Environmental Toxicology and Chemistry, Vol. 27, No. 5, pp.

11351140, 2008
2008 SETAC
Printed in the USA
0730-7268/08 $12.00 .00

RESPONSE OF A NEW ZEALAND MAYFLY (DELEATIDIUM SPP.) TO ACID MINE

DRAINAGE: IMPLICATIONS FOR MINE REMEDIATION

KATHRYN OHALLORAN,* JO-ANNE CAVANAGH, and JON S. HARDING

Landcare Research, P.O. Box 40, Lincoln 7640, New Zealand
School of Biological Sciences, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand

( Received 11 March 2007; Accepted 19 November 2007)

AbstractInvestigating the toxicity of acid mine drainage (AMD) on benthic communities in receiving waters can be highly
challenging because of the difficulty in unraveling the effects of acidity, dissolved metals, and precipitates. Furthermore, the survival
of different species may vary depending on any natural adaptation they may have acquired to low pH, metals, or sedimentation.
We investigated the effect of different pHs and AMD on the survival of a common New Zealand leptophlebiid mayfly (Deleatidium
spp.) in 96-h laboratory trials. Our results indicate that the primary driver of toxicity in AMD was pH, although some mortality
could be attributable to the presence of dissolved heavy metals at low pH (3.6). Mayflies sourced from three naturally acidic
streams (pH 5.76.5) had a distinctly higher tolerance to AMD and low pH (3.54.0) compared to mayflies sourced from three
circumneutral streams (pH 7.07.4). This indicates that the chemistry of the natal stream strongly influences the sensitivity of
mayflies to AMD, which, in turn, could have consequences for the successful remediation of a given AMD-impacted stream.
Furthermore, the water chemistry of unimpacted streams that could be sources of potential recolonists might provide ecologically
relevant water-quality targets for remediation of AMD-damaged streams. Understanding the variable tolerances of common lotic
benthic taxa can provide ecologically relevant water-quality criteria for mine remediation.

KeywordsBenthic invertebrates Acid mine drainage Toxicity Remediation

INTRODUCTION streams of naturally low pH also may tolerate anthropogeni-

Acid mine drainage (AMD) is a long-term, global issue cally acidic waters better than the same taxa in streams of
affecting almost all countries with abandoned or active coal higher natural pH [5,7,11,12]. This variability in tolerances of
mines [1,2]. Acid mine drainage is produced when water mixes the same taxa creates complexity in determining what consti-
with mineral deposits containing sulfides, such as iron disulfide tutes successful remediation of an AMD-impacted stream and,
(FeS2, pyrite) in an oxidizing environment. Sulfuric acid is where new mining operations are being considered, establish-
formed, lowering the pH of receiving waters while reacting ing appropriate water-quality criteria to avoid significantly im-
with elevated dissolved metal concentrations (predominantly pacting the existing ecosystem.
iron and aluminum ions). Changes in pH, which might result The aim of the present study was to evaluate the survival
from dilution of waters from tributary streams or heavy rain- of a common indigenous freshwater invertebrate to a range of
fall, can cause the precipitation of metals (e.g., iron hydroxide). mine-impacted waters in New Zealand. Specifically, we used
Metal precipitate smothers the streambed substrate, reducing the leptophlebiid mayfly (Deleatidium spp.) as a representative
the habitat for stream fauna [3]. Acid mine drainage frequently organism to evaluate the effect of dilution and pH modification
is persistent and extreme in its effects on water chemistry and on AMD toxicity, to delineate pH tolerance in the presence
biotic communities [4]. Toxicity can arise from the increased and absence of AMD-associated metals, and to investigate the
acidity as well as from the presence of soluble and particulate influence of natal stream chemistry on sensitivity to AMD.
metals. The interplay between pH and metals in solution makes
MATERIALS AND METHODS
these systems complex to investigate and manage. Conse-
quently, environmental agencies dealing with the mining sector Test organism
currently rely on generic water-quality criteria to predict the We selected the leptophlebiid mayfly (Deleatidium spp.
potential toxicity of AMD on stream communities. These ge- [Ephemoptera: Leptophlebiidae]) as a representative organism
neric criteria usually do not account for any specialized or for the evaluation of AMD tolerances. Although not widely
adaptive characteristics of species inhabiting the receiving- used as a toxicity-testing organism, this mayfly has proved to
water environments of concern. This aspect is of particular be useful in other New Zealand studies [13,14]. It is endemic
interest in New Zealand, because naturally acidic streams, usu- to New Zealand and is among the most abundant taxa collected
ally with a pH of approximately 4 [5], frequently occur on the in New Zealand streams [15,16], thus allowing large numbers
west coast of the South Island. to be collected as required for use in replicated toxicity tests.
In New Zealand, many freshwater benthic invertebrates Ephemoptera, which have a reported sensitivity to acidic con-
have highly variable tolerances to pH and AMD [511]. Fur- ditions [17], commonly are used in biotic indices of stream
thermore, several studies have indicated that organisms in condition, including receiving waters in the vicinity of mines
* To whom correspondence may be addressed [18,19]. Frequently, it is one of the earliest colonizers of
(ohallorank@landcareresearch.co.nz). streams [20] and could be an important indicator of ecosystem
Published on the Web 1/4/2008. recovery in remediated mine streams.
1135
1136 Environ. Toxicol. Chem. 27, 2008 K. OHalloran et al.

