - Office of Water. (2016d). Report card
for the North Coast Volcanics Groundwater Source.
DPI Water, NSW Department of Primary Industries - Office of Water. (2016e). Rules
Summary Sheet. Water Sharing Plan for the North Coast Fractured and Porous Rock
Groundwater Sources
DPI Water, NSW Department of Primary Industries - Office of Water. (2016f). Water Sharing
Plan for the North Coast Fractured and Porous Rock Groundwater Sources.
Background document.
Eamus, D., Hatton, T., Cook, P., & Colvin, C. (2006). Ecohydrology: vegetation function,
water and resource management: CSIRO Publishing.
Eamus, D., Zolfaghar, S., Villalobos-Vega, R., Cleverly, J., & Huete, A. (2015). Groundwater-
dependent ecosystems: Recent insights from satellite and field-based studies.

122
 Hydrology and Earth System Sciences, 19(10), 4229-4256. doi: 10.5194/hess-19-
4229-2015
Eco Logical. (2016). Groundwater Assessment 477 Urliup Road Bilambil. Prepared by Eco
Logical Australia for the Karlos Family Trust
Eco Logical. (2018a). Hydrogeological Assessment 350 Rowlands Creek Road. Prepared by
Eco Logical Australia for Jim Glazebrook & Associates Pty Ltd
Eco Logical. (2018b). Hydrogeological Assessment. Kyogle Road Water Bottling Facility.
Prepared by Eco Logical Australia for Mount Warning Spring Water Company
Australia Pty Ltd
EFBW, European Federation of Bottled Waters. (2019). Key statistics. Retrieved 30
September 2019, from https://www.efbw.org/index.php?id=90
Ellen MacArthur Foundation. (2014). The New Plastics Economy. Rethinking the Future of
Plastics. Prepared under Project MainStream by the World Economic Forum and the
Ellen MacArthur Foundation
EMM. (2015). Coastal Porous Rock Rainfall Recharge Study. Prepared for NSW Department
of Primary Industries - Office of Water
EMM. (2018). Water rights - Ballina Tweed area. Maps and Graphics provided to the
Department of Primary Industries - Office of Water
Envisage Works. (2018). 2016-17 Australian Plastics Recycling Survey. National report.
Prepared for Commonwealth Department of the Environment and Energy
Envisage Works. (2019). 2017-18 Australian Plastis Recycling Survey. National report.
Prepared for Commonwealth Department of the Environment and Energy
EPA, NSW Environment Protection Authority. (2013). Managing vehicle noise. Report No.
EPA 2013/0197. Retrieved from https://www.epa.nsw.gov.au/-/media/epa/corporate-
site/resources/noise/130197vehicle.pdf
EPA, NSW Environment Protection Authority. (2014). NSW Waste Avoidance and Resource
Recovery Strategy 2014-21. Report No. EPA 2014/0876
EPA, NSW Environment Protection Authority. (2018). Too Good To Waste. Discussion paper
on a circular economy approach for NSW. Report No. EPA 2018P1061
European Commission. (2015). The Impact of the Use of "Oxo-degradable" Plastic on the
Environment. Prepared for the European Commission DG Environment.
European Commission. (2018). Report from the Commission to the European Parliament
and the Council on the impact of the use of oxo-degradable plastic, including oxo-
degradable plastic carrier bags, on the environment. Report No. COM (2018) 35 final
Evian. (2017). Evian, A Unique Water Source Steeped in History. Retrieved 30 September
2019, from
http://www.evianfactory2017.com/static/documents/Fiches%20individuelles%20EN/1.
%20EVIAN%C2%AE,%20A%20UNIQUE%20WATER%20SOURCE%20STEEPED%
20IN%20HISTORY.pdf
Fletcher, S., Lickley, M., & Strzepak, K. . (2019). Learning about climate change uncertainty
enables flexible water infrastructure planning. Nature Communications, 10(1), 1782.
Freight on Rail. (2019). Lorries cause more damage to roads than cars from
http://www.freightonrail.org.uk/HotTopicsLorriesCauseMoreDamageToRoadsThanCa
rs.htm
Green, R. (2006). Alstonville Groundwater Investigations. Status Report. Department of
Natural Resources
Hastie, T.J., & Tibshirani, R.J. (1990). Generalized Additive Models. Chapman & Hall/CRC.
Hodgett, R.E., & Siraj, S. (2019). SURE: A method for decision-making under uncertainty.
Expert Systems with Applications, 115, 684-694. doi:
https://doi.org/10.1016/j.eswa.2018.08.048
IBISWorld. (2019). Sparkling future: Rising disposable incomes are projected to boost
industry revenue. Bottled Water Manufacturing in Australian Industry Report
prepared for NSW Department of Industry
Institute of Medicine. (2013). Environmental Decisions in the Face of Uncertainty.

123
IPCC, Intergovernmental Panel on Climate Change. (2013). Summary for Policymakers. In:
Climate Change 2013: The Physical Science Basis. Contribution of Working Group I
to the Fifth Assessment Report of the Intergovernmnetal Panel on Climate Change.
Cambridge University Press, Cambridge, United Kingdowm and New York, NY, USA.
Retrieved from
https://www.ipcc.ch/site/assets/uploads/2018/02/WG1AR5_SPM_FINAL.pdf
Kelly, S., Lewis, H., Atherton, A., & Downes, J. (2017). Packaging Sustainability Framework
for APCO Members. Final Report. Institute for Sustainable Futures. University of
Technology Sydney
Kochenderfer, M.J., Amato, C., Chowdhary, G., How, J.P., Reynolds, H.J.D., Thornton, J.R.,
Torres-Carrasquillo, P.A., Ure, N.K., & Vian, J. (2015). Decision Making Under
Uncertainty. Theory and Application. MIT Press
Kuginis, L., Dabovic, J., Byrne, G., Raine, A., & Hemakumara, H. (2016). Methods for the
identification of high probability groundwater dependent vegetation ecosystems.
Published by the Department of Primary Industries, a Division of NSW Department of
Industry, Skills and Regional Development. Retrieved from
https://www.industry.nsw.gov.au/__data/assets/pdf_file/0010/151894/High-
Probability-GDE-method-report.pdf
Kunsch, H.R. (1989). The Jackknife And The Bootstrap For General Stationary
Observations. The Annals of Statistics, 17(3), 1217-1241. doi:
10.1214/aos/1176347265
Linfox. (2019). Transport management systems. Retrieved 24 October 2019, from
https://www.linfox.com/supply-chain-solutions/logistics-it/transport-management-
systems/
Locock, K.E.S., Deane, J., Kosior, E., Prabaharan, H., Skidmore, M., & Hutt, O.E. (2017).
The Recyled Plastics Market: Global Analysis and Trends. CSIRO, Australia
Madden, B., & Florin, N. (2019). APCO Packaging Material Flow Analysis 2018.
Characterising the material flows through the Australian waste packaging system.
Prepared by Institute for Sustainable Futures, University of Technology Sydney
Mau, D.P., & Winter, T.C. (1997). Estimating Ground-Water Recharge from Streamflow
Hydrographs for a Small Mountain Watershed in a Temperate Humid Climate, New
Hampshire, USA. Ground Water, 35(2), 291-304. doi: https://doi.org/10.1111/j.1745-
6584.1997.tb00086.x
Meldrum-Hanna, C., Davies, A., & Richards, D. (2017). Recycling companies stockpiling
thousands of tonnes of glass as cheap imports leave market in crisis, ABC News.
Retrieved from https://www.abc.net.au/news/2017-08-07/recycling-companies-
forced-to-stockpile-glass-industry-crisis/8778088
Moore, T. (2014). Future Water Strategy. Integrated Water Planning Process. Prepared by
MWH for Rous Water
Nathan, R.J., & McMahon, T.A. (1990). Evaluation of automated techniques for base flow
and recession analyses. Water Resources Research, 26(7), 1645-1473. doi:
https://doi.org/10.1029/WR026i007p01465
National Uniform Drillers Licensing Committee. (2012). Minimum Construction Requirements
for Water Bores in Australia, Third Edition
NHMRC, National Health and Medical Research Council, & NRMMC, National Resource
Management Ministerial Council. (2011). National Water Quality Management
Strategy. Australian Drinking Water Guidelines 6. Version 3.4. Updated October
2017. Commonwealth of Australia, Canberra
NRC, Natural Resources Commission. (2013). Review of 2004 water sharing plans
NRC, Natural Resources Commission (2005). Standard for Quality Natural Resource
Management.
NSW Innovation and Productivity Council. (2019). Innovatoin in the NSW environmental
goods and services sector.
NSW Office of Water. (2011). NSW policy for managing access to buried groundwater
sources. Report No. NOW 11_133

124
NTC, National Transport Commission. (2019). About the HVNL. Retrieved 24 October 2019,
from https://www.ntc.gov.au/laws-and-regulations/heavy-vehicle-national-law
Nu-Pure. (2019). Quality & Manufacturing. Retrieved 15 October 2019, from http://nu-
pure.com.au/quality-manufacturing/
OEH, NSW Office of Environment & Heritage. (2014). North Coast. Climate change
snapshot. Report No. OEH 2014/0824
OEH, Office of Environment and Heritage (2018). Adaptive Management Position Statement
Retrieved from https://www.environment.nsw.gov.au/research/adaptive-
management.htm
Oregon State University. (2019). Groundwater and Wells
Retrieved 18 October 2019, from http://wellwater.oregonstate.edu/groundwater-and-wells
PACIA, Plastics and Chemicals Industries Association Incorporated. (2007). Using
Degradable Plastics in Australia. A Product Stewardship Guide and Commitment.
Prepared in partnership with the Commonwealth Department of the Environment and
Water Resources
Parliament of NSW. (2012). 3120 - Noise Camera on Mount Ousely Road. Retrieved 24
October 2019, from https://www.parliament.nsw.gov.au/la/papers/Pages/qanda-
tracking-details.aspx?pk=207525
Parsons Brinckerhoff. (2011). Future Water Strategy Groundwater Options - position paper.
Prepared for Rous Water
Peterson, T.J. , & Western, A.W. (2014). Nonlinear time-series modeling of unconfined
groundwater head. Water Resources Research, 50(10), 8330-8355. doi:
doi:10.1002/2013WR014800
Polasky, S., Carpenter, S.R., Folke, C., & Keeler, B. (2011). Decision-making under great
uncertainty: environmental management in an era of global change. Trends in
Ecology and Evolution, 26, 398-404.
Preston, B.J. . (2017). The Judicial Development of the Precautionary Principle.
PricewaterhouseCoopers. (2008). Cost Schedule for Food Labelling Changes. Final Report.
Food Standards Australia New Zealand. Retrieved from
https://www.foodstandards.gov.au/publications/Documents/Final%20report-
%20FSANZ%20-%207%20March%202008%20(2).pdf
Queensland Government. (2019). SILO - Australian Climate Data from 1889 to Yesterday.
Retrieved 24 October 2019, from https://www.longpaddock.qld.gov.au/silo/
Raiber, M., Cui, T., Pagendam, D. , Rassam, D., Gilfedder, M., Crosbie, R., Marvanek, S., &
Hartcher, M. (2016). Observations analysis, statistical analysis and interpolation for
the Clarence-Moreton bioregion. Product 2.1-2.2 from the Clarence-Moreton
Bioregional Assessment. Prepard for the Departmnet of the Environment and
Energy, Bureau of Meteorology, CSIRO and Geoscience Australia
RBA, Reserve Bank of Australia. (2019). Statement on Monetary Policy. August 2019.
Retrieved 22 October 2019, from https://www.rba.gov.au/publications/smp/2019/aug/
Richardson, S., Irvine, E., Froend, R., Boon, P., Barber, S., & Bonneville, B. (2011).
Australian groundwater-dependent ecosystems toolbox part 2: assessment tools.
Waterlines report, National Water Commission, Canberra. Retrieved from
http://www.bom.gov.au/water/groundwater/gde/GDEToolbox_PartTwo_Assessment-
Tools.pdf
RMS, NSW Roads and Maritime Services. (2019). General access heavy vehicles.
Retrieved 24 October 2019, from https://www.rms.nsw.gov.au/business-
industry/heavy-vehicles/road-access/general-access-vehicles.html
Rogers, K. (2011). Scientific Modeling. from https://www.britannica.com/science/scientific-
modeling
Rojas, R., Kahunde, S., Peeters, L., Batelaan, O., Feyen, L., Dassargues, A. . (2010).
Application of a multimodel approach to account for conceptual model and scenario
uncertainties in groundwater modelling. Journal of Hydrology, 394(3-4), 416-435.
Rous Water. (2014). Future Water Strategy. Water Security for our future.

