UK Gasline
UK Gasline
UK Gasline
system in the UK
Abstract
The UK has an extensive gas system but studies of decarbonisation pathways using the UK MARKAL
least-cost energy system optimisation model have concluded that the natural gas system will be
mostly abandoned by 2050 to achieve a 90% reduction in CO2 emissions in 2050 relative to 1990. In
this study we examine the possible future role of the gas system in the UK energy system. Firstly, we
improve the representation of the gas system in UK MARKAL and add bio-methane and hydrogen
injection as gas decarbonisation options. Secondly, we increase the flexibility of the model with the
addition of biomass CCS technologies and the use of hydrogen in the industrial sector as new
decarbonisation options.
Both bio-methane and hydrogen injection contribute to gas system decarbonisation but are
insufficient alone to avoid the gas system being abandoned in the reference scenario. However, the
addition of biomass CCS to the model, for electricity generation or hydrogen production, allows the
model to remove CO2 from the atmosphere and relaxes emission cuts in other sectors of the
economy. The use of hydrogen to decarbonise industrial processes is also strongly favoured by the
model. With these technologies, the gas system continues to dominate space and water heating
demand to 2050; conversely, if CCS plants prove uneconomic then the gas system will almost
certainly have to be abandoned to achieve the UK emission targets.
1 Introduction
The UK has an extensive natural gas transmission and distribution networks supplying 22.9 million
customers (DECC, 2011). Around half of the annual gas consumption is used for space and water
heating in the residential and service sectors, with the remainder used in electricity generation and
other industrial plants.
The UK Climate Change Act 2008 requires the UK government to reduce UK greenhouse gas
emissions in 2050 by 80% relative to 1990 levels (HM Parliament, 2008). Studies of UK
decarbonisation pathways to meet this target, underpinned by the UK MARKAL energy systems
model, have invariably suggested that the gas system will be mostly decommissioned by 2050 with
heating provided by either electric or biomass boilers (e.g. Hawkes et al., 2011). The government
has yet to adopt a position on the long-term future of the gas system (HM Government, 2011, p.34).
Three options for the gas system can be envisaged under decarbonisation scenarios. First, it is
possible that the gas system will indeed be decommissioned, as previous analysis with UK MARKAL
has tended to suggest. Second, gas could be decarbonised through the injection of bio-methane
and/or hydrogen in the distribution networks. Third, natural gas deliveries to customers in the
current form could continue if carbon reduction targets could be met entirely through
decarbonisation of other sectors of the economy.
The aim of this study was to examine the possible future role of the gas system in the UK energy
system. The objectives were to:
1. reappraise the residual capacity of the gas system and the pipeline investment costs;
2. improve the representation of the gas system in UK MARKAL using the new data;
3. add bio-methane and hydrogen injection to the gas system in UK MARKAL;
4. increase the flexibility of the model to decarbonise other sectors of the economy, with a
focus on biomass and hydrogen processes and the industrial sector; and,
5. assess the individual impacts of each of these changes on the future of the gas system.
This paper is split into two parts. The first part examines the residual capacity and cost of the gas
system and describes our changes to the UK MARKAL model. The second part presents the UK
MARKAL results for the two gas system representations in four decarbonisation scenarios.
First, we reassess the residual capacity of the system that will exist independent of decisions made
by the energy system. The residual capacity in modelling terms is the infrastructure that is built
prior to the model base year that operates during the model time horizon, for which the model has
no investment costs. Unlike previous work with UK MARKAL, we take into account a national
programme that is replacing many of the original iron distribution and service pipes with new plastic
pipes for safety reasons (HSE, 2001). This programme slowed in the late 1990s but was then
accelerated in 2002 following several fatal incidents, with the aim of replacing all iron pipes within
30 m of buildings by 2030 (Cambridge Economic Policy Associates, 2011). Since this programme is
based on health and safety concerns rather than energy system policy, we assume for the purposes
of this work that the programme will continue to 2030 as planned independent of energy system
choices.
Second, we revise the structure and costs of the gas system. In the current version of UK MARKAL,
the gas system is represented as several independent networks each servicing a sector of the
economy. This approach has the disadvantage that increased consumption in a sector always
requires additional capital investment, even if it is matched by reductions in another sector. Our
revised representation disaggregates the system into transmission, distribution and service parts.