Midinstar nymphs of Deleatidium spp. were collected using every 24 h (i.e., five times). Experiments were run for 96 h
a hand-net and transported in the source stream water at 10 in a environment-controlled room set to simulate summer
to 15C in insulated, aerated containers to the laboratory. When stream conditions at 15C with a 16:8-h light:dark photoperiod.
possible, organisms were collected within 24 h of the start of
each experiment. Nymphs with dark wing pads were excluded Chemical analysis
to reduce the possibility of nymphs emerging as adults during Water samples were analyzed for heavy metals (aluminum,
the experiments. arsenics, iron, manganese, nickel, and zinc), pH, and electrical
The AMD dilution and pH manipulation experiments were conductivity by R.J. Hills Laboratories (Hamilton, New Zea-
performed on nymphs collected from a single, naturally acidic land). Total and dissolved metals were filtered and analyzed
stream (Carton Creek [17150.666E, 4208.526S], Reefton, by digesting samples through boiling in nitric acid and then
New Zealand) with a background pH of 5.7. Variations in analyzed using inductively coupled plasmamass spectropho-
tolerance to AMD also were examined in mayflies sourced tometry according to American Public Health Association
from five additional New Zealand streams (Cust River [172 method 3125 [21]. Method detection limits were as follows:
37.609E, 4322.327S], Donegals Stream [17114.727E, Iron, 0.02 g/L; aluminum, 0.003 g/L; arsenic and zinc, 0.001
4243.240S], Soldiers Creek [17151.126E, 427.974S], g/L; and manganese and nickel, 0.0005 g/L. Electrical con-
Lankey Creek [17154.057E, 428.866S], and Otira Spring ductivity and pH were measured according to American Public
[17133.865E, 4249.006S]). Health Association methods 2510 and 4500, respectively [21].