125
San Pellegrino. (2019). S.Pellegrino Celebrates Its 120th Anniversary. Retrieved 30
September 2019, from https://www.sanpellegrino.com/au/en/120-years
Scanlon, B.R. , Healy, R.W. , & Cook, P.G. . (2002). Choosing appropriate techniques for
quantifying groundwater recharge Hydrogeology Journal, 10(1), 18-39. doi:
https://doi.org/10.1007/s10040-001-0176-2
Scientific Committee, NSW Threatened Species Scientific Committee. (2019). Lowland
rainforest on floodplain in the NSW North Coast Bioregion - endangered ecological
community listing. Retrieved 28 October 2019, from
https://www.environment.nsw.gov.au/topics/animals-and-plants/threatened-
species/nsw-threatened-species-scientific-committee/determinations/final-
determinations/1996-1999/lowland-rainforest-on-floodplain-nsw-north-coast-
bioregion-endangered-ecological-community-listing
Selke, S., Auras, R., Nguyen, T.A., Aguirre, E.C., Cheruvathur, R., & Liu, Y. (2015).
Evaluation of Biodegradation-Promoting Additives for Plastics. Environmental
Science & Technology, 49, 3769-3777. doi: DOI: 10.1021/es504258u
Shi, X., Vaze, J., & Crosbie, R. (2015). Evaluation of AWRA recharge: Comparison to field
recharge estimates across Australia. Technical Report. CSIRO. Retrieved from
https://publications.csiro.au/rpr/download?pid=csiro:EP157510&dsid=DS2
Spiegelhalter, D. (2017). Risk and uncertainty communication. Annual Review of Statistics
and Its Application, 4, 31-60.
Spiegelhalter, D., Pearson, M., & Short, I. (2011). Visualizing Uncertainty About the Future.
Science, 333(6048), 1393-1400. doi: 10.1126/science.1191181
Statista. (2019). Bottled Water. Retrieved 30 September 2019, from
https://www.statista.com/outlook/20010000/107/bottled-water/australia
Tallaksen, L.M. (1995). A review of baseflow recession analysis. Journal of Hydrology,
165(1-4), 349-370. doi: https://doi.org/10.1016/0022-1694(94)02540-R
Transport and Infrastructure Council. (2017). Heavy Vehicle Charges Model Law. Published
by the Parliamentary Counsel's Committee. Retrieved from
https://www.pcc.gov.au/uniform/Heavy-Vehicle-Charges-Model-Law.pdf
TSC, Tweed Shire Council. (2016). CP04. Section 7.11 Plan (formerly Section 94). No. 4 -
Tweed Road Contribution Plan. Version 6.3.8. Indexed July 2019. Retrieved from
https://www.tweed.nsw.gov.au/Controls/Planning/Documents/PlanningDocs/CP04.pd
f
TSC, Tweed Shire Council. (2019b). Planning Proposal to Remove Enabling Clause 7.15 for
Water Bottling Facilities in the RU2 Zone. Version 2 Exhibition. Version 2 Exhibition.
July 2019. Report No. TSC File PP18/0004
VOSS. (2019). About VOSS. Retrieved 30 September 2019, from
https://vosswater.com/about/
Waste Management Review. (2019). Trevor Evans MP opens APCO Collective Action
Group. Retrieved 15 October 2019, from
http://wastemanagementreview.com.au/cag-trevor-evans/
WaterNSW. (2017). Coastal groundwater. Test pumping groundwater assessment
guidelines for bore licence applications. Retrieved 8 October 2019, from
https://www.waternsw.com.au/__data/assets/pdf_file/0009/128826/3.-
avail_ground_coastal_test_pumping_bore_licence_guide.pdf
WaterNSW. (2019). NSW Water Register. Retrieved 28 October 2019, from
https://waterregister.waternsw.com.au/water-register-frame
West, C., Sabburg, L., Arnold, S., Reading, L., McIntyre, N., Larsen, J., Bulovic, N.,
Bloomfield, G., & Finch, W. (2018). Recharge Estimation in the Surat Basin. Final
Report. The University of Queensland, Australia
Willems, P. (2009). A time series tool to support the mutli-criteria performance evaluation of
rainfall-runoff models. Environmental Modelling & Software, 24(3), 311-321. doi:
https://doi.org/10.1016/j.envsoft.2008.09.005

126
Wood, W.W. (1999). Use and Misuse of the Chloride-Mass Balance Method in Estimating
Ground Water Recharge. Ground Water, 37(1), 2-3. doi: 10.1111/j.1745-
6584.1999.tb00949.x

127
APPENDIX 1: TERMS OF REFERENCE (UPDATED FEB 2019)

The Chief Scientist & Engineer is requested to conduct an independent review and provide
expert advice on the impacts on groundwater quantity arising from extraction by the bottled
water industry in the Northern Rivers region of NSW.

1. In undertaking the review, the Chief Scientist & Engineer will:

a. review existing data and information on the bottled water industry’s
entitlements and extractions in the context of:
i. total water access rights (basic landholder rights and access licences),
and
ii. extraction limits established in statutory water sharing plans
b. provide advice on the sustainability of the extraction limits in the relevant
water sharing plans for groundwater sources in the Northern Rivers of NSW
c. provide advice on whether the current or proposed groundwater monitoring
bores on the Northern Rivers are sufficient.
2. Provide advice on potential impacts:
a. on groundwater resources, having regard to the sustainable take of the
resource and the scale of the current bottled water industry and proposed or
potential expansion of the industry
b. of the groundwater take of the bottled water industry on surface water.
3. Provide advice on:
a. scientific and technical approaches to examining socio-economic factors and
impacts and possible solutions using locally relevant examples
b. localised environmental consequences related to extraction for bottled water.
4. As needed, the Chief Scientist & Engineer may:
a. seek advice from relevant Government agencies and other organisations
b. draw on additional sources of advice and expertise
c. commission or recommend studies.
5. The Chief Scientist & Engineer will:
a. consult with key local stakeholders
b. provide an initial report by 1 February 2019
c. provide a final report by Oct 2019.

128
APPENDIX 2: SITE VISITS, CONSULTATIONS, WORKSHOP AND
SUBMISSIONS

Table 19: Site visit consultations

Date Location Facility
9 December 2018 Urliup Karlos Family Trust
 Larry Karlos
10 December 2018 Uki Mount Warning Spring Water
 Shaun Martin
 Tessa Martin
20 January 2019 Kynnumboon Pristine Water Supply Pty Ltd
 Steve Bell
20 January 2019 Nobbys Creek Rosehill Estate 1890 Pty Ltd
 Gary Appleby
 Trevor Johnson
21 January 2019 Lynwood Prime Flowers Pty Ltd
 Geoffrey Bottomley
 Ian Cooke
21 January 2019 Alstonville Rous County Council Groundwater Bore
 Michael McKenzie

Table 20: Consultations

Date Location Present
9 December 2018 Murwillumbah Northern River Guardians
 Daniele Voinot
 Marian van Gestel
 Gwyn Hooper
 Scott Sledge
 Julie Beesley
Water Dragons
 Greg O’Donnell
 Michele Bevis
9 December 2018 Murwillumbah Dungay Action Group
 Betty Wood
 Lucy Campeanu
 Joy Baker
 Jack Griffis
 Dale Holt
9 December 2018 Murwillumbah Tweed Water Alliance
 Pat Miller
 Pamela Veness
 Denise White
 Trevor White
 Pamela Smith
9 December 2018 Murwillumbah Bilambil Urliup Action Group
 Anna Champ
 Jasmin Derrington
 Peter McIlveen
 Barbara Downes
 Louis Lambert
9 December 2018 Murwillumbah Bunjalung community members
 John Hunt
 Thomas Paulson
 Murray
10 December 2018 Murwillumbah Tweed Shire Council
 Michael Banks
 Robyn Eisermann

129
  Iain Lonsdale
 Denise Galle
 Danny Rose
 Ray Clark
 Tracey Stinson
Lismore City Council
 Leonie Walsh
Richmond Valley Council
 Mike Perkins
10 December 2018 Murwillumbah Tweed Shire Council
 Warren Polglase
 James Owen
 Katie Milne
 Reece Byrnes
 Pryce Allsop
 Troy Green
Hon Justine Elliot MP’s Office
 Jurgen Schanzenbacher
10 December 2018 Murwillumbah Rous County Council
 Phillip Rudd
 Michael McKenzie
10 December 2018 Uki  Graham Dietrich

13 December 2018 Sydney Australian Beverages Council

 Alby Taylor
 Shae Courtney
20 January 2019 Murwillumbah Combined Tweed Rural Industries Association
 Colin Brooks
20 January 2019 Murwillumbah Richmond Wilson Combined Water Users Association
 Chris Magner
 Catherine Richardson-Magner
20 January 2019 Ballina  Ceridwen Quick
 Clive Quick
21 January 2019 Rous Mill  Bryan Douglas

21 January 2019 Alstonville Nu-Pure Beverages

 Brendan Moroney
 Bruce Taylor
Black Mount Spring Water
 Tim Carey
21 January 2019 Alstonville Save Alstonville Aquifer
 Michael Hogan
 Troy Outerbridge
 David Huett
21 January 2019 Alstonville Ballina Shire Council
 Sharon Parry
 Eoin Johnson
 Ben Smith
 Phillip Meehan
 Sharon Cadwallader
 Matthew Wood
 David Wright
 Andrew Smith
 Simon Scott
 Georgia Lee
 Keith Williams (Chair of Rous County Council)
Byron Shire Council
 Jason Stanley
 Andrew Cameron
 Michael Bingham
 Bryan Green

130
1 August 2019 Level 48, MLC Centre Australian Beverages Council
19 Martin Place, Sydney  Shae Courtney
Blackmount Spring Water
 Tim Carey
 Scott Wallace
15 October 2019 Phone consultation Kevin Graham
17 October 2019 Phone consultation Lance Rawson