The industrial, electricity generation and hydrogen sectors use gas directly from the transmission
network while other sectors are connected to the distribution or service pipes.
Finally, we revise the options for decarbonising delivered natural gas by improving the
representation of bio-methane and hydrogen injection in the model.
2.1 Estimate of the residual capacity of the gas system
Gas has been delivered to households in Britain by pipeline since Victorian times. Town gas, a
mixture of hydrogen, carbon monoxide, methane and other gases, was produced from coal for
household use until the 1970s, when it was replaced by off-shore natural gas (Rooke 1996). A new
transmission grid was built during the 1970s and early 1980s to deliver natural gas from under the
North Sea; UK regions were systematically switched to natural gas and town gas deliveries ceased in
1988 (DECC, 2011).
The national transmission network has a total length of 7600 km (National Grid, 2011a). It was
originally designed to transport gas from the north-east coast of Scotland to the principle demand
areas of England, hence the four southward-bound pipelines leaving St Fergus in Figure 1, but as
indigenous production has reduced in recent years a much greater proportion of the UK supply has
come from The Netherlands or from LNG imports (DECC, 2011).
Gas leaves the transmission network at 53 locations (National Grid, 2011a). Some large power
generation and industrial consumers are supplied directly but most consumers receive low-pressure
gas from one of eight regional mains distribution networks. These networks are much larger than
the transmission network with a total length in 2010 of 275,000 km (Ofgem, 2009, p. 27).
Service pipelines link smaller buildings to the mains distribution networks. They are the narrowest
and shortest pipes in the system but since there are approximately 23 million of them across the
country, they represent a substantial investment that is not borne by large gas consumers.
The highest annual gas delivery through the transmission network since 1998 was 3732 PJ in 2004.
Adding the Milford Haven pipeline, which has a capacity of 20% of the entire network (ICIS Heren,
2007), we assumed a residual capacity of 4439 PJ in UK MARKAL with the entire network operating
to 2050.
8000
Annual pipeline construction (km)
6000
4000
2000
0
1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
Figure 2: Estimated annual pipeline construction rates over 5-year periods from 1890 to 2010. Both
new and replacement pipelines are included. The peak in 1915-1919 is caused by a 50% increase in
gas consumption between 1919 and 1920, presumably the result of a major government-backed
building programme and post-war industrial capacity.
We assume that the current replacement programme will continue to 2030 at the planned rate of
2928 km per annum until there are no more iron pipes near buildings. We then assume that
pipeline replacement will cease unless the model chooses to invest in new capacity. The residual
capacity calculated here does not include any additional capacity after 2014.
To calculate the total energy capacity of the distribution networks, we examine the actual delivered
energy each quarter over the period 1998-2010 in terms of the total length of all of the networks.
The networks satisfied demand throughout the period so the actual capacity must be higher than
the peak supply rate but we conservatively assume here that the redundant capacity is necessary to
ensure a robust and continuous supply and so would be the same relative size in the future. The
highest quarterly throughput as a function of the total mains pipeline length was 7069 GJ/km/a in
2004 (the same year that was used for the current MARKAL residual capacity).
We calculate the residual capacity in energy terms using this throughput. The rate of pipeline
obsolescence is very sensitive to the assumed lifetime of the pipes; older pipes have a greater
likelihood of failure so the lifetime practically depends on the failure tolerance. The 70,000 km of
iron pipes that remain in the system have been in service for between 50 and 100 years (HSE, 2001)
and new stainless steel transmission pipelines are anticipated to have a lifetime of 80 years (National
Grid, 2011b). There is little data to determine the lifetimes of plastic pipes (they have only been
deployed in the UK since 1970); Ofgem assumes a period of 45 years for accounting purposes
(Cambridge Economic Policy Associates, 2011, p25) but it seems reasonable to expect them to last at
least as long as iron pipes so we assume an average lifetime of 80 years.
Figure 2 compares our estimate of the residual capacity of the mains distribution system with the
residual capacity currently in UK MARKAL. There is little difference until 2020 but the data diverge
after then; our analysis suggests that the system will continue to be fully operational in 2050 if the
current accelerated replacement programme continues while the current UK MARKAL
representation assumes that the whole network would be obsolete by then.
2500
2000
Gas capacity (PJ/a)
1500
1000
Current UK MARKAL
Revised mains distribution
Revised service pipelines
500
0
2000 2010 2020 2030 2040 2050
Figure 2: Current UK MARKAL residual capacity of the gas network compared with the revised
capacities of the mains and service pipelines.