Test water Statistical analysis

All AMD water samples used in the present study were The proportion of mayfly nymphs surviving after 96 h were
sourced from West Sullivan Mine on the Denniston Plateau square-root arcsin transformed to normalize error variance
north of Westport, on the west coast of South Island, New [22]. Data were analyzed using the statistical package
Zealand. Natural stream water upstream of the mine discharge S-PLUS (Ver 6.2; Insightful, Seattle, WA, USA). Differences
(pH 4.6) was used for dilution water and control treatments. between treatments after 96 h of exposure were assessed using
To simulate the effect of simple remediation measures, such one-way analysis of variance, with a significance level of p
as dilution and neutralization, AMD water collected from the 0.05. Tukeys multiple-range post hoc test was used to de-
mine discharge point was manipulated. Initial pH of the AMD termine when significant differences occurred.
was recorded (Mettler Toledo SevenEasy pH meter; Mettler The influence of AMD metals on the proportion of mayflies
Toledo International, Schwerzenbach, Switzerland) before surviving at various pH was analyzed using binomial regres-
making a series of dilutions (i.e., 0.5, 0.25, 0.125, and sion analysis of pH versus the transformed survival data in
0.0625) to simulate natural dilution of AMD waters. A du- AMD and non-AMD water [23]. The difference between the
plicate dilution series was then modified with alkali (NaOH) slopes of the two lines was compared using analysis of de-
to neutralize the pH up to background levels (pH 4.6). Samples viance (Genstat 8th ed; VSN International, Hempstead, UK)
were filtered through a 0.45-M membrane filter to remove at a significance level of p 0.05.
precipitated material and then assessed for toxicity, pH, con- To evaluate the importance of their natal stream water
ductivity, and total and dissolved metals. chemistry, the proportion of mayflies surviving from either
To differentiate between metal and pH toxicity, additional acidic or circumneutral streams was modeled with a gener-
tests were conducted using uncontaminated water that had been alized linear model using binomial errors (Genstat). Data col-
acidified (with HCl) to match the pH of the unmodified dilution lected at the lowest pH were excluded, because no animals
series described above. survived at this pH.
Variations in AMD tolerance were then examined in may-
flies sourced from three circumneutral and three naturally acid- RESULTS
ic streams. The mayflies were exposed to AMD dilutions that Both undiluted (pH 3.0) and 0.5 AMD (pH 3.2) resulted
gave low pH (i.e., 3.3), medium pH (i.e., 3.5), and higher pH in 100% mortality in mayflies sourced from streams of natu-
(i.e., 4.0). The pH varied between 3.2 and 3.3 in the low-pH rally low pH, but with each subsequent dilution, mayfly sur-
samples on different experimental occasions. Therefore, it is vival increased (Fig. 1A). Significantly greater survival oc-
reported as a range in figure legends. These dilutions repre- curred with dilutions of fourfold or greater (F5,42 72.8, p
sented the range of pH found above the AMD discharge point 0.0001), which corresponds to pH 3.5 (Fig. 1A and Table
and in the receiving waters of Sullivan Mine. No mortality 1). Mayfly survival of 80% or greater was observed in 0.125
occurred in the source water (i.e., the water from where the AMD or lower (i.e., pH 3.8) (Table 1). Dilution also mark-
mayflies were collected) controls (data not shown). edly reduced metal concentrations, with iron and aluminum
effectively reduced by 50% with each dilution (Table 1) and
Toxicity testing remaining largely in a dissolved form despite the progressive
Toxicity tests were conducted in the laboratory on active increase in pH up to background levels (data not shown).
mayfly nymphs (n 5). Nymphs were randomly assigned to Following modification to pH 4.6, most of the iron and up
replicate plastic, 200-ml containers (n 8 unless otherwise to 39% of the aluminum in the AMD was precipitated. Mayfly
stated), each with 100 ml of treatment water. Air was supplied survival was high (77%) in all pH-modified AMD dilutions,
to the containers from an electric aerator for 15 min twice and no significant difference was found between any of the
daily. Mayflies were checked every 24 h for mortality and AMD samples or the control (F5,42 1.39, p 0.25) (Fig.
evidence of molting. Individuals lacking vitality and indepen- 1B).
dent mobility were considered to be dead, and dead mayflies Other metals analyzed (manganese, arsenic, nickel, and
and exoskeletons were removed daily. During the experiment, zinc) existed as dissolved ions in both unmodified and pH-
the pH of the test samples (n 8 per treatment) was recorded modified samples, indicating that these metals are most likely
Mayfly response to acid mine drainage Environ. Toxicol. Chem. 27, 2008 1137