Table 21: Expert Workshop Participants – 6 September 2019

Participant Name Affiliation
Peter Cook Flinders University
Grant Hose Macquarie University
Lucy Reading Queensland University of Technology
Liliana Pagliero University of Queensland
Neil McIntyre University of Queensland
Louise Ryan University of Technology Sydney
Ian Acworth UNSW Sydney
Jason Evans UNSW Sydney
Lucy Marshall UNSW Sydney
Mahmood Sadat-Noori UNSW Sydney
Martin Andersen UNSW Sydney
Scott Sisson UNSW Sydney
William Glamore UNSW Sydney
Daniel Deere (independent facilitator) Water Futures
Luk Peeters CSIRO
Tomonori Hu NSW Smart Sensing Network
Jenny Johnson DPIE
Jon Stone DPIE
Danielle Doughty DPIE Water
Fabienne d’Hautefeuille DPIE Water
Lynn Tamsitt DPIE Water
Mark Simons DPIE Water
Richard Green DPIE Water

Table 22: Submissions

No. Organisation
SUB 001 Ballina Shire Council
SUB 002 Bryan Douglas
SUB 003 Michael Hogan, Save Alstonville Aquifer
SUB 004 David Huett, Save Alstonville Aquifer
SUB 005 Australian Beverages Council
SUB 006 NSW Irrigators Council
SUB 007 Save Alstonville Aquifer
SUB 008 Duncan Dey
SUB 009 CONFIDENTIAL
SUB 010 Australian Beverages Council
SUB 011 CONFIDENTIAL
SUB 012 CONFIDENTIAL
SUB 013 CONFIDENTIAL
SUB 014 CONFIDENTIAL

131
APPENDIX 3: EXCERPTS FROM INITIAL REPORT – EXTRACTION
LIMITS AND THE WATER SHARING PLAN

The following are extracts from the Initial Report providing background information about the
North Coast Fractured and Porous Rock Groundwater Sources Water Sharing Plan; and
how extraction limits are calculated and allocated. These sections are brought forward to
provide context for the analysis undertaken by the Review. Note that some reference in the
following text refer to Tables and Figures in the Initial Report.
Beginning of text from the Initial Report
The first Term of Reference entails an examination of extraction levels in the WSP and
seeks advice on extraction levels for the bottled water industry and future monitoring. In
working to address these, the consultation process proved to be a rich source of information,
data and ideas from the community and other stakeholders. The issues identified in those
fora (set out in Chapter 2 of the Initial Report) that relate to the volume of water and
allocation processes and issues are the focus of this section.
The quantity of water that is available for the bottled water industry in the Northern Rivers
region, as with other extractors for commercial purposes is established through the Water
Management Act 2000 and the WSP instruments.
Two fundamental principles for the WSP that are important for the framing of this Review
are:
1. there are established priorities of allocation with environmental and ecological first,
basic rights/stock and domestic second, and industrial and commercial extraction
last
2. within the category of industrial and commercial, there is no distinction made
between different ‘product categories’ or end uses – water involved in producing
food, drink, minerals, manufactured products and services are all considered on a
level playing field.
The following sections discuss in more detail the relevant WSP for groundwater in the
region. Further information about the regulatory framework is in Appendix 4 (of the Initial
Report) and the rules applying to the four groundwater sources relevant to the Review are at
Appendix 5 (of the Initial Report).

WATER SHARING PLAN FOR NORTH COAST FRACTURED

AND POROUS ROCK GROUNDWATER SOURCES
Under the Water Management Act 2000, WSPs have been developed for many groundwater
and surface water systems in NSW to control and limit usage of water resources, ensure that
Basic Landholder Rights (BLR) can be met and ensure that there is sufficient water reserved
as environmental water to support dependent ecosystems and maintain aquifer health. Table
23 sets out the WSPs in the Northern Rivers region.

132
Table 23: Northern Rivers region groundwater and surface water sharing plans
Water Sharing Plan Plan Status Supporting Documentation Cease Date
Brunswick Unregulated and Commenced  Brunswick water source rules July 2026
Alluvial July 2016  Background document
North Coast Coastal Sands Commenced  North Coast Coastal Sands Groundwater July 2026
Groundwater Sources July 2016 source rules
 Background document
North Coast Fractured and Commenced  North Coast Fractured and Porous Rock July 2026
Porous Rock Groundwater July 2016 Groundwater source rules
Sources  Background document
Richmond River Area Commenced  Richmond River are water source rules July 2021
Unregulated, Regulated and Dec 2010  Background document
Alluvial
Tweed River Area Commenced  Tweed River area water source rules July 2021
Unregulated and Alluvial Dec 2010  Background document
Alstonville Plateau Commenced  Replaced by Water Sharing Plan for the June 2016
Groundwater Sources 2004 - North Coast Fractured and Porous Rock
Repealed Groundwater Sources
Source: DOI (2019)

The Review is focusing on four specific groundwater sources in the Northern Rivers region
where there is current, proposed or potential historical groundwater extraction for bottled
water (Table 24). If the Review is made aware of additional groundwater sources in the
Northern Rivers region used by the industry, these will be added to the scope as the Review
progresses.

Table 24: Groundwater sources and descriptions

Groundwater Source Description
Alstonville Basalt A fractured rock aquifer system in which Tertiary basalt extends to a depth of up to
Plateau Groundwater 150 metres. Groundwater is contained in fractures in the basalt. The hydrology of
Source the area is complex and the degree of connectivity (both vertical and horizontal) is
not uniform. The groundwater in this source is used for town water supply and
irrigated agriculture. Discharge at the surface provides baseflow to surface waters
and is important to the environment
Clarence Moreton A porous rock aquifer system, overlain by the Mount Warning complex (comprised
Basin Groundwater of the North Coast Volcanics and the Alstonville Plateau groundwater sources). On
Source the eastern extent it is overlain by alluvial and coastal sand deposits. Groundwater
is both contained within the system, and moves through it, due to the primary
porosity of the rock as well as the fractures present due to the folding and faulting
of the rock formation. Low bore yields of 1L/s, rising to up to 10L/s in highly
fractured fault systems. All surface units are recharged by direct rainfall recharge
with subsequent vertical leakage. Generally used for stock and domestic purposes
with some sporadic irrigation/commercial supplies.
New England Fold Belt A fractured rock aquifer system, overlain by the Clarence Moreton Basin and North
Coast Groundwater Coast Volcanics groundwater sources. On the eastern extent it is overlain by
Source alluvial and coastal sand deposits. Groundwater is contained within, and moves
through, fractures in the rock due to the folding and faulting of the rock formations.
Low bore yields of 1L/s, rising to up to 10L/s in highly fractured fault systems.
Recharge is typically by direct rainfall infiltration and, combined with the degree of
mineral leaching that has occurred over time, has resulted in good quality water.
Generally used for small scale irrigation, stock and domestic purposes.
North Coast Volcanics A fractured rock aquifer system comprised of the Lamington Volcanics, associated
Groundwater Source with the Mount Warning Complex. It is situated on top of the New England Fold
Belt Coast and Clarence Morton Basin groundwater sources. Typically composed
of basalt and rhyolite, the groundwater is contained within, and moves through,
fractures formed as a result of the rock cooling as well as the vesicular structures
of basalt flows. Moderate bore yields of 5L/s, rising to up to 10L/s in highly
fractured fault systems. Recharge is typically by direct rainfall infiltration, resulting
in excellent quality water. Used for stock, domestic and irrigation water supplies.
Stream and spring flow is reliant on groundwater discharge during non-rainfall
periods. As a result, groundwater-dependant ecosystems are common with the
groundwater source.
Source: (DPI Water, 2016c, 2016d, 2016b, 2016a)

133
These sources are covered by the Water Sharing Plan for the North Coast Fractured and
Porous Rock Groundwater Sources (DPI Water, 2016f) (WSP) released in September 2016.
The WSP covers 13 groundwater sources from Gosford to Tweed Heads, 10 of which had
not previously been covered by a WSP. The total area covered by the WSP is approximately
76,000 km2 (DPI Water, 2016f). The groundwater sources covered by this plan are defined
either as porous or fractured rock aquifers, all with the following connectivity characteristics
according to the WSP:
 low-moderate connection between surface and groundwater
 low impact on in-stream values
 years to decades travel time between groundwater and surface water (DPI Water,
2016f)
Prior to the commencement of the WSP, of the four groundwater sources in scope, only the
Alstonville Plateau Groundwater Source was the subject of a WSP as noted in Table 24
above.
Prior to the finalisation of the WSP, report cards for each of the groundwater sources were
issued in February 2016. For the Alstonville Basalt Plateau Groundwater Source, which had
been subject to a prior plan, the proposed rule changes and technical specifications like
calculation of recharge were compared against the original WSP for Alstonville.
One significant change was that the previous division of the Alstonville Plateau into six
groundwater sources changed to the groundwater sources merged into one with two
management zones - the Alstonville-Tuckean and the Bangalow-Wyrallah management
zones. These were based on levels of extraction, particular intensity of extraction in the
Alstonville-Tuckean area, and to prevent localised impacts.
Within the whole groundwater source, no water is being made available for new licences, as
the source has been capped at the current entitlement. However, trading is allowed within
the groundwater source, but not if the trade results in a net increase to the sum of share
components in the Alstonville-Tuckean Management Zone. There are no restrictions to
trades within each of the management zones or from the Alstonville-Tuckean into the
Bangalow-Wyrallah zones.
The WSP is a regional water planning tool and a macro-scale instrument that considers a
geographically large area to set guidelines and regulations to limit overuse or stress on an
aquifer at a regional scale. The WSP does not specifically assess local scale risks, other
than setting broad limits on the proximity of groundwater extractions to groundwater
dependent ecosystems (GDEs), other groundwater users, aboriginal heritage sites and
major water supply bores. These are summarised for the four groundwater systems in the
rules in Appendix 3 (of the Initial Report). The issue of potential localised impacts from
extraction will be further investigated by the Review over the coming months.
Under the WSP, priority is given to environmental water and basic landholder rights (BLR).
The plan allocates an amount of water that is to be reserved for these priority uses, and
prevents licensed extractions from accessing a portion of the estimated groundwater source.
Licensed extractions for all other uses are secondary to BLR and environmental water.
Some additional priority is given to groundwater extracted for local and major utilities that are
typically licensed to extract reasonably large amounts for water security and to licensed
stock and domestic bores.
Under the Water Management Act 2000, granting of commercial water licences (e.g.
groundwater extractions that are not for BLR or utilities) is independent of their intended use.
This allows ready trade and aims to promote efficient use of resources, as well as allowing
new industries to develop and water to be allocated to the highest value use (DPI Water,
2016f). Under the Water Management Act 2000 and the WSP, extraction for bottled water is
treated the same as any other commercial extraction, including commercial irrigation and

134
horticulture. Further information about the WSPs, including audit and review processes are
contained in Appendix 4 (of the Initial Report).
The following sections review the purpose and methods used in the WSP that governs
groundwater licenses for the areas relevant to this review.