We calculate the residual capacity of the service pipes using a similar methodology as for the mains
distribution network. Following Mitchell et al. (1990), we assume an average service pipe length of
11.3 m for all years, in the absence of better information, and calculate the total length of pipes as a
function of the number of customers. This average length depends on the characteristics and spatial
aspects of the buildings that are connected to the mains network over time and could well have
reduced in recent years as many large buildings have been progressively converted to flats in cities.
For the accelerated replacement programme, Cambridge Economic Policy Associates (2011, p49)
assume that 62 service pipes are replaced for each kilometre of mains distribution that is upgraded.
Service pipes are smaller than mains pipes and are more likely to be affected by changes to the build
environment so we assume an average lifetime of 60 years, which is 20 years shorter than the
assumed lifetime of mains distribution pipes.
Figure 1 shows that our revised service pipeline residual capacity is lower than the mains distribution
residual capacity. The rate of service pipeline obsolescence towards 2050 is higher because the
assumed lifetime is lower, but a substantial residual capacity is still available in 2050.
where c is a constant that accounts for the larger size of this pipeline (1.2 m diameter) compared to
the average. Material costs are lower for small pipelines but these only account for 40% of the total
costs, while labour, access and planning costs depend more on the terrain and land use than the size
of the pipe (van der Zwaan et al., 2011). Setting c to 75% and taking the capacity of the system from
the highest recent quarterly throughput (1239 PJ in the first quarter of 2010) gives an investment
cost for transmission pipeline investment of 2.77 £(2000)m/PJ/a. Since the total length of the
network is likely to reduce in the future as gas imports dwarf domestic production, this cost is likely
to be an upper bound.
2.2.2 Investment cost for the mains distribution networks
The capital cost of extending the mains distribution networks depends primarily on the total length
of the networks (and hence the number of customers). Most gas pipes service urban areas where
the building density is likely to be relatively constant across the country, so the spatial dependence
of the mains distribution network is likely to be small.1
The cost per metre of building mains distribution pipes depends primarily on the diameter of the
pipe (Figure 3). Costs for all pipeline sizes are higher in London than elsewhere but it is a reasonable
approximation to represent the whole country with a single representative region. The cost of
building the whole network depends on the proportion of each pipeline size within the network.
This information was not available to us but the length of each pipeline size being replaced in the
period to 2030 is available from Frontier Economics (2011). We estimate the total pipeline length at
each diameter using the assumption that proportion of iron pipes across the network is independent
of the pipeline size. This assumption is likely to underestimate the proportion of larger pipes
because these are less likely to be located within 30 m of a building and are more likely to have been
built since 1970 to connect the regional distribution networks to the national transmission network
that was constructed at that time. Most of these larger pipelines are likely to be operational until
2050 so our assumption is reasonable as long as substantial additional capacity is not added to the
network before 2050.
800
London
700
South East
600 Wales & South West
Pipeline cost (£/m)
East of England
500 West Midlands
North West
400
North East & Yorkshire
300 Scotland
200
100
0
<3 4 to 5 6 to 7 8 to 9 10 to 12 to 18 to over 24
12 18 24
Pipeline diameter (inches)
Figure 3: Cost of replacing pipelines in each UK mains distribution network as a function of the
pipeline diameter (data from Frontier Economics, 2011).
1
We can test this assumption by comparing the number of customers in each region with the length of
pipelines being replaced. The lowest replacement rate is in London, where the housing density is higher than
elsewhere in the country. The highest rate is in the South East, which has many rural areas with particularly
old pipelines that require replacing. In the other regions, the replacement rate varies in the range 2.8–3.7
km/1000 customers. If the gas networks in each of these regions have similar age distributions then the
assumption of little spatial dependence is reasonable for the purposes of this calculation.
We derive a mains replacement costs of 0.141 £(2008)m/km from Cambridge Economic Policy
Associates (2011, p51) and 0.117 £(2008)m/km from Frontier Economics (2011). Taking into account
the nature of the replacement (some new pipes are smaller than existing pipes and in some cases
are even fitted into existing pipes) and the higher costs of building new capacity than replacing
existing capacity, we use the higher of these values. A further consideration is that neither of these
costs includes overheads of approximately 0.034 £(2008)m/km (see the following section). Adding
this to mains replacement cost, we calculate the MARKAL investment costs using:
where the 0.79 factor is an inflation adjustment factor to the year 2000.