2). Survival at pH 3.3 (90%) was not significantly different

from survival in control water (Fig. 2). Furthermore, the slopes
of the lines derived from pH versus survival data at 96 h for
mayflies exposed to AMD and non-AMD water were signif-
icantly different (F1,116 14.90, p 0.001). An increased mor-
tality in the AMD-contaminated samples occurred at pH 3.6
(p 0.05), indicating higher toxicity in the metal-rich AMD
water at these lower pH levels.
Preadaptation to low pH based on water chemistry in natal
streams was evaluated in mayflies sourced from three circum-
neutral and three naturally acidic streams (Fig. 3). A significant
difference in the response to different AMD dilutions of may-
flies sourced from different streams was observed (F4,78 5.62,
p 0.001). These differences were explained by significantly
lower survival in AMD water (F1,78 188.22, p 0.001)
observed in mayflies sourced from circumneutral streams (Fig.
3AC) compared to mayflies from streams with naturally low
pH (Fig. 3DF). After the 96-h exposure period, mayflies
sourced from circumneutral streams showed 100% mortality
at pH 3.5. At pH 4.0, markedly higher numbers survived,
although mortality was still considerable (between 62.5 and
82.5%) (Fig. 3AC), whereas mayflies sourced from naturally
acidic streams had a higher tolerance to AMD-contaminated
water. In these trials, mayfly mortality was still 100% at pH
3.3, but survival improved in all treatments with pH 3.5,
with no mortality occurring at pH 4.0 (Fig. 3DF).

DISCUSSION
Mayfly survival improved with progressive dilutions of
AMD and mayflies were able to tolerate 96-h exposures to
AMD diluted to less than 0.125 (i.e., at pH 3.8), with no
statistically significant mortality occurring. Modifying to pH
4.6 (equivalent to the background pH of naturally acidic
streams) abrogates the toxicity observed in the unmodified
AMD samples, despite the continued presence of metals in
solution. Our findings indicate that pH probably is the main
factor causing mortality in these mayflies. Low pH may hinder
osmoregulatory processes in these animals, disturbing the cel-
lular ionic balance. In particular, significant losses of sodium,
calcium, potassium, and chloride ions may occur [24].
Evaluating the role of metal ions in AMD toxicity is com-
plex at lower pHs where dissolved metals may exist. The re-
sults of the present study show that mayflies can tolerate lower
Fig. 1. Survival of mayflies (sourced from stream with low pH) in
(A) differing dilutions of unmodified acid mine drainage (AMD) and pH if the water is free of metals, suggesting that dissolved
(B) AMD dilutions where pH was modified to background level (pH metals contribute additional toxicity at lower pH. Elevated iron
4.6). Values are presented as the mean standard error (n 8 rep- and aluminum concentrations are common in AMD [25], but
licates). Letters denote significantly different groups at 96 h. information on the toxicity of dissolved iron to freshwater
undiluted AMD; 0.5 AMD; 0.25 AMD; 0.125
AMD; 0.062 AMD; control water. invertebrates is limited. Water-quality criteria for the protec-
tion of aquatic life range from 0.3 to 1.5 mg/L of iron [26
29] (http://www.ccme.ca/publications/ceqgrcqe.html and
to exist in a soluble form in both diluted and undiluted AMD http://www.environment-agency.gov.uk/). Although iron is not
samples. considered to be an overtly toxic or priority pollutant, these
Neutralization of AMD reduces its acute toxicity (Fig. 1B), criteria are still below the concentration of iron in the AMD
suggesting that low pH is driving toxicity in the unmodified samples that we tested, in which concentrations of dissolved
AMD samples (Fig. 1A). Modification of pH, however, also iron of up to 2.29 mg/L had no adverse influence on the sur-
causes precipitation of iron and aluminum; therefore, the con- vival of Deleatidium spp. Some suggest that the presence of
tribution of these metal ions to the toxicity observed in the iron can reduce the toxicity of other metals to aquatic inver-
unmodified samples cannot be dismissed entirely. Mayflies tebrates. Soucek et al. suggest the mechanism may be via the
exposed to non-AMD water that had been artificially acidified formation of iron oxyhydroxides and its coprecipitation of or
to the same range of pH as in the first trial showed compar- absorption to other metals [30]. Furthermore, it has been sug-
atively greater survival at lower pH (Fig. 2, compare to Fig. gested that iron additions (up to 3.9 mg/L) increased survival
1B). Survival at pH 3.3 was low (40%) but significantly of Deleatidium and Zealandiobus spp. at low pH (i.e., pH 3)
greater than survival at pH 3.1 (F7,32 15.66, p 0.01) (Fig. (M.K. Anthony. 1999. Masters thesis. University of Canter-
1138 Environ. Toxicol. Chem. 27, 2008 K. OHalloran et al.