EXTRACTION LIMITS
A major output of the WSP is the LTAAEL for each of the 13 groundwater sources. The
LTAAEL determines the maximum average volume of water that can be extracted from one
source in a given year. To prevent the overuse of groundwater resources, the water sharing
plans generally enforce the LTAAEL to be a portion of the average recharge of the aquifer
(i.e. versus water that is already stored in the aquifer). To quantify an appropriate LTAAEL,
average recharge must be defined.
Groundwater sources in the WSP are described as ‘less highly-connected groundwater
sources’ (DPI Water, 2016f), so the estimates of aquifer recharge only consider direct
rainfall. While some surface water – groundwater connectivity may exist, no surface water
contribution to recharge is considered in these calculations. Under that assumption, the
LTAAEL for fractured rock aquifers is determined by the following simplified procedure (DPI
Water, 2015):
1. estimation of the annual average rainfall (discussed in Section 3.2.1)
2. estimation of the recharge as a fixed percentage of rainfall (discussed in Section
3.2.2)
3. determination of areas of high environmental value and non-high environmental
value
4. determination of current and future water requirements (the latter increased by 10
percent to ensure a conservative estimate)
5. determination of the UEL, which is equal to the recharge in the non-high
environmental value area multiplied by a sustainability index (described in Section
3.2.3)
6. determination of LTAAEL as the UEL or a lower value based on estimates of existing
and future extractions.
The current total water access rights, including an estimate of BLR (which does not require a
licence for extraction), was calculated for each of the groundwater sources as part of the
WSP in 2016. Where the total water access rights are less than the 80 percent of the
LTAAEL, water can be made available as new licence allocations by state government
agencies via a controlled allocation process.
When the WSP was first released in 2016, the Alstonville Basalt Plateau Groundwater
Source was the only source subject to the Review where licence allocations were at the
LTAAEL, whereas (the other three sources subject of this Review were less than 60 percent
allocated). There is a provision for the LTAAEL of fractured rock aquifers to be increased to
a maximum of the UEL if demand for water increases beyond the predicted amount.
Within the area covered by the WSP, licences entitle their holders to a certain ‘share’ of the
water resource. Under ordinary circumstances, one share is equal to an entitlement of
1ML/y. However, at the Minister’s discretion, the allocation of water per share can be
reduced to minimise environmental or socio-economic impacts, such as during a drought or
in response to a growth in use of local water utility or BLR use.
The following sections review the process for water allocation determination in the WSP.

135
Average Annual Rainfall
Annual rainfall is variable over the relatively large area covered by the WSP. Rainfall data
used by the WSP was sourced from the BOM gridded rainfall data (approximately 5 km2
grid) from 1901 – 2011. More information on the rainfall data can be found on BOM (2015).
This rainfall model uses algorithms to estimate a weighted average rainfall in each grid
based on the observations at the nearest BOM stations (see Section 1.2). This accounts for
the spatial variability of rainfall throughout large catchments and is based on the best
available data.
Recharge Rates
The WSP simplifies the aquifer recharge to the relationship shown in Equation 1. The
systems considered in the WSP are considered to be ‘less highly-connected groundwater
sources’ (DPI Water, 2016f), which means there is little recharge through creek beds, and
therefore, only recharge through rainfall was considered.
Calculating recharge through this relationship assumes that the aquifer is homogenous,
which is a simplification of the complex geology that occurs in porous and fractured rock
aquifers. It also assumes that the recharge is generated over the entire surface area of the
groundwater source that is not overlain by another defined groundwater source (i.e. the
outcropping area).

Average Recharge (ML/yr) = Average Rainfall (mm/yr) x Area (km2) x Recharge Rate (%)/100
Equation 1: Average recharge

While rainfall and area are measurable (although the area over which the source is
recharged is sometimes less clear, which is discussed further below, the recharge rate is
more difficult to define. The transmissivity of different aquifers can vary significantly
depending on the geology, and different recharge rates were applied by the WSP for each of
the groundwater sources. DPI Water (2015) provides some guidance on the recharge rates
applied for different groundwater source types (Table 25).

Table 25: Recharge rates recommended by DPI Water (2015)

Hydrogeological Type Recharge Rate Comment
Coastal Porous Rock 1 – 6% Based on the findings of Coastal
Porous Rock Rainfall Recharge
Study
Inland Porous Rock 6%
Fractured Rock 4%
(excl. North Coast Volcanics)
North Coast Volcanics 8% Source is unclear

Table 26 shows the recharge rates adopted in the WSP for the groundwater sources
relevant to this report. DPI Water (2016f) and DPI Water (2015) acknowledge that regional
estimates of recharge of large aquifers is not an exact science, and they state they that due
to this uncertainty have taken a precautionary approach. DPIE Water has advised that the
precautionary approach was based on using zero percent recharge estimates for high value
area, no allowance for recharge from anything other than direct rainfall, and sustainability
indexes to ensure that use is significantly less than recharge (DOI Water, 2019).

136
Table 26: Rainfall recharge rates adopted in the Water Sharing Plan
Groundwater Source Rainfall Recharge Based on
Rate adopted
Clarence Moreton Basin 6% There is little direct data and very little demand for
groundwater, therefore the NSW default 6 percent
was recommended, based on the Coastal Porous
Rock Rainfall Recharge Study (DPI Water, 2016f)
North Coast Volcanics 8% DPI Water (2016f)
New England Fold Belt 4% DPI Water (2016f)
Coast
Alstonville Basalt Plateau 8% Based on preceding WSP

Sustainability Index
The sustainability index (SI) is a qualitative risk based approach used in water sharing plans
to account for the relative social, economic and environmental risks of extracting
groundwater from a particular water source.
The environmental risk considers the prevalence of high priority groundwater dependent
ecosystems and the risk to the groundwater source itself. It considers water quality, ecology
and aquifer integrity. Environmental risk is rated as high (e.g. permanent and significant
change), moderate (temporary change) or low (no change anticipated) and is a simple
relative measure. If there are any mitigation actions (e.g. groundwater modelling or distance
rules from sensitive areas), these may be considered to lower the environmental risk.
Socio-economic risk considers the financial and social dependence of local communities on
a groundwater resource. For example, the socio-economic risk considers whether there is
any readily available alternative to groundwater extraction, the contribution of groundwater
dependent industry on the local economy (including employment rates) and the dependence
of the local communities on groundwater resources for drinking water supplies. As per the
environmental risk, the socio-economic risk is assigned a relative rating (high, moderate or
low).
Following these assessments, the environmental (known as the ‘aquifer risk’) risk and the
socio-economic risk are input into the matrix shown in Table 27 to define the final
sustainability index. For example, if the aquifer risk is classified as ‘High’ and the socio-
economic risk is ‘Medium’, the sustainability index would be 25 percent as illustrated in Table
27.
Table 27: Sustainability index matrix, with an example calculation of a high aquifer, medium socio-
economic risk sustainability index of 25%
Aquifer Risk High 5% 25% 50%
Medium 25% 50% 60%
Low 50% 60% 70%
High Medium Low
Socio-Economic Risk
Source: (DPI Water, 2016f)

The sustainability index is used to define the upper extraction limit (UEL – the maximum
allowable extraction from the groundwater source) as per Equation 2 below. The
sustainability index is the portion of estimated recharge that can be assigned to the UEL.
A lower sustainability index indicates less water is to be available for extraction (i.e. more
water is assigned as environmental water). All the catchments are split into two areas – high
conservation areas (e.g. National Parks) and the remaining areas. For all WSP groundwater
sources, the sustainability index over high conservation areas is, by default, 0 percent. This
means that recharge over these areas is preserved for environmental use. The sustainability
index calculated in Table 27 only relates to the remaining areas.

137
 UEL (ML/yr) = Recharge over non- high environmental areas (ML/yr) x SI(%)
Equation 2: UEL

Table 28 summarises the sustainability indexes for the four groundwater sources considered
in this report, including the assigned socio-economic and environmental risk. Environmental
risk of the North Coast Volcanics is high due to the prevalence of springs, rainforests and
groundwater dependent soils. The socio-economic risk in the Clarence Moreton Basin is
largely due to the predicted (at the time) reliance of the coal seam gas industry on
groundwater resources, as well as the dependence of the smaller industries on groundwater.
No socio-economic or environmental risk was provided for the Alstonville Basalt Plateau in
the WSP, as there were limited changes to the allowable extraction from the previous Water
Sharing Plan for the Alstonville Plateau Groundwater Source.

Table 28: Sustainability index for relevant groundwater sources

Groundwater Source Socio-Economic Risk Environmental Risk Sustainability Index
New England Fold Belt Low Moderate 25%
Coast
Clarence Moreton Basin Moderate Low 60%
North Coast Volcanics Moderate High 25%
Alstonville Basalt Plateau - - ~20%*
This is not presented in the current WSP but is based on the preceding legislation Water Sharing Plan for the Alstonville
Plateau Groundwater Source

Estimates of LTAAEL
The LTAAEL is calculated differently depending on whether the groundwater source is
defined as a porous or fractured rock aquifer. For fractured rock aquifers (New England Fold
Belt Coast, North Coast Volcanics and the Alstonville Plateau), the UEL is calculated as per
Equation 2, as a direct relationship between the recharge and sustainability index. However,
in an acknowledgement of the uncertainties surrounding the recharge estimates for fractured
rock, the upper extraction limit is compared to the current and estimated future requirements
for water (including a 10 percent buffer on the future requirements). The future estimated
requirements were calculated considering the following (DPI Water, 2015):
 growth in BLR as a result of increasing populations. BLR was assumed to grow in
proportion with population. Population forecasts were based on Department of
Planning estimates
 increase in requirements for dewatering, based on dewatering in the previous decade
increasing proportionally with population growth
 growth in town water supply requirements, sourced from future water strategies and
consultation with the relevant councils;
 growth in agricultural, which was determined by the North Coast Interagency
Regional Panel based on local knowledge and present agricultural requirements
 growth in mining requirements, based on industry statistics reviewed by the North
Coast Interagency Regional Panel.
Once the future estimated requirement for groundwater was calculated, the following rules
are applied to determine the LTAAEL:
1. if the future estimated requirement for groundwater (+10 percent) < 10 percent of
UEL, LTAAEL = 10 percent of UEL
2. if the future estimated requirement for groundwater (+10 percent) > UEL, LTAAEL =
UEL
3. otherwise, LTAAEL = future requirement for groundwater (+10 percent).

138
In cases where the LTAAEL<UEL, the LTAAEL can be increased during the life of the WSP
if the entitlement reaches 80 percent of the LTAAEL. This would require a review of the
LTAAEL (of one particular groundwater source) by the North Coast Interagency Region
Panel or some other similar interagency panel (DPI Water, 2016e). DPI Water (2016f) notes
that the future requirement estimates were ‘generous’ implying that it was considered
unlikely that there would be an increase in LTAAEL in the life of the plan.
For the New England Fold Belt Coast and the North Coast Volcanics, the LTAAEL is
substantially smaller than (< 25 percent of) the UEL (Table 29). This provides a suitable
buffer to account for the uncertainty related to the recharge rates for these areas, and results
in what is likely a conservative allocation of groundwater resources. For the Alstonville
Plateau, where the LTAAEL is based on the preceding WSP, the LTAAEL is relatively high
compared to the average annual recharge.