The service pipe replacement costs are estimated by Frontier Economics (2011) to be 44% of the
mains distribution replacement costs. Using their data, 55347 km of service pipes will be replaced
by 2032 at a cost of £4056m (excluding overheads). This is equivalent to a cost of 0.073 £m/km.
The two costs are substantially different because only one includes overheads. Overheads should be
included in the investment costs because they are an integral part of the upgrade programme.
Assuming that overheads account for the entire difference, the total overhead cost is 0.068 £m/km
or 0.034 £m/km each for the mains and the service pipes, assuming an equal split. The service pipe
investment cost is then:
The service pipe costs only include connections to the gas meter. Gas pipes beyond the gas meter
are the responsibility of the property owner but could represent a substantial additional cost when
adding gas connections to new properties. These costs are not currently considered in UK MARKAL
but could be evaluated in the future.
The owner of much of the system, National Grid, identifies bio-methane injection as a long-term
decarbonisation strategy to allow continued use of the system (National Grid, 2009). Bio-methane is
produced by upgrading biogas, a product of biomass gasification and anaerobic digestion plants, so
is considered to be carbon neutral.
Bio-methane injection is modelled as a sensitivity study option by Usher & Strachan (2010) and
Hawkes et al. (2011). In both cases, bio-methane is assumed to use a different pipeline system to
natural gas so it is necessary to invest in new pipeline capacity whenever bio-methane capacity is
increased. Our revised gas system representation assumes that bio-methane is injected directly into
the existing distribution mains networks. We use the same costs for upgrading biogas to bio-
methane as Usher & Strachan (2010).
Another gas decarbonisation option is to inject small amounts of hydrogen into the distribution
networks. The level of hydrogen that can be safely added depends on the characteristics of the
existing natural gas in the system and on the design of existing appliances (NaturalHy, 2010); for
example, appliances have been successfully tested with 18-50% hydrogen by volume. In the
Netherlands, a recently concluded four-year field trial used hydrogen and natural gas blends (of up
to 20% hydrogen by volume) in off-the-shelf gas appliances and highlighted no serious problems in
operation (Kippers et al., 2011).
For hydrogen injection in UK MARKAL we assume that the hydrogen volume cannot exceed 20% of
the total gas volume in line with a recent field trial in the Netherlands (Kippers et al., 2011). Since
the volumetric energy density of hydrogen is substantially lower than that of natural gas, the
maximum hydrogen injection in our study represents only 6% of the energy content of the blended
gas. Hydrogen injection is only possible for runs using the new gas system representation in our
study.
3 Methods
MARKAL is a widely-applied bottom-up, dynamic, linear programming optimisation model (Loulou et
al., 2004). The UK MARKAL model (Kannan et al., 2007, Anandarajah and Strachan, 2010) has been
developed over the last decade and portrays the entire UK energy system from imports and
domestic production of fuel resources, through fuel processing and supply, explicit representation of
infrastructures, conversion of fuels to secondary energy carriers (including electricity, heat and
hydrogen), end-use technologies and energy service demands of the entire economy. It is calibrated
to the UK energy consumption in the year 2000. The initial energy service demands to 2050 are fully
described in Usher and Strachan (2012).
We use the MARKAL elastic demand variant in this study in which welfare (defined as the sum of
producer and consumer surplus) is maximised. Behavioural change in response to increasing energy
costs is simulated using reductions in the energy service demands.
Carbon dioxide is the only greenhouse gas simulated in UK MARKAL. In both Usher & Strachan
(2010) and Hawkes et al. (2011), the 80% emissions reduction target in 2050 is represented by a 90%
reduction in CO2 in the model. This additional effort recognises the uncertainties in the contribution
of non-CO2 GHGs, the emissions from land-use change and emissions from international bunker fuels
(Usher and Strachan, 2010). It is currently uncertain whether emissions from international aviation
and shipping should be included in the 90% target; Usher & Strachan (2010) include these while
Hawkes et al. (2011) do not. These sectors are not included in this study.
3.1 Scenarios
We test the old and new gas system representations in UK MARKAL using four scenarios (Table 2).