Table 1. pH and dissolved metal concentrations (mg/L) in acid mine drainage (AMD)

Treatmenta Sample pH Fe Al As Mn Ni Zn

U AMD 3.1 22.4 8.9 0.005 0.291 0.066 0.381

D 0.5 AMD 3.2 11.2 4.56 0.001 0.152 0.034 0.202
D 0.25 AMD 3.5 3.87 2.24 0.001 0.077 0.017 0.108
D 0.125 AMD 3.8 2.29 1.22 0.001 0.045 0.010 0.066
D 0.062 AMD 4.0 1.05 0.631 0.001 0.026 0.005 0.040
U, P AMD 4.6 0.07 5.44 0.001 0.263 0.063 0.351
D, P 0.5 AMD 4.6 0.09 2.50 0.001 0.144 0.031 0.192
D, P 0.25 AMD 4.6 0.11 1.35 0.001 0.076 0.016 0.110
D, P 0.125 AMD 4.6 0.12 0.80 0.001 0.043 0.009 0.067
D, P 0.062 AMD 4.6 0.12 0.24 0.001 0.023 0.004 0.040
Dilution water 4.6 0.05 0.11 0.001 0.008 0.001 0.018
a D diluted; U undiluted; P pH modified to 4.6.

bury, Christchurch, New Zealand). This finding is contrary to tions exceeded generic water-quality criteria of 0.011 and
our results, in which we observed a higher toxicity at low pH 0.008 mg/L, respectively [26], and, therefore, may contribute
in the presence of dissolved Fe at concentrations greater than to any observed toxicity in the AMD sample. Deleatidium
approximately 3.0 mg/L; however, we cannot overlook the spp., however, have been reported to be particularly tolerant
presence of dissolved Al, as well as that of other dissolved to significantly higher concentrations of zinc [13], and Havas
metals, in our samples. and Hutchison [9] found that only aluminum (20 mg/L) and
Aluminum also occurred at a significant concentration in iron (30 mg/L) had any influence on toxicity to Daphnia mid-
the AMD, and a considerable proportion remains in solution dendorffiana at pH 4.5. Those authors concluded that alumi-
even in pH-modified samples. International water-quality cri- num was the key additional factor in the toxicity of pH (H
teria (intended to protect freshwater aquatic life) range from ions). These amendments [9] were within a concentration range
87 to 150 g/L of aluminum [26,27,29]. Lower criteria for similar to that of our AMD samples.
aluminum have been suggested for pH 6.5 (i.e., 0.85 Our experiments comparing mayflies sourced from circum-
g/L) [26,27]. Even in the greatest dilution used in our tests, neutral and naturally acidic streams indicated that depending
the aluminum concentration exceeded these criteria. Although on the water-quality conditions in their natal streams, mayflies
aluminum is reportedly not as toxic to aquatic invertebrates may be more or less tolerant of changing water chemistry.
as it is to fish [31], acute effects have been reported between Previous studies of New Zealand stream invertebrates have
2.3 and 36.9 mg/L [26]. shown that taxa can have highly variable tolerances to differing
The dissolution status of all other metals that we measured water-quality conditions [5,6,8,10]. Specifically, many catch-
(i.e., arsenic, manganese, nickel, and zinc) were not noticeably ments within the west coast of South Island have naturally
influenced by pH and remained primarily in solution even at acidic, brown waters [6]. These naturally low-pH systems re-
pH 4.6. The concentrations of dissolved manganese and arsenic ceive fluvic and humic acids generated by percolation of rain-
were at very low levels and were unlikely to be influencing fall through organic soils and surface vegetation [6]. Healthy
toxicity in the AMD sample. Both nickel and zinc concentra- communities of benthic fauna that exist in these low-pH sys-
tems exhibit a degree of tolerance to acidic conditions, and
taxonomic richness can be similar to that of circumneutral
streams [5]. Intraspecific variability in tolerances to acidic con-
ditions between populations has been shown in stream inver-
tebrates [7], amphibians [32], and fish [33]. Mackie [12] ob-
served that populations of both Hyalella azteca and Amnicola
limosa sourced from low-alkalinity waters could tolerate lower
pH compared with populations sourced from high-alkalinity
waters. France and Stokes [7] also showed that populations of
the Canadian amphipod H. azteca sourced from acidic lakes
in Ontario survived longer at lethal pH than did conspecifics
from circumneutral lakes. Furthermore, those authors sug-
gested that acid tolerance could be determined genetically.
Most acid-tolerant species have physiological adaptations that
result in an enhanced ability for osmoregulation at low pH,
such as a greater buffering capacity of the hymolymph or a
