Vol.

539: 241–254, 2015 MARINE ECOLOGY PROGRESS SERIES

Published November 12
doi: 10.3354/meps11496 Mar Ecol Prog Ser

Harmful effects of sediment-induced turbidity

on juvenile fish in estuaries
M. L. Lowe1, 2,*, M. A. Morrison1, R. B. Taylor2
1
National Institute of Water and Atmospheric Research, Private Bag 99940, Auckland, New Zealand
2
Leigh Marine Laboratory, Institute of Marine Science, University of Auckland, PO Box 349, Warkworth 0941, New Zealand

ABSTRACT: Estuaries are important nursery habitats for juvenile fishes of many species, but are
under increasing pressure from anthropogenic stressors. We examined the impacts of suspended
sediments/turbidity on the foraging success and health of juvenile snapper Pagrus auratus F. Spari-
dae, which are abundant in many northern New Zealand estuaries and comprise a major coastal
fishery as adults. In the laboratory, short-term exposure to turbidity (range <10 to 160 NTU)
reduced foraging success of juvenile snapper, while month-long exposure caused higher rates of
gill ventilation, gill deformation (epithelial hyperplasia, fusion of the lamellae), weight loss and
mortality. In 7 northern New Zealand estuaries with varying catchment land uses, total suspended
sediments were negatively correlated with capture rates of juvenile snapper and the condition of
individuals, and positively correlated with rates of gill deformation and gill parasite loads. Fish
diets changed from zooplankton to benthic prey as turbidity increased. Collectively, our results
indicate that elevated turbidity levels have strong negative effects on the health and abundance
of juvenile snapper. These turbidity impacts compound the negative effects of sediments on
important biogenic fish nursery habitats such as subtidal seagrass beds, likely reducing sub-
sequent recruitment into the fishable adult stock. Management of exploited fish species that have
an estuarine life stage must consider the effects of both fishing and sediment runoff from the
catchment.

KEY WORDS: Foraging success · Condition · New Zealand snapper · Nursery habitat · Pagrus
auratus · Sedimentation · Suspended sediments · Water quality
Resale or republication not permitted without written consent of the publisher

INTRODUCTION Lohrer et al. 2004), while suspended sediments in-

crease turbidity, reducing visibility and depth of the
Estuaries comprise the first and often ultimate re- photic zone and leading to declines in primary pro-
ceiving environment for contaminants generated on ducers with resultant impacts on both pelagic and
the land, and are amongst the most degraded aquatic benthic prey (e.g. Berry et al. 2003, Thrush et al.
habitats worldwide (Syvitski et al. 2005). While high 2004, Morrison et al. 2009).
suspended sediment loads are a natural feature of Estuaries provide crucial nursery habitats for many
many estuarine waters, the rate and extent of ter- species of fish, including the juveniles of recreation-
restrial sediment delivery has accelerated through ally and commercially important species (e.g. Beck et
human activities (Ellis et al. 2004, Thrush et al. 2004). al. 2001, Orth et al. 2006, Lowe 2013, Morrison et al.
Increased sedimentation into the coastal zone alters 2014). These small fish are potentially vulnerable to
ecological functioning in numerous ways. Deposition the effects of increased sedimentation because the
of fine sediments (silt and clay) can abrade, clog and biogenic habitats (e.g. seagrass meadows) that many
smother benthic organisms (Norkko et al. 2002, rely on for food and shelter are themselves suscepti-

*Corresponding author: [email protected] © Inter-Research 2015 · www.int-res.com

242 Mar Ecol Prog Ser 539: 241–254, 2015

ble to sedimentation (Duarte 2002, Green & Short MATERIALS AND METHODS
2003, Grech et al. 2012). Suspended sediment/
turbidity can also affect fish more directly, by reduc- Lab experiments were conducted from May to Sep-
ing both reactive distance and foraging efficiency tember 2004 at the University of Auckland’s Leigh
(e.g. Vinyard & O’Brien 1976, Rowe & Dean 1998, Marine Laboratory. The field survey was done from
Manning 2013), particularly for planktonic prey May to July 2006.
(Wenger et al. 2012, Johansen & Jones 2013), and by
damaging their gills, sometimes fatally (e.g. Sherk et
al. 1975, Au et al. 2004, Wong et al. 2013). Suspended Laboratory experiments
sediments can also modify other ecological processes
including prolonging larval development and the Juvenile snapper Pagrus auratus of 55 to 90 mm
ability to select habitat at settlement (e.g. Wenger et fork length were captured from the Rangaunu and
al. 2011, 2014). Mahurangi Harbours using beach seines and baited
Most of our knowledge of the effects of suspended opera traps (Morrison et al. 2002, Morrison & Car-
sediments on fish is based on studies of (1) freshwater bines 2006), respectively. Fish were initially held in
species (e.g. Newcombe & MacDonald 1991) rather 500 l flow-through tanks, and fed daily with diced
than estuarine species, the latter of which may be mussels and shrimp pellets. After 2 wk, fish were
accustomed to much higher sediment loads, (2) acute transferred to the experimental tanks and fed mysid
effects measured in the lab, rather than chronic effects shrimps Tenagomysis sp. daily to acclimate to the
as experienced by fish under natural field conditions experimental conditions.
(Au et al. 2004), and (3) fish >1 yr old, with limited Preliminary observations revealed that aggression
study on larvae/juveniles, particularly for estuarine increased when fish in a tank differed in length by
fishes (but see Partridge & Michael 2010). Early life >15 mm. To counteract this, fish were allocated to
stages are thought to be more susceptible than older tanks so that their lengths within any given tank dif-
individuals to elevated levels of suspended sedi- fered by no more than 10 mm. Tanks were then ran-
ments (Sigler et al. 1984, Wilber & Clarke 2001), domly assigned to treatments. The overall length
which is of particular concern given that events range of fish in the short-term experiments on for-
occurring in the first few months of life can strongly aging success (run earlier in the season) ranged from
influence subsequent performance (Houde 1987, 55 to 74 mm, with an average (± SE) length of 66 ±
Sogard 1992, Francis 1994, Manderson et al. 2002, 2 mm. For the