We start with the most recent ‘standard’ version of the model, UK MARKAL v3-26-5, and then
progressively implement new technologies and constraints in subsequent scenarios. We perform
each scenario with the old and new gas system representations. For each of these eight cases we
produce a reference case (no decarbonisation) and an elastic demand case for a 90% emissions
reduction in 2050 relative to 1990. The reference case prices are used for the elastic demand
calculations in each scenario. In total there are 16 simulations.
Scenario Description
DECC UK MARKAL v3-26-5 used for the DECC decarbonisation pathways study (AEA 2010),
with bio-methane production technologies from Usher & Strachan (2010)
BIOCCS The DECC scenario with revised biomass CCS electricity generation technologies
HYIND The BIOCCS scenario with a revised hydrogen production and delivery sector, including
biomass CCS for hydrogen generation, and the use of hydrogen in industrial processes
NOCCS The HYIND scenario but with all CCS technologies disabled
Table 2: UK MARKAL scenarios examined in this study.
Hawkes et al. (2011) add several additional model constraints in the DECC study with the assumption
that all current short-term government decarbonisation targets will be met; these constraints are
not included in our simulations.
A potentially important use of biomass that has not previously been available in UK MARKAL is the
production of hydrogen in a CCS plant. In this scenario we revamp the hydrogen production sector
with updated technology data for SMR, coal and biomass gasification technologies, with and without
CCS. We also add a flexible coal integrated gasification combined cycle (IGCC) technology and
update the electrolysis technologies. These changes are fully described in Dodds and McDowall
(2011b).
We also revamp the representation of hydrogen delivery technologies. For gaseous hydrogen, we
introduce a separate transmission/distribution pipeline system similar to that outlined above for the
natural gas system. We update the liquefaction and road tanker technologies to properly examine
alternative methods of hydrogen delivery. These changes are fully described in Dodds and
McDowall (2011a).
The representation of the industrial sector in UK MARKAL is less detailed than most other sectors
and there are few options available for model decarbonisation. This partly reflects the diversity of
the sector and the relative paucity of low-cost options for decarbonisation. Previous studies have
identified industrial processes where energy efficiency options might be effective (e.g. Hodges and
Hawkes, 2010) but few have considered the potential for hydrogen, produced from electrolysis or
from CCS plants, as a clean energy carrier to replace carbon-intensive fuels. In this scenario we
extend the UK MARKAL industrial sector to use gaseous hydrogen for drying, low temperature heat
and other processes where a range of fuels are acceptable.
4 Results
The results are presented in three parts. First, we look at how the change to the gas system
representation in UK MARKAL affects each scenario. Second, we examine how the gas system is
affected by the changes to the model in each scenario. Third, we examine the prospects for
decarbonising the gas system through bio-methane and hydrogen injection.
Since CCS is never used in reference cases because of the additional costs, the DECC and BIOCCS
reference cases are identical, as are the HYIND and NOCCS reference cases. These two reference
pairs are each presented as a single line in all of the graphs to aid visual interpretation.
20%
Change in UK gas consumption
Figure 5: Simulated change to the UK gas consumption in each scenario caused by updating the
model gas system representation. Solid lines show the reference cases (no emissions limit). Dotted
lines represent a 90% reduction in CO2 emissions in each scenario.
4000
3500
Total UK gas consumption (PJ)
0
2000 2010 2020 2030 2040 2050
Figure 6: Total simulated UK gas consumption to 2050 using the new gas system representation.
The residential and service sector gas consumption for space and water heating in each scenario are
shown in Figure 7. Gas continues to be the dominant fuel source to 2050 in the reference cases.
The decarbonisation runs show a strong dichotomy. For the DECC and NOCCS scenarios, gas is not
used for space and water heating after about 2040 as the carbon emissions are too high. In the
BIOCCS and HYIND scenarios, the availability of biomass CCS plants relaxes the tight carbon emission
constraints on the model and gas remains the optimum fuel for heating. The mix of fuels used for
heating in 2050 is relatively unchanged in the latter scenarios as shown in Figure 8. For the DECC
and NOCCS scenarios, however, the optimum use of biomass when biomass CCS plants are not
available is for residential and service sector heating. Electricity supplies about 30% of heating in the
DECC scenario but is less important in the other scenarios.