lower permeability of the cuticle or gill epithelium to hydrogen
ions [34]. Collier et al. [5] suggested that intraspecific variation
in acid tolerance may be widespread in some parts of New
Fig. 2. Survival of mayflies (sourced from stream with low pH) in Zealand. Our mayfly data strongly support this contention.
uncontaminated stream water that was artificially acidified to various Identification to the species level would have allowed more
pH levels with the addition of HCl. Values are presented as the mean
standard error (n 5 replicates). Letters denote significantly dif-
certainty regarding whether differences in sensitivity to AMD
ferent groups at 96 h. pH 3.1; pH 3.3; pH 3.5; were the result of varied species compositions at the six dif-
pH 3.8; pH 4.0; # pH 4.2; pH 4.5; pH 5.7. ferent locations where mayflies were sourced. Taxonomically,
Mayfly response to acid mine drainage Environ. Toxicol. Chem. 27, 2008 1139

Fig. 3. Mayflies from three circumneutral and three acidic streams exposed to acid mine drainage waters diluted to give low-pH (3.23.3),
medium pH (3.5), and higher pH (4.0) levels. Survival in mayflies sourced from (AC) circumneutral streams (pH 7.07.4) and (DF) acidic
streams (pH 5.76.5). Values are presented as the mean standard error (n 8 replicates). pH 3.23.3; pH 3.5; pH 4.0.

Deleatidium spp. are difficult to identify confidently to the pacted streams could provide an ecologically relevant water-
species level, and it is standard biomonitoring practice in New quality target for remediation of AMD-damaged streams.
Zealand not to distinguish this taxa below the genus level [35]. For a population to be successful, organisms need to be
The time-consuming nature of species identification in the field able to survive all life stages, reproduce successfully, and have
would make running the laboratory-based toxicity tests re- access to adequate food resources. Although the mayfly sur-
ported in the present study unfeasible. Retrospective species vival assays reported here only indicate acute toxicity, they
identification at the six streams may provide further confir- have allowed us to differentiate toxicity and have provided
mation; however, there does appear to be a definite increased critical information regarding mayfly tolerances according to
pH tolerance in Deleatidium spp. sourced from acidic streams. where mayflies were sourced. These kinds of data can provide
The longer-term sublethal impacts of heavy metals con- valuable information regarding the likelihood of remediated
tained in AMD were not evaluated in the present study, which waters to support aquatic invertebrate life. Understanding the
only looked at one kind of AMD from a single source. Based variable tolerances of common lotic benthic taxa can provide
on these 96-h mortality tests, we cannot attempt to draw con- ecologically realistic water-quality criteria for mine remedia-
clusions regarding the chronic effects on organisms inhabiting tion.
the receiving waters of all kinds of AMD. For the type of
AMD that we studied, however, our acute toxicity data do
AcknowledgementWe thank Ingra Smith, Denise Jones, Veronica
indicate that pH probably is the main factor underpinning acute McLeod, Annabel Barnden, and the staff from CRL Energy who
toxicity to mayflies, but overall tolerance to AMD appeared assisted with field and laboratory work; Guy Forrester for assisting
to be strongly influenced by the chemistry of their natal stream. with statistical analysis; and Christine Bezar for editorial advice. We
This has important consequences in setting water-quality cri- also thank several reviewers for their help improving the manuscript.
teria for remediation and rehabilitation of AMD-affected sys- The study was funded by the Foundation for Research, Science, and
Technology (CRLX0401).
tems. If the aim of AMD remediation is to improve water
quality to an extent that it supports healthy biotic communities, REFERENCES
then remediation thresholds should reflect the likely water 1. Gray NF. 1997. Environmental impact and remediation of acid
chemistry of streams that may act as sources for recolonists mine drainage: A management problem. Environ Geol 30:6271.
of a restored stream. The water chemistry of nearby unim- 2. DeNicola DM, Stapleton MG. 2002. Impact of acid mine drainage
1140 Environ. Toxicol. Chem. 27, 2008 K. OHalloran et al.