Table 29: LTAAEL in fractured rock aquifers as reported in February 2019 (Initial Report)
Groundwater Average Recharge Estimate Future UEL 10% of UEL LTAAEL
Source over non-high Requirement (+10) (ML/yr) (ML/yr) (ML/yr)
environmental areas (ML/yr)
(ML/yr)
New England 1,500,000 60,000 375,000 37,500 60,000
Fold Belt Coast
North Coast 220,000 13,000 55,000 5,500 13,000
Volcanics
Alstonville 50,000 - - - 8,895
Plateau*
Source: DPI 2016
Based on the preceding legislation Water Sharing Plan for the Alstonville Plateau Groundwater Source, no future requirement
or UEL was presented

For porous rock aquifers (Clarence Moreton Basin), the WSP states a higher degree of
confidence in the recharge rates due to the results of the Coastal Porous Rock Rainfall
Recharge Study. Further investigation is needed to determine the basis for this higher
degree of confidence.

Table 30: LTAAEL for porous rock aquifers as reported in February 2019 (Initial Report)
Groundwater Source Average Recharge over Current Requirement LTAAEL (ML/yr)
non-high environmental (ML/yr)
areas (ML/yr)
Clarence Moreton 500,000 4,562 300,000
Basin
Source: DPI Water (2016f)

LTAAEL values in the porous rock aquifers were calculated as per Equation 2, where the
LTAAEL is equal to the UEL. Unlike fractured rock aquifers, no reduction is made in the
LTAAEL to account for cases with low current and estimated future requirements for
groundwater extractions.
As a result, the LTAAEL for the Clarence Moreton Basin, shown in Table 30, is large
compared to the current water extraction. While this is an indicator that the groundwater
source is unlikely to be currently under stress, there is no trigger for review of the LTAAEL if
there is a large growth in extraction (as would be required for the New England Fold Belt
Coast or the North Coast Volcanics). However, the whole WSP is reviewed after a period of
ten years, so any significant growth in these porous rock aquifers could be reviewed at this
time.
Environmental Water
The WSP requires an assignment of a portion of the annual average recharge to be classed
as environmental water. As mentioned previously, 100 percent of recharge over high

139
conservation areas, such as National Parks, is preserved for environmental water. The total
volume of water assigned as RRE is defined by the relationship in Equation 3.

RRE (ML/yr) = Average Recharge (ML/yr) - LTAAEL

Equation 3: Recharge Reserved for the Environment (RRE)

Table 31 shows the RRE for the four groundwater sources of interest. The allotment of total
estimated recharge is illustrated graphically in Figure 24.
With the exception of the Clarence Moreton Basin, the RRE is in excess of 80 percent of the
estimated recharge. RRE is typically higher in fractured rock aquifers due to the more
conservative approach used to obtain a value of LTAAEL.

Table 31: Recharge amount reserved for the environment as reported in February 2019 (Initial Report)
Groundwater Source Total Estimated LTAAEL RRE RRE as a percentage of
1 1,2 1
Recharge (ML/yr) (ML/yr) (ML/yr) estimated recharge
New England Fold Belt 1,980,000 60,000 1,920,000 97%
Coast
Alstonville Plateau 50,079 8,895 41,184 82%
Clarence Moreton Basin 576,000 300,000 276,000 48%
North Coast Volcanics 310,000 13,000 297,000 96%
Source: (DPI Water, 2016f)
1. All numbers presented in this table are over the whole groundwater source and include recharge and environmental
water from high-conservation areas and less environmentally sensitive areas combined, which may differ from numbers
expressed in the WSP
2. Table 15 in WSP Background document refers to these values as planned environmental water ‘PEW’

Figure 24: Allotment of estimated recharge to Recharge Amount Reserved for the Environment (total
volumes differ between aquifers) as reported in February 2019 (Initial Report)

140
APPENDIX 4: BASEFLOW FILTERING TECHNIQUE

Baseflow was filtered from the total daily flow time series using a recursive digital filter
(Arnold et al., 1995; Arnold & Allen, 1999). It corresponds to an adaptation of digital filter
methods used in signal analysis, and assumes that low frequency baseflow could be
distinguish from high frequency flows. The filter equation is:
1+𝛽
𝑞𝑡 = 𝛽𝑞𝑡−1 + (𝑄𝑡 − 𝑄𝑡−1 ) Equation 1
𝑠

Where, qt is the filtered surface runoff (quick response) at day number t, Q is the original
streamflow, and β is the filter parameter. The filter can be applied to the stream flow data, in
both forward and reverse directions, as many times as desired. The number of times
determines the degree of smoothing of the baseflow hydrograph. In general, each pass will
result in less baseflow as a percentage of total flow. For this review, the filter parameter β
was set to 0.925, as determined by Nathan and McMahon (1990) and Arnold et al. (1995) to
provide realistic results.
Baseflow, bt, was then calculated via:
𝑏𝑡 = 𝑄𝑡 − 𝑞𝑡 Equation 2
Although the technique has no physical basis, it is objective and reproducible and has been
successfully compared with graphical (manual) methods of baseflow separation (Arnold et
al., 1995; Mau & Winter, 1997) and with measured field estimates (Arnold & Allen, 1999).
For this study the filter was passed three times; forward, backwards and forward for
smoothing the baseflow hydrograph based on the dataset length.

141
Baseflow filtering results

Figure 25: Baseflow filtering results - monitoring point 201001: Oxley River at Eungella

142
Figure 26: Baseflow filtering results monitoring point 201005: Rous River at Boat Harbour No. 3

143
Figure 27: Baseflow filtering results monitoring point 201012: Cobaki creek at Cobaki

144
Figure 28: Baseflow filtering results - monitoring point 201015: Tweed River D/S Palmers Road Crossing

145
Figure 29: Baseflow filtering results - monitoring point 201900: Tweed River at Uki

146
Figure 30: Baseflow filtering results - monitoring point 202001: Brunswick river at Durrumbul (sherrys crossing)

147
Figure 31: Baseflow filtering results - monitoring point 202002: Burringbar creek at Burringbar

148
Figure 32: Baseflow filtering results - monitoring point 203012: Byron Creek at Binna Burra

149
Figure 33: Baseflow filtering results - monitoring point 203014: Wilsons River at Eltham

150
Figure 34: Baseflow filtering results - monitoring point 203057: Houghlahans Creek at upstream Teven

151
Figure 35: Baseflow filtering results - monitoring point 203059: Maron Creek at Graham road

152
APPENDIX 5: MODELLING OF ALSTONVILLE PIEZOMETRIC DATA

Professor Louise Ryan

This appendix describes the GAM analysis for each of the Bore, Hole and Pipe combinations
listed in Table 12. As discussed in the main report, the Review used a statistical technique
called the Generalized Additive Model or GAM to explore how water levels varied over time,
and also to assess the extent to which rainfall and seasonal effects could explain the
observed levels.
We developed GAM models that predict piezometer level (in metres below measuring point)
as a function of time, rainfall and season. The Review explored a range of options for how to
best incorporate rainfall into the model. It found that ‘lagged rainfall’ averages, where rainfall
is reported for each day as the average rainfall over the previous x days (where x was
allowed to range from 10 days up to 240 days) provided a better explanation of variability in
the piezometers than daily rainfall.
For each piezometer, the Review team re-ran the GAM models to identify the most
appropriate lag time (x). The Review then explored the extent to which each component
contributes to the variability of each piezometer. To ensure that the significant tests were
appropriately adjusted for autocorrelation induced by the time-series nature of the data, a
technique called the block bootstrap (Kunsch, 1989) was used, which has been implemented
using the boot package in the statistical programming environment, R (Canty & Ripley,
2019). This analysis is an alternative to using the Seasonal Kendall Trend test which is
popular in hydrogeology.
Although the primary focus in this section of the Report lies in the period from 2009 onwards,
we included data from 2006 onwards in these analyses in order to boost statistical power to
detect effects. These figures were computed by running models leaving out each factor and
comparing the deviance explained by that model to the deviance explained by the model
with all three factors in included. Statistical significance was assessed through use of a
likelihood ratio test, using the block bootstrap to adjust for autocorrelation. Numbers that are
statistically significantly different from 0 at p<0.05 are indicated by an asterix. A higher
percentage indicates that the water levels in that piezometer are more highly correlated to
that particular variable.
This appendix shows the results of fitting the GAM models to each of the bores. The results
are summarised in the Table given in the main body of the Report. Note that no data were
available after 2006 for the following bores:
 GW036701 hole 1 pipe 4
 GW036702 hole 1 pipe 4
 GW081004 hole 1 pipe 1

153
Analysis for site GW036701 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 240 days. Adding
rain into the model explained an additional 32.61 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 11.11 percent of variability compared
to a model with just rain and time. However, this effect was not statistically significant (p=
0.128).
Adding time into the model explained an additional 41.26 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There were statistically significant non-linearities in the time effect (p= 0.018).

154
Analysis for site GW036701 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 7.6 percent of variability compared with a model
that included just time and season. But this increase was not statistically significant (p=
0.217).
Adding season into the model explained an additional 3.26 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.462).
Adding time into the model explained an additional 97.91 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There was no statistically significant non-linearity in the time effect (p= 0.106).

155
Analysis for site GW036702 hole 3 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 49.42 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 4.27 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.296).
Adding time into the model explained an additional 8.01 percent of variability compared to a
model with just rain and season, but this effect was not statistically significant (p= 0.052).
There was no statistically significant non-linearity in the time effect (p= 0.058).

156
Analysis for site GW036702 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 28.9 percent of variability compared with a model
that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 21.11 percent of variability compared
to a model with just rain and time. However, this effect was not statistically significant (p=
0.066).
Adding time into the model explained an additional 17.83 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.01).
There was no statistically significant non-linearity in the time effect (p= 0.193).

157
Analysis for site GW040999 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 30 days. Adding rain
into the model explained an additional 57.37 percent of variability compared with a model
that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 8.34 percent of variability compared to
a model with just rain and time. This effect was statistically significant (p= 0.009).
Adding time into the model explained an additional 10.62 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.002). There
were statistically significant non-linearities in the time effect (p= 0.007).

158
Analysis for site GW040999 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 40 days. Adding rain
into the model explained an additional 1.46 percent of variability compared with a model that
included just time and season. but this increase was not statistically significant (p= 0.734).
Adding season into the model explained an additional 7.27 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.267).
Adding time into the model explained an additional 98.24 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001). There were
statistically significant non-linearities in the time effect (p= 0.016).

159
Analysis for site GW041000 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 16.59 percent of variability but this increase was
not statistically significant (p= 0.14).
Adding season into the model explained an additional 2.3 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.743).
Adding time into the model explained an additional 93.79 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There were statistically significant non-linearities in the time effect (p<.001).

160
Analysis for site GW041000 hole 1 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 20.68 percent of variability but this increase was
not statistically significant (p= 0.061).
Adding season into the model explained an additional 3.03 percent of variability compared to
a model with just rain and time.
However, this effect was not statistically significant (p= 0.576).
Adding time into the model explained an additional 95.01 percent of variability compared to a
model with just rain and season.
This effect was statistically significant (p= 0.001).
There were statistically significant non-linearities in the time effect (p<.001).

161
Analysis for site GW041001 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 30 days. Adding rain
into the model explained an additional 68.44 percent of variability compared with a model
that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 8.47 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.126).
Adding time into the model explained an additional 8.18 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.031).
There were statistically significant non-linearities in the time effect (p= 0.018).