1800
Gas consumption for heating in the
residential and service sectors (PJ)
1600
1400
DECC & BIOCCS REF
1200
HYIND & NOCCS REF
1000
DECC 90% CO2
800
BIOCCS 90% CO2
600
HYIND 90% CO2
400
NOCCS 90% CO2
200
0
2000 2010 2020 2030 2040 2050
Figure 7: Simulated gas consumption for space and water heating in the residential and service
sectors using the new gas system representation.
100%
Residential and service sector heat mix
80%
Solar
District heating
60%
Solid/wood fuel
20% Electricity
Gas
0%
DECC & HYIND & DECC 90% BIOCCS HYIND 90% NOCCS 90%
BIOCCS REF NOCCS REF 90%
Figure 8: Fuels used for space and water heating in the residential and service sectors in 2050 in each
model scenario.
We can better understand the gas system trends in each scenario by examining the electricity
generation mix in 2050 (Figure 9). Coal dominates generation in the reference scenarios although
some gas is also used in the HYIND & NOCCS REF case because coal replaces gas for hydrogen
production (not shown). Nuclear power and renewables are the dominant generation technologies
in all four decarbonisation scenarios. In the DECC scenario, cofiring CCS and biomass CCS are used to
sequester small amounts of CO2 from the atmosphere. In the BIOCCS scenario, where the biomass
CCS technologies can sequester a much greater proportion of the biomass carbon, the model invests
heavily in biomass CCS up to the resource supply limit, which lead to strong ‘n gat v m ss ons’ n
the electricity sector that reduce emission constraints in other sectors and permit the continued use
of gas for heating. Biomass CCS plants are available but are not selected in the HYIND scenario, with
coal CCS plants being deployed instead. This reflects the influence of the new hydrogen scenario
and the use of hydrogen in industry; biomass is more efficiently used for hydrogen production than
for electricity generation and the additional hydrogen is used to decarbonise much of the industrial
sector as shown in Figure 10. Hydrogen production starts 15 years earlier and is four times higher in
2050 in the HYIND scenario than in the BIOCCS scenario. Approximately 90% of this additional
production is used in the industrial sector.
2500
Other renewable
Electricity generation mix (PJ)
The second metric is the marginal cost of CO2 removal from the energy system, which is shown for
the four scenarios in Figure 11 for the elastic demand variant. Higher values indicate that the
scenario is more constrained by the emissions reduction. The DECC and NOCCS scenarios have high
marginal CO2 costs by 2050 of £300–£500. Conversely, the other two scenarios have very similar
and much lower marginal costs of around £110 in 2050, despite having evolved having quite
different energy systems by then.
1600
Hydrogen
1400 Renewable (waste)
Steam
Industrial energy use (PJ)
1200
Liquid Petroleum Gas
1000
Light Fuel Oil / Bio-oil
800 Heavy Fuel Oil
Blast Furnace Gas
600
Coke Oven Gas
400 Coke
200 Coal
Electricity
0
2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 Natural Gas / Biomethane
Figure 10: Industrial fuel use in the HYIND scenario for a 90% reduction in CO2 emissions.
600
500
Marginal cost of CO2 (£/te)
400
DECC 90% CO2
100
0
2000 2010 2020 2030 2040 2050
Figure 11: Marginal cost of reducing carbon emissions in each scenario (£/te).
Hydrogen injection is constrained in the model to 6% of the total supplied energy so has a more
minor role than bio-methane injection. The maximum amount of hydrogen injection is adopted by
the model from 2040 in the DECC scenario and 2050 in the NOCCS scenario. Only a small amount is
injected in the BIOCCS scenario and none at all is used in the HYIND scenario, where the industrial
sector is a lower-cost alternative.
70%
Biomethane fraction of the total UK gas
60%
50%
DECC 90% CO2
40%
supply
10%
0%
2000 2010 2020 2030 2040 2050
Figure 12: Bio-methane content of the total UK gas supply in each model scenario.
5 Discussion
Our reappraisal of the UK gas system identifies two inaccuracies in the UK MARKAL model. First, the
model currently assumes that the whole network will become obsolete by 2040 but our analysis
shows that the majority of the existing network will still be operational in 2050, if the safety-related
iron pipeline replacement programme continues to 2032 as planned. Second, the cost of replacing
or building new pipelines is three times higher than currently assumed in the model.