on benthic communities in streams: The relative roles of sub- roinvertebrate response to acid mine drainage: Community met-
stratum vs. aqueous effects. Environ Pollut 119:303315. rics and on-line behavioral toxicity bioassay. Environ Pollut 130:
3. Letterman RD, Mitsch WJ. 1978. Impact of mine drainage on a 263274.
mountain stream in Pennsylvania. Environ Pollut 17:5373. 20. Scrimgeour GJ, Winterbourn MJ. 1989. Effects of floods on epi-
4. Harding JS. 2005. Impacts of metals and mining on stream com- lithon and BE3-benthic invertebrate populations in an unstable
munities. In Moore TA, Black A, Centeno AA, Harding JS, New Zealand river. Hydrobiologia 171:3344.
Trumm DA, eds, Metal Contaminants in New Zealand. Caxton, 21. Clesceri LS, Greenberg AE, Eaton AD, eds. 1998. Standard
Christchurch, New Zealand, pp 343358. Methods for the Examination of Water and Wastewater, 20th ed.
5. Collier KJ, Ball OJ, Graesser AK, Main MR, Winterbourn MJ. American Public Health Association, American Water Works As-
1990. Do organic and anthropogenic acidity have similar effects sociation, and Water Environment Federation, Washington, DC.
on aquatic fauna? Oikos 59:3338. 22. Zar JH. 1996. Biostatistical Analysis. Prentice Hall, Upper Saddle
6. Collier KJ, Winterbourn MJ, Jackson RJ. 1989. Impacts of wet- River, NJ, USA.
land afforestation on the distribution of benthic invertebrates in 23. McConway KJ, Jones MC, Taylor PC. 1999. Statistical Modeling
acid streams of Westland, New Zealand. N Z J Mar Freshw Res Using GENSTAT. Arnold, London, UK.
23:479490. 24. Poleo ABS, Lydersen E, Rosseland BO, Kroglund F, Salbu B,
7. France RL, Stokes PM. 1987. Life stage and population variation Vogt RD, Kvellestad A. 1994. Increased mortality of fish due to
in resistance and tolerance of Hyalella azteca (Amphipoda) to changing Al-chemistry of mixing zones between limed streams
low pH. Can J Fish Aquat Sci 44:11021111. and acidic tributaries. Water Air Soil Pollut 75:339351.
8. Harding JS, Quinn JM, Hickey CW. 2000. Effects of mining and 25. Pentreath RJ. 1994. The discharge of waters from active and
forestry production. In Collier KJ, Winterbourn MJ, eds, New abandoned mines. In Hester RE, Harrison RM, eds, Mining and
Zealand Stream Invertebrates: Ecology and Implications for its Environmental Impact. The Royal Society of Chemistry, Cam-
Management. New Zealand Limnological Society, Christchurch, bridge, UK, pp 121131.
pp 230259. 26. Australian and New Zealand Environment and Conservation
9. Havas M, Hutchison TC. 1982. Aquatic invertebrates from Smok- Council, Agriculture and Resource Management Council of Aus-
ing Hills, NWT: Effects of pH and metals on mortality. Can J tralia and New Zealand. 2000. Australian and New Zealand guide-
Fish Aquat Sci 39:890903. lines for fresh and marine water quality. National Water-Quality