162
Analysis for site GW041001 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 40 days. Adding rain
into the model explained an additional 2.7 percent of variability but this increase was not
statistically significant (p= 0.406).
Adding season into the model explained an additional 1.71 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.568).
Adding time into the model explained an additional 99.02 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There were statistically significant non-linearities in the time effect (p<.001).

163
Analysis for site GW041002 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 200 days. Adding
rain into the model explained an additional 61.02 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p= 0.001).
Adding season into the model explained an additional 4.54 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.151).
Adding time into the model explained an additional 75.22 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.005).
There were statistically significant non-linearities in the time effect (p= 0.016).

164
Analysis for site GW041003 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 30 days. Adding rain
into the model explained an additional 38.81 percent of variability compared with a model
that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 10.76 percent of variability compared
to a model with just rain and time. This effect was statistically significant (p= 0.001).
Adding time into the model explained an additional 7.78 percent of variability compared to a
model with just rain and season, but this effect was not statistically significant (p= 0.259).
There was no statistically significant non-linearity in the time effect (p= 0.228).

165
Analysis for site GW041003 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 180 days. Adding
rain into the model explained an additional 63.43 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p= 0.001).
Adding season into the model explained an additional 5.04 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.383).
Adding time into the model explained an additional 18.75 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.014).
There were statistically significant non-linearities in the time effect (p= 0.023).

166
Analysis for site GW041004 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 50.21 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p= 0.001).
Adding season into the model explained an additional 11.22 percent of variability compared
to a model with just rain and time. This effect was statistically significant (p= 0.049).
Adding time into the model explained an additional 7.78 percent of variability compared to a
model with just rain and season, but this effect was not statistically significant (p= 0.093).
There was no statistically significant non-linearity in the time effect (p= 0.07).

167
Analysis for site GW041004 hole 2 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 50.63 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 11.08 percent of variability compared
to a model with just rain and time. This effect was statistically significant (p= 0.039).
Adding time into the model explained an additional 8.17 percent of variability compared to a
model with just rain and season but this effect was not statistically significant (p= 0.191).
There was no statistically significant non-linearity in the time effect (p= 0.178).

168
Analysis for site GW041005 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 30 days. Adding rain
into the model explained an additional 53.08 percent of variability compared with a model
that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 2.76 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.169).
Adding time into the model explained an additional 5.17 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.044).
There were statistically significant non-linearities in the time effect (p= 0.028).

169
Analysis for site GW041007 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 100 days. Adding
rain into the model explained an additional 54.98 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 7.11 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.078).
Adding time into the model explained an additional 18.85 percent of variability compared to a
model with just rain and season but this effect was not statistically significant (p= 0.056).
There were statistically significant non-linearities in the time effect (p= 0.008).

170
Analysis for site GW041007 hole 1 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 120 days. Adding
rain into the model explained an additional 50.81 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 6.99 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.062).
Adding time into the model explained an additional 22.23 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.007).
There were statistically significant non-linearities in the time effect (p= 0.003).

171
Analysis for site GW041008 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 120 days. Adding
rain into the model explained an additional 56.38 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 10.49 percent of variability compared
to a model with just rain and time. However, this effect was not statistically significant (p=
0.087).
Adding time into the model explained an additional 36.79 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.007).
There were statistically significant non-linearities in the time effect (p= 0.01).

172
Analysis for site GW041008 hole 1 pipe 2 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 57.62 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 15.87 percent of variability compared
to a model with just rain and time. This effect was statistically significant (p= 0.004).
Adding time into the model explained an additional 36.41 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.017).
There were statistically significant non-linearities in the time effect (p= 0.006).

173
Analysis for site GW081000 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 46.88 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 3.58 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.202).
Adding time into the model explained an additional 15.33 percent of variability compared to a
model with just rain and season but this effect was not statistically significant (p= 0.072).
There was no statistically significant non-linearity in the time effect (p= 0.07).

174
Analysis for site GW081001 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 2.46 percent of variability but this increase was
not statistically significant (p= 0.421).
Adding season into the model explained an additional 10.95 percent of variability compared
to a model with just rain and time. This effect was statistically significant (p= 0.044).
Adding time into the model explained an additional 99.35 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There were statistically significant non-linearities in the time effect (p<.001).

175
Analysis for site GW081002 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 44.91 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p= 0.001).
Adding season into the model explained an additional 7.05 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.462).
Adding time into the model explained an additional 71.2 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There was no statistically significant non-linearity in the time effect (p= 0.277).

176
Analysis for site GW081003 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 150 days. Adding
rain into the model explained an additional 66.13 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 13.57 percent of variability compared
to a model with just rain and time. However, this effect was not statistically significant (p=
0.129).
Adding time into the model explained an additional 35.11 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.004).
There were statistically significant non-linearities in the time effect (p= 0.009).

177
Analysis for site GW081005 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 280 days. Adding
rain into the model explained an additional 1.77 percent of variability but this increase was
not statistically significant (p= 0.48).
Adding season into the model explained an additional 0.83 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.585).
Adding time into the model explained an additional 99.23 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p<.001).
There were statistically significant non-linearities in the time effect (p<.001).

178
Analysis for site GW081006 hole 1 pipe 1 data from 2006 only
The highest percent variation explained was given by lagging rain over 120 days. Adding
rain into the model explained an additional 43.52 percent of variability compared with a
model that included just time and season. This effect was statistically significant (p<.001).
Adding season into the model explained an additional 5.02 percent of variability compared to
a model with just rain and time. However, this effect was not statistically significant (p=
0.264).
Adding time into the model explained an additional 42.1 percent of variability compared to a
model with just rain and season. This effect was statistically significant (p= 0.023).
There were statistically significant non-linearities in the time effect (p= 0.019).

179
APPENDIX 6: CASE STUDY - LUMLEY PARK AND CONVERY’S
LANE, ALSTONVILLE

This case study provides one of the few examples for the region and illustrates the
complexities of evaluating and ascertaining potential connections between the deeper
groundwater system, the shallower aquifers and impacts on local assets. However, it is not
provided to specifically examine connectivity and drawdown effects.
Rous County Council has TWS bores at Lumley Park and Convery’s Lane that draw from the
deeper aquifer in the Alstonville Basalt Plateau Groundwater system (Table 32). These
bores have allocations that provide town water supply at around ten times or higher than
most licences for bottled water extraction. These bores are also in close proximity to the
DPIE Water monitoring bores.
Table 32: Details for TWS bores, the allocation, depth of the bore, and nearby DPIE groundwater
monitoring piezometers
TWS Bore TWS bore TWS depth (m DPIE Depth of Monitoring
Name allocation below monitoring monitoring bore year
(ML/y) measuring bore piezometer installed
point) (screen
interval)

Convery’s Lane 253 111 GW036702 17-21m 1987

GW036702 150-168m 1987

Lumley Park 530 82 GW081005 60-71 m 1999

(max historical
extraction 192)

GW081006 7.5-12 m 1999

GW41001 -1 Shallow* 2005-06

GW41001 -2 Deep* 2005-06

*the exact depth was not available

In 2006 a comparison was made of the Convery’s Lane TWS bore pumping data with the
deeper groundwater levels at the nearby DPIE Water monitoring bore GW036702 around
one km away. At the time, investigations found that the pressure heads in the deeper
groundwater system dropped significantly when the Convery’s Lane TWS bore was
pumping, indicating that pumping was not sustainable. Once these pumps ceased it took
several years for pressure heads to recover (Green, 2006) (Figure 36). In 2003, Rous
County Council ceased using the Convery’s Lane bore due to its effect on the deeper
groundwater levels (Rous Water, 2014). Further investigation would be required to ascertain
the hydrogeologic relationship between the deep level pumping, the shallow water levels and
rainfall.

180
Figure 36: Convery's Lane TWS bore usage (ML/y) with GW036702 (deeper and shallow) levels and rainfall
data.
The Convery’s Lane usage shows gaps in extraction in 2001-2002 and from 2003. The bore usage is shown in
subset graph in black columns. Figure modified from (Parsons Brinckerhoff, 2011).The nearby bores shows a
response from the deeper monitoring bore (depth to groundwater shown as below measuring point in green),
while the monitoring bore GW036702.3.1 (shallow) may be associated with rainfall (nearby Bureau of
Meteorology rainfall stations, shown in blue columns). Data source: WaterNSW and BoM

In the case of the TWS bore at Lumley Park, in contrast to that at Convery’s Lane,
investigations of the groundwater pressure heads at the nearest DPIE Water monitoring bore
around one km away (GW081005 (deeper) and GW081006 (shallow)) in 2006 indicated that
drawdown during pumping at this site had limited influence on groundwater levels (Parsons
Brinckerhoff, 2011). This may indicate a confined or partially confined aquifer and help to
ensure that a remnant subtropical rainforest (Scientific Committee, 2019) is not impacted by
the TWS. This rainforest is likely a GDE that contains highly diverse vegetation,
invertebrates and fauna including a small melaleuca swamp community with platypus
(Moore, 2014). However, it has been estimated that this rainforest most likely relies on the
shallow groundwater zone during dry periods and contains at least one spring feeding into
Maguires Creek (Green, 2006; Parsons Brinckerhoff, 2011).
Previous analysis indicated that groundwater pressure heads within the TWS bore at Lumley
Park, or in close proximity to it, were likely to be temporarily lowered during the period it was
operational (2002-2006) (Parsons Brinckerhoff, 2011). However, the deeper groundwater
levels at the monitoring bore in proximity remained around 45 m below ground level
(Parsons Brinckerhoff, 2011) (Figure 37). The TWS bore was used again for a period in

181
between late August and late December 2007, but has not been operational since (Rous
County Council, 2019a, 2019b).
Additional monitoring piezometers were installed in 2005-2006 (GW041001_1 and _2) near
the TWS bore at Lumley Park (~10-20 m away) and Lumley Cutting (~50-100 m away). The
deeper levels during this period were generally stable, with the four dips potentially attributed
to measurement error, periods when the data from the logger was being downloaded or the
logger was down (Figure 37 and Figure 38). Small rises in the shallow levels seem to follow
the significant rainfall events. It was previously reported that groundwater levels in the
shallow aquifer are rapidly recharged with rainfall events (Green, 2006), which can be seen
in Figure 38.

Figure 37: Lumley Park TWS bore usage (black columns) and water level and/or pressure observed at DPIE
monitoring bores GW081006 (shallow) and GW081005 (deeper) from 2002 to 2009.
Source: Rous County Council (2019a) and WaterNSW data register (WaterNSW, 2019)

182
Figure 38: Groundwater levels from DPIE monitoring bores near Lumley Park plotted with rainfall
Groundwater levels from DPIE monitoring bores near Lumley Park: GW081006 (shallow, purple line) and GW081005 (deeper,
green line) and two more recent monitoring bores GW041001.1.1 (shallow, light green line) and GW041001.2.2 (deeper, light
blue line, note the four dips are most likely associated with logger servicing); and the rainfall amount (nearby Bureau of
Meteorology rainfall stations, shown in blue columns). Data source: WaterNSW and BoM

At Lumley Park, the monitored levels did not appear to support an immediate link between
extraction and the deeper and shallow levels nearby; whereas at Convery’s Lane there
appeared to be a link between extraction and drawdown from the deeper aquifer. This
example highlights the complexity of the spatial and temporal variation across the fractured
rock aquifers despite monitoring of the deeper and shallow aquifers over extended periods.
However, these investigations also highlight the value of properly conducted investigations,
which include field investigations, conceptual model development, and bore testing.
As these cases do not include monitoring of the local GDEs, it is difficult to draw conclusions
about environmental impacts. Further monitoring at Lumley Park (Parsons Brinckerhoff,
2011) and an assessment of the interaction with the GDE could determine if the pumping
rate may cause unacceptable drawdown (Moore, 2014). Monitoring suggestions, highlighted
in reports prepared for Rous, included monthly data collection of groundwater levels and
parameters including: EC, pH, temperature and redox potential, as well as annual monitoring
for major ions, metals, and nutrients to detect any potential changes in groundwater quality
(Parsons Brinckerhoff, 2011).

183
APPENDIX 7: SETBACK RULES FROM THE WSP

Table 33: Minimum distance rules to minimise interference between bores in fractured rock groundwater
sources (Alstonville Basalt Plateau, New England Fold Belt Coast, and North Coast Volcanics)
Other bore/asset type Minimum distances
An existing bore that is not used for basic rights 200m (bores < 20ML/yr)
400m (bores > 20ML/yr)
An existing bore that is used for basic rights 200m
The boundary of the property (unless consent gained from neighbour) 100m
A local or major water utility bore 500m
A bore used by the Department for monitoring purposes 400m
Exceptions – the above restrictions do not apply if either:
 The bore is used solely for basic rights;
 The bore is a replacement bore;
 The bore is used for monitoring, environmental management or remedial works; or
 The location of the bore would result in no more than minimal impact on existing extractions within the
water source.
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 40

Table 34: Minimum distance rules to minimise interference between bores in porous rock groundwater
sources (Clarence Morton Basin)
Other bore/asset type Minimum distances
An existing bore that is not used for basic rights 400m
An existing bore that is used for basic rights 100m
The boundary of the property (unless consent gained from neighbour) 50m
A local or major water utility bore 1000m
A bore used by the Department for monitoring purposes 200m
Exceptions – the above restrictions do not apply if either:
 The bore is used solely for basic rights;
 The bore is a replacement bore;
 The bore is used for monitoring, environmental management or remedial works; or
 The location of the bore would result in no more than minimal impact on existing extractions within the
water source.
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 40

Table 35: Minimum distance rules to minimise contamination

Contamination source Minimum distances
The plume associated with a Any distance from the plume that is insufficient to protect the groundwater
contamination source source; or
250-500m if no drawdown will not occur within 250 m of plume; or
or 250m
Exceptions – the above restrictions do not apply if either:
 The distance is adequate to protect the groundwater source, its dependent ecosystems and public
health and safety; or
 The bore is used for monitoring, environmental management or remedial works.
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 41

184
Table 36: Minimum distance rules to minimise impacts on GDEs and environmentally sensitive areas
GDE or environmentally sensitive area Minimum distances
A high-priority GDE 100m (for bores used for
basic rights)
A high-priority GDE or the outside perimeter of a National Park estate 200m (for bores not used for
basic rights)
A high-priority karst environment GDE 500m (for bores not used for
basic rights)
A river or stream (1st, 2nd or 3rd order) 40m (for bores not used for
basic rights)
An escarpment 100m (for bores not used for
basic rights)
Exceptions – the above restrictions do not apply if either:
 The water supply works (bores) are used for monitoring, environmental management purposes or
remedial work;
 A hydrogeological study demonstrates no drawdown of the groundwater at the outside edge of the GDE;
or
 No more than minimal impact will occur to any groundwater dependent vegetation in the nearby National
Park estate.
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 42

Table 37: Minimum distance rules to minimise impacts on groundwater-dependent culturally significant
sites
Site Minimum distances
A groundwater-dependent culturally significant site 100m (for bores used for
basic rights)
A groundwater-dependent culturally significant site 200m (for bores not used for
basic rights)
Exceptions – the above restrictions do not apply if either:
 The bore is used for monitoring, environmental management or remedial works; or
 The location of the bore at a lesser distance would result in no greater impact on the groundwater
source and its groundwater dependent culturally significant sites.
Water Sharing Plan for the North Coast Fractured and Porous Rock Groundwater Sources 2016 cl 43

185
APPENDIX 8: DECISION MAKING AND UNCERTAINTY

As part of Term of Reference 1, the Review was requested to provide advice on the
sustainability of the extraction limits in the relevant Water Sharing Plan (WSP) for
groundwater sources in the Northern Rivers.
While in common use, the term ‘sustainability’ is complex and underpinned by principles that
require interpretation and consideration of multiple, changing and sometimes competing
factors. The Protection of the Environment Administration Act 1991 considers the
sustainability to be informed by integration of social, economic and environmental
considerations in decision-making processes. For the purposes of this report, ‘groundwater
sustainability’ relates to managing the current (environmental and anthropogenic) use of the
resource to provide for ensure long-term water security.
Regulatory instruments provide a framework and strategies to help achieve this long-term
security, including adaptive management, risk assessment of proposed developments and
approval conditions for licences and development applications. However, within these
frameworks, judgements still need to be made. Salient questions in exercising judgement
include for example, how much risk is acceptable, to whom, under what circumstances, with
what information and what consequences and the degree of confidence in the assumptions
made, what tools can be drawn on to reduce uncertainty in a way that is cost effective and
proportional to the level of risk.
The Review recognises the community concern about water allocations and use, and the
desire for greater certainty and more definitive information about sustainability to inform
decision making at a regional, local and individual scale. Yet all decisions entail some
degree of uncertainty, and all decisions are made in the context of imperfect knowledge.
At an individual- and community-level we are faced on a daily basis with the challenge of
making decisions that balance risks and potential benefits of actions. These range from the
routine to significant, may be implicit or explicit and are dynamic in light of changing
knowledge – with advances and evolutions in science often shifting the balance. All are
influenced by the actions and/or opinions of others, communal decisions made more
complex by diverse stakeholder priorities and values.
In undertaking its work, the Review considered risk and uncertainty and how these have
been managed by decision-makers and proponents from a scientific perspective based on
currently available knowledge. This Section sets out how these concepts were approached
by the Review and informed consideration of sustainability and impacts under the Terms of
Reference.

TYPES OF UNCERTAINTY
Much has been written, both for academic and popular science audiences, on the topic of
decision making under uncertainty. Some issues are well recognised – for example the need
to account for sample variability, computational uncertainty or establishing appropriate
margins of error. However, increasing recognition is being given to the importance and best
means of communicating uncertainty to diverse audiences and recognition of the emotional
side of decision making under uncertainty. Some offer practical guidance for individuals.
The prestigious US National Academy of Sciences (NAS) convened an expert panel several
years ago to develop guidelines around environmental decision making under uncertainty
(Institute of Medicine, 2013). The report had a strong focus around human health risks
associated with environmental exposures. Notwithstanding this focus, the report offers many
useful principles that can help in contexts where a regulatory body is faced with the
challenge of making a complex decision.

186
One of the first steps in any uncertainty analysis should be a clear articulation of the various
sources of, and nature of, the uncertainties involved and it is useful to distinguish between
the different types of uncertainty.
Statistical variability, sometimes referred to as aleatory uncertainty, refers to natural variation
in the physical environment and in human behaviour and biology. In the context of this
Review, there will be statistical variability associated with daily rainfall levels or with how
much water is actually extracted by an individual licensee within the maximum amount
allowed.
This kind of variability is inherent to the system and cannot be reduced by collecting further
data, though the latter may be extremely useful in helping decision makers to understand
further this kind of uncertainty. This first type of uncertainty is, at least in broad principle,
easy to accommodate through statistical modelling.
Model uncertainty refers to the fact that in virtually every area of enquiry, there will be
uncertainties associated with the conceptual understanding of the relevant science that
describes the context of interest. A useful definition of scientific modelling that recognises
both its values and limitations is
“The generation of a physical, conceptual, or mathematical representation of a real
phenomenon that is difficult to observe directly. Scientific models are used to explain and
predict the behaviour of real objects or systems and are used in a variety of scientific
disciplines, ranging from physics and chemistry to ecology and the Earth sciences. Although
a central component of modern science, scientific models at best are approximations of the
objects and systems that they represent—they are not exact replicas. Thus, scientists
constantly are working to improve and refine models” (Rogers, 2011).
It is useful to subdivide model uncertainty into uncertainty associated with the broad nature
of the model itself as well as uncertainty associated with particular parameters or inputs
needed to characterise a particular model.
In some contexts, there can be substantial scientific debate about the appropriate
conceptual model for a situation at hand. In the Review, an example of this uncertainty is in
the appropriate method to calculate the recharge rate of groundwater. As will be seen later in
Section 4.3.1, there are two approaches that can be used to model the recharge rate of
groundwater, being Chloride Mass Balance and the baseflow filtering approach. The choice
of which model to use may be based on a conceptual understanding of the mode of
recharge (local or widespread, rainfall to groundwater versus rainfall to surface water to
groundwater), previous experience, availability of data and tools.
Even in settings where scientists agree in broad terms over the appropriate conceptual
modelling framework, there will often be uncertainty over the particular inputs needed to
precisely define that model. This is referred to as parameter uncertainty.
Parameter uncertainty can generally be reduced through additional data collection, though
this may involve time consuming and costly effort. In the context of this Review, it is likely
that hydrogeologists would agree in broad terms about how one should go about
constructing a model to characterise the aquifers in the Northern Rivers region. However,
there might be variations in opinions regarding precise approaches.
The greatest source of uncertainty in this context arises from limitations in the availability of
data to inform the right inputs to these models and to help define the needed model
parameters. This would include data from geological surveys designed to help characterise
the nature and structure of the aquifers and to elucidate their recharge behaviours.
There is a fourth kind of uncertainty referring to settings where there may be fundamental
disagreements about the nature of the processes driving the situation of interest or where it
is impossible to collect all the data needed to properly inform the system due to cost and
time considerations. The term deep uncertainty is sometimes used to describe this kind of

187
uncertainty. This kind of uncertainty classically arises in settings where decisions may have
long-term consequences, but where it is not possible to accurately predict the future with full
accuracy.
The process of decision making under uncertainty naturally follows several phases. The first
phase involves problem formulation and scoping, creating an inventory or even a taxonomy
of the uncertainties associated with a particular decision making context. This would involve
listing out the various sources of statistical variation and heterogeneity. As part of the
process in identifying and listing these sources, it is important to assess whether a particular
source of heterogeneity might have impact on the decision and hence need to be
incorporated specifically, or whether it is simply a source of heterogeneity that can be noted
and then set aside and not considered further. In the hydrogeological context for example, it
is typical to recognise that while there will always be a lot of small-scale fluctuations in the
structure of a porous aquifer, it is not necessary to capture these precisely and only a
general, larger-scale description of the aquifer characteristics may be needed.
As part of the first scoping phase of a decision, it is very important to assess whether some
of the sources of uncertainty could be reduced relatively easily and in an acceptable
timeframe through additional data collection or even research. It will also be critical to
identify any sources of deep uncertainty and also to decide on the broader strategies that will
be used to incorporate the identified uncertainties into the decision making process and
ongoing risk management. Applying appropriate strategies to account for and manage those
uncertainties correspond to the second and third phases of decision making under
uncertainty, the focus of the following section.

APPROACHES TO INCORPORATING AND MANAGING UNCERTAINTY

A variety of modern-day tools are available to help with the incorporation of uncertainty
considerations into decision making. Indeed, the science of Decision Theory goes back to
the work of probability theorists such as Pascal and Bernoulli in the 17th and 18th centuries
who discovered identified that people do not always react completely rationally and
predictably when it comes to making decisions under uncertainty. These early developments
were largely done in the context of gambling games where the choices and associated
losses or gains were fairly simple.
The concept of Utility was developed to measure the value that people place on certain
outcomes happening and then the decision making could be framed in terms of choosing the
action with the highest expected utility. Alternatively, strategies such as minimax (choosing
the option that minimizes the worst outcome) can be used in settings where it is difficult to
assign probabilities to the relevant scenarios. Polasky et al. (2011) discuss these ideas in
the context of environmental impact assessment. However, they make the point that these
fairly simple classical decision theory tools work well only in settings where existing
information is extensive and where the probabilities, risks and benefits associated with
various decisions are well delineated. In most complex real world settings, more
sophisticated tools are needed.
Modern decision science has evolved considerably in order to have relevance in and
applicability in complex real-world settings. For example, there have been extensions to so-
called multi-attribute utility analysis for settings involving multiple different outcomes. Cost-
benefit analysis is an example. Extensions to the setting of multiple decision makers led to
the field of game theory which has found wide application and interest from economists.
Tools such as probabilistic risk assessment were proposed in the late 80s and 90s as a
means of incorporating uncertainty into the modelling process.
While probabilistic risk assessment cannot remove uncertainty, it provides a means of
enabling decision makers to gain a clearer understanding of the impact of various sources of
uncertainty on the outcomes of interest. Probabilistic risk assessment typically uses Monte

188
Carlo simulation and Bayesian methods to add extra layers to the modelling process. It
works very well in terms of addressing the second type of uncertainty, model and parameter
uncertainty. An example would be in areas of water quality and risk assessment, some
stakeholders may have concerns that variation in the amount of water drunk by individuals
each day might affect the estimated dose-response of contaminants and hence impact on
the decision making process. By extending classic dose-response modelling to incorporate
this variability, it is possible to explicitly assess the impact of this variability. Probabilistic
Risk Assessment has also been adopted by the US Nuclear Regulatory Commission.
In very complex settings, the number of scenarios needing to be considered can easily
balloon out to an unmanageable level. Some new computational tools have been recently
developed to handle this. For example, MIT researchers utilise Bayesian networks to
efficiently evaluate and compare thousands of decision options in the context of robotics and
autonomous vehicle management (Kochenderfer et al., 2015; Hodgett & Siraj, 2019)
describe a computational tool that builds uncertainty into a complex decision framework via a
series of triangular distributions.
Bayesian modelling approaches can also be used in settings where there are uncertainties
about the model to be used. In data-rich settings, statistical methods can be used to guide
the choice between different models or even to build a “meta-model” that includes multiple
models as special cases. In complex settings such as groundwater modelling, model
specification requires the input of experts with deep knowledge of the subject. Once a model
has been specified, there will still be a need to use a combination of data and informed by
expert knowledge to estimate model parameters. (Peterson & Western, 2014) used this kind
of approach in the context of groundwater modelling.
Rojas (2010) refer to this as a multi-model approach and discuss how this kind of approach
can be used to consider the impact of various future scenarios. However, this kind of
approach can be difficult to apply in practice. While it naturally allows for a wide range of
opinions about the right conceptual model, it still requires that there be enough data
available to help quantify the different sources of uncertainty. These approaches can also be
computationally very complex when the individual models in the multi-model all require the
running of a time consuming hydrogeological model. This can also make such models very
expensive to develop. Asher et al. (2015) discuss a more computationally feasible approach
based on surrogate models that approximate a complex hydrogeological model with an
empirical model that captures the relationship between various model inputs and expected
outcomes.
However, the greatest challenge in complex real-world settings is not so much running the
models, but delineating all the different elements involved in the decision making and
characterising the probabilities and uncertainties associated with these events. In settings
that are data-poor or subject to deep uncertainties, the more mathematical tools described
above become less relevant since it becomes almost impossible to attach realistic
probabilities to the various settings being considered. While the ideal is of course to reduce
uncertainty in order to create more reliable predictive models of environmental systems, this
step can be time consuming, expensive and potentially unfeasible in the timeframe needed
for decision making. Polasky et al. (2011) In such cases other more pragmatic solutions may
be taken such as adaptive management, with monitoring and feedback steps to maintain an
up-to-date view on the trajectory of an issue so that changes can be made, including
potentially decisions to cease activity, informed by new information.
Adaptive management is a precautionary measure in certain cases where there is
uncertainty, defined as a “procedure for implementing management while learning about
which management actions are most effective at achieving specified objectives.” (OEH,
2018). It is an “iterative based approach involving explicit testing of the achievement of
defined goals” (Preston, 2017).

189
The Water Management Act 2000 provides that “the principles of adaptive management
should be applied, which should be responsive to monitoring and improvements in
understanding of ecological water requirements.” The NSW Land and Environment Court of
NSW has held that an adaptive management approach might involve monitoring
management impacts, research, periodic evaluation of outcomes and learning reviewing and
adjusting in light of these and establishing effective compliance systems.
Scenario planning provides an appealing method to facilitate thinking and planning about
potentially complex future events and outcomes. Scenario planning is less quantitative than
traditional decision theory approaches, relying instead on a set of detailed stories that reflect
possible changing conditions over time. An advantage of the scenario approach is that it
allows the incorporation of complex interplays between social, economic and physical factors
such as climate. However, this flexibility and capacity also leads to the main weakness of the
approach, namely the difficulty in quantifying the relative likelihoods of the various scenarios.
Also important is to prepare responses to potential scenarios with action ‘trigger’ points,
thereby avoiding both the risk of automatically defaulting to a ‘middle’ option or over-
investing to manage theoretical extremes unless required.
A threshold approach to decision making involves identifying critical boundaries that might
have major implications if crossed. Setting emissions caps in the context of planning related
to climate change is an example of a threshold approach to environmental management.
Resilience thinking refers to the idea of organizing decisions so that they can adapt or
transform to a new mode of operation should the old mode become unworkable. Adaptive
monitoring in those settings, emphasizing the importance of having access to good quality
data that can be used to monitor the context of interest and potentially being used to trigger
alerts should problems arise. In the context of aquifer management, having access to
reliable data from monitoring bores can play a critical role in terms of assessing the long-
term viability of the system and activities.
Polasky et al. (2011) also argue that most situations can benefit from the use of multiple
tools and stress the importance of thinking of decision-making as a dynamic process that
can responsively adapt in the face of change and of new information. Fletcher, Lickley, and
Strzepak (2019) discuss similar ideas in the context of water resource planning. This is
consistent with statutory and policy approaches described earlier. In the context of the
Review, this kind of adaptive planning and decision making would require that reliable data
be available to inform on the state of the various aquifers. Section 3.4 discusses how the
data from the network of 29 functional monitoring bores in the Alstonville Plateau region can
potentially be monitored in a real-time manner and how such analyses can either provide
reassurance that the system is in good health or perhaps trigger a warning that some
change might be needed.

Communicating uncertainty
Once the various sources of uncertainty have been identified and a strategy developed for
decision making in that context, the next step involves ensuring that issues of uncertainty are
communicated to various stakeholders and other audiences. There has been significant
research undertaken into the effectiveness of visual and descriptive versus numerical
representations of the uncertainty in risk. Professor David (Spiegelhalter, 2017) has written
for both the scientific community and the general public about the importance of using clear
language and graphical displays to help audiences understand the nature and sources of
uncertainty and magnitude of consequences. An overriding principle is that information
needs to be presented in a clear and digestible way. Greater attention is also needed to
evidence about how visual representations including infographics are processed and
understood by different reading audiences (Spiegelhalter, Pearson, & Short, 2011).

190
APPLICATION OF PRINCIPLES AND CONCEPTS BY THE REVIEW
The Review has examined the potential impacts and consequences of groundwater
extraction for bottling purposes having regard to the statutory context in which water
resources are allocated and managed and approaches of decision makers at regional and
local levels to understand and manage risk and uncertainty.
In so doing, the Review has analysed the assumptions underpinning the relevant WSP,
including the strategies deployed and level of conservatism applied to assumptions to
manage uncertainty. The Review undertook further analyses and gave consideration to
comparable and alternate approaches to managing uncertainty.
The Review was cognoscente of the complexity of the groundwater system, including
potential groundwater and surface water interactions in confined and unconfined aquifers
and implications this has for any extraction. The Review accepted the assumption that
drawing groundwater from a bore will have some impact on the water balance, both spatially
and temporally, and may have potential consequences for other water assets in the vicinity,
including the environment and other groundwater users. These consequences can be
related to changes in both water quantity and quality that may not emerge in the short term.
At the same time, an effect on a system may be a measurable effect but may not have
significant consequences or be of lesser significance relative to other factors at play.
Insofar as possible the Review has sought to provide pragmatic and feasible suggestions to
improving knowledge and understanding. While not directly in its Terms of Reference, it has
also made observations about communication and data arrangements as they relate to the
management of water resources and transparency and confidence in decisions made.

191
ACRONYMS

Table 38: Acronyms

Acronym Complete Term
ACT Australian Capital Territory
ADR Australian Design Rules
AIP Aquifer Interference Policy
APCO Australian Packaging Covenant Organisation
ATC Automatic Tube Counts
AWD Available water determination
AWRA-L Australian Water Resource Assessment Landscape model
BLRs Basic Landholder Rights
BOM Bureau of Meteorology
CCA Coca-Cola Amatil
DPE Department of Planning and Environment (now DPIE)
DPI Water Department of Primary Industries Water
DPIE Department of Planning, Industry and Environment
EGS Environmental goods and services
EPA Environmental Protection Authority
ESD Ecologically sustainable development
GAM Generalized Additive Model
GDE's Groundwater dependent ecosystems
GMP Good Manufacturing Practice
HDPE High density polyethylene
HVNL Heavy Vehicle National Law
HVSS Heavy Vehicle Safety Stations
KM Kilometres
LEP Local Environment Plan
LGA Local Government Area
LTAAEL Long Term Average Annual Extraction Limit
ML Mega Litres
ML/y Mega Litres per year
mm/yr millimetres per year
MRFs Materials Recycling Facilities
NAS National Academy of Science
NRAR Natural Resources Access Regulator (NRAR)
NRC Natural Resources Commission
NSW New South Wales
NTC National Transport Commission
NWI National Water Initiative

192
OCSE Office of the Chief Scientist & Engineer
PCT Plant Community Type
PET Polyethylene terephthalate
PEW Planned Environmental Water
RCP Road Contribution Plan
REA Representative Elementary Area
REV Representative Elementary Volume
RMS Roads and Maritime Services
RRE Recharge amount reserved for the environment
SEED Sharing and Enabling Environmental Data
SI Sustainability Index
SILO Scientific Information for Land Owners
SSDs State significant Developments
SSIs State Significant Infrastructure
TAD Total Available Drawdown
TOR Terms of Reference
UEL Upper Extraction Limit
UNSW University of New South Wales
WAL Water Access Licences
WSP Water Sharing Plans

193