Our revised representation of the gas system in UK MARKAL fixes these inaccuracies and introduces
a more accurate structure that better represents the flexibility of the network. Injection of bio-
methane and hydrogen to the distribution networks is also possible in the new structure. Despite
the breadth of these changes, they cause only relatively minor variations in the utilisation of the gas
system in all of the scenarios, particularly when a 90% reduction of CO2 emissions in 2050 is
required. One reason is that the additional cost of replacing the gas system due to the unnecessarily
early retirement in the current model is partially offset by exceptionally low investment costs for
replacement capacity. Our concern when commencing this study was that the model might be
abandoning the gas system as a result of inaccuracies in the representation of the system but this
proved not to be the case.
The gas system has a long lifetime (for example, some distribution mains pipes are now more than
100 years old). UK MARKAL only simulates the UK energy system to 2050 as this is the current target
date for UK long-term emissions targets. While most of the current network will still be operational
in 2050, the majority will require replacing by 2100. The long-term role of the gas system in the UK
energy system could be better assessed in the future by extending the model horizon to 2100.
The crucial change in the BIOCCS and HYIND scenarios is the availability of biomass CCS plants which
effectively remove CO2 from the atmosphere to underground storage. These substantially reduce
the emission constraints in the other sectors and also substantially reduce the marginal carbon price.
No large-scale biomass CCS plants have ever been constructed, for electricity or hydrogen
production, and more research is required to better understand the costs and efficiencies of these
plants. The fraction of the biomass carbon that can be sequestered is likely to be a crucial
determinant of the efficacy of these plants. It will also be important to include the full lifecycle costs
and emissions of biomass in future modelling studies as these tend to be higher than those of fossil
fuels because the energy density of biomass is much lower and because land use change can be an
important factor.
The use of hydrogen in the industrial sector has received little attention in the literature but the
take-up is very high from 2030 in the HYIND scenario because there are few other decarbonisation
options for this sector. The uptake here must be regarded as somewhat speculative because of the
aggregated nature of the industrial sector representation in UK MARKAL. One of the impediments to
the use of hydrogen in the transport sector is the lack of existing delivery infrastructure; liquid
hydrogen is expensive to produce while a gaseous hydrogen pipeline system is very expensive to
build when the consumption is very low in the early stages (Strachan et al., 2008). If hydrogen was
adopted in the industrial and transport sectors at the same time then the total demand would be
much higher and the delivery costs would reduce. This strategy deserves further attention in the
future.
Hydrogen injection is adopted in all but the HYIND scenario, where the industrial sector is the
preferred destination. However, the quantities of hydrogen are very small, partly because the
distribution networks are abandoned in two scenarios, so we believe that hydrogen injection will not
have a major impact on UK decarbonisation strategy but might make a small contribution. Hydrogen
injection potentially offers an additional benefit not captured by MARKAL of facilitating load levelling
of the electricity network if there is a large deployment of wind turbines in the future.
One option not examined here is the conversion of the natural gas distribution networks to deliver
100% hydrogen. The hydrogen would be either combusted in a boiler or used to fuel stationary
micro-CHP units in buildings. This is a complex option with many technical and economic issues that
we plan to address in a future study.
6 Conclusions
The aim of this study was to examine the possible future role of the gas system in the UK energy
system. Previous studies (e.g. Hawkes et al., 2011) using the UK MARKAL model conclude that the
gas system will be abandoned by 2050. Our improvements to the gas system representation in UK
MARKAL do not alter this conclusion for those scenarios but we do find two mitigating factors. The
first is to partially decarbonise the gas system through bio-methane and hydrogen injection.
However, these technologies are insufficient on their own to avoid the gas system being abandoned.
The second factor is the potential for higher emission reductions in other sectors to offset emissions
from gas combustion. Our scenarios examine a number of options in the biomass, hydrogen and
industrial sectors and identify biomass CCS as a crucial technology, despite it receiving little
attention to date in the literature. The use of hydrogen in industrial processes is also strongly
favoured by the model and should be examined further. Deployment of these technologies enables
the gas system to continue supplying 80% of space and water heating demand to 2050.
7 References
ANANDARAJAH, G. & STRACHAN, N. 2010. Interactions and implications of renewable and climate
change policy on UK energy scenarios. Energy Policy, 38, 6724-6735.
CAMBRIDGE ECONOMIC POLICY ASSOCIATES 2011. 10 year review of the Iron Mains Replacement
Programme. Cambridge, UK: HSE/Ofgem.