10. Winterbourn MJ, McDiffett WF. 1996. Benthic faunas of streams Management Strategy, Vol 1. Canberra, ACT, Australia.
of low pH but contrasting water chemistry. Hydrobiologia 341: 27. Canadian Council of Ministers of the Environment. 2005. Ca-
nadian Environmental Quality Guidelines. Update 5.0. Winni-
101111.
peg, MB.
11. Winterbourn MJ. 1998. Insect faunas of acidic coal mine drain-
28. UK Environment Agency. 2007. Environmental quality standards
ages in Westland, New Zealand. N Z Entomol 21:6572.
for list 2 dangerous substances (response to EC dangerous sub-
12. Mackie GL. 1989. Tolerances of five benthic invertebrates to
stances directive 76/464/EEC). Bristol.
hydrogen ions and metals (Cd, Pb, Al). Arch Environ Contam 29. U.S. Environmental Protection Agency. 2002. National recom-
Toxicol 18:215223. mended water-quality criteria: 2002. EPA-822-R-02-047. Office
13. Hickey CW, Vickers ML. 1992. Comparison of the sensitivity of of Water, Washington DC.
heavy metals and pentachlorophenol of the mayflies Deleatidium 30. Soucek DJ, Cherry DS, Zipper CE. 2001. Aluminium-dominated
spp. and the cladoceran Daphnia magna. N Z J Mar Freshw Res acute toxicity to the cladoceran Ceriodaphnia dubia in neutral
26:8793. waters downstream of an acid mine drainage discharge. Can J
14. Hickey CW, Clements WH. 1998. Effects of heavy metals on Fish Aquat Sci 58:23962404.
benthic macroinvertebrate communities in New Zealand streams. 31. Wren CD, Stephenson GL. 1991. The effect of acidification on
Environ Toxicol Chem 17:23382346. the accumulation and toxicity of metals to freshwater inverte-
15. Harding JS, Winterbourn MJ, McDiffett WF. 1997. Stream faunas brates. Environ Pollut 71:205241.
and ecoregions in South Island, New Zealand: Do they corre- 32. Pierce BA. 1985. Acid tolerance in amphibians. Bioscience 35:
spond? Arch Hydrobiol 140:289307. 239243.
16. Winterbourn MJ, Gregson KLD, Dolphin CH. 2006. Guide to the 33. Mills D. 1989. Conservation and management in brown trout,
aquatic insects of New Zealand. Bull Entomol Soc N Z 14:108. Salmo trutta, in Scotland: A historical review and the future.
17. Garcia-Criado F, Tome A, Vega FJ, Antolin C. 1999. Performance Freshw Biol 21:8798.
of some diversity and biotic indices in rivers affected by coal 34. Last J, Johnson KS, Herrick G. 2002. High tolerance of Alderfly
mining in northwestern Spain. Hydrobiologia 394:209217. larvae (Sialis spp. [Megaloptera]) to metals is not affected by
18. Courtney LA, Clements WH. 1998. Effects of acidic pH on ben- water pH. Bull Environ Contam Toxicol 69:370377.
thic macroinvertebrate communities in stream microcosms. Hy- 35. Stark JD. 1998. SQMCI: A biotic index for freshwater macro-
drobiologia 3:135145. invertebrate coded-abundance data. N Z J Mar Freshw Res 32:
19. Gerhardt A, Janssens de Bisthoven L, Soares AMVM. 2004. Mac- 5566.