DECC 2011. Digest of United Kingdom Energy Statistics 2011. London, UK: Department of Energy and
Climate Change.
DODDS, P. E. & MCDOWALL, W. 2011a. A review of hydrogen delivery technologies for energy
system models. UKSHEC Working Paper. London, UK: UCL.
DODDS, P. E. & MCDOWALL, W. 2011b. A review of hydrogen production technologies for energy
system models. UKSHEC Working Paper. London, UK: UCL.
FRONTIER ECONOMICS. 2011. Evaluating the gas mains replacement programme – preliminary
findings [Online]. London, UK. Available: http://www.ofgem.gov.uk/Networks/GasDistr/RIIO-
GD1/WorkingGroups/Documents1/Frontier%20repex.pdf [Accessed].
HAWKES, A., NARKEVICIUTE, R., MORRIS, S. & SOLANO-RODRIGUEZ, B. 2011. Pathways to 2050 –
Detailed Analyses. London, UK: AEA Technology plc.
HM GOVERNMENT 2011. The Carbon Plan: delivering our low carbon future. In: DECC (ed.). London,
UK.
HM PARLIAMENT 2008. The Climate Change Act 2008. In: DECC (ed.). London, UK.
HODGES, R. & HAWKES, A. 2010. Analysing the Opportunities for Abatement in Major Emitting
Industrial Sectors. Harwell, UK: AEA Technology plc.
HSE 2001. Th H alth and Saf ty Ex cut v ’s nforc m nt ol cy for th r lac m nt of ron gas
mains. London, UK: HSE.
ICIS HEREN. 2007. UK energy minister opens Milford Haven gas pipeline [Online]. Available:
http://www.icis.com/heren/articles/2007/11/30/9300867/uk-energy-minister-opens-milford-haven-
gas-pipeline.html [Accessed 1 Nov 2011].
KANNAN, R., STRACHAN, N., PYE, S., ANANDARAJAH, G. & BALTA-OZKAN, N. 2007. UK MARKAL
Model Documentation.
KIPPERS, M. J., DE LAAT, J. C., HERMKENS, R. J. M., OVERDIEP, J. J., VAN DER MOLEN, A., VAN ERP, W.
C. & VAN DER MEER, A. 2011. Pilot Project on Hydrogen Injection in Natural Gas on Island of
Ameland in the Netherlands. International Gas Union Research Conference 2011. Seoul, South Korea.
LOULOU, R., GOLDSTEIN, G. & NOBLE, K. 2004. Documentation for the MARKAL Family of Models.
ETSAP, IEA.
MITCHELL, C., SWEET, J. & JACKSON, T. 1990. A study of leakage from the UK natural gas distribution
system. Energy Policy, 18, 809-818.
NATIONAL GRID 2009. The potential for Renewable Gas in the UK. National Grid.
NATIONAL GRID 2011a. Annual Report and Accounts 2010/11. London, UK: National Grid plc.
NATIONAL GRID. 2011b. Milford Haven Gas Connection Projects [Online]. London, UK: National Grid.
Available: http://www.nationalgrid.com/uk/Gas/Pipelines/milfordhaven [Accessed].
NATIONAL GRID. 2012. UK electricity and gas transmission [Online]. London, UK. Available:
http://www.nationalgrid.com/annualreports/2006/05_opfinrev/ukelecgastransmission.html
[Accessed 30 Jan 2012].
NATURALHY 2010. Final publishable activity report. Preparing for the hydrogen economy by using the
existing natural gas system as a catalyst. Groningen, The Netherlands: N.V. Nederlandse Gasunie.
OFGEM 2009. Gas distribution annual report for 2007-2008. London, UK: Office of Gas and Electricity
Markets, Ofgem.
STRACHAN, N., BALTA-OZKAN, N., KANNAN, R., HUGHES, N., MCGEEVOR, K. & JOFFE, D. 2008. State-
of-the-art modelling of hydrogen infrastructure development for the UK. London, UK: Policy Studies
Institute.
US EPA 2005. US Regional MARKAL model documentation. Office of Research and Development, US
Environmental Protection Agency.
VAN DER ZWAAN, B. C. C., SCHOOTS, K., RIVERA-TINOCO, R. & VERBONG, G. P. J. 2011. The cost of
pipelining climate change mitigation: An overview of the economics of CH4, CO2 and H2
transportation. Applied Energy, 88, 3821-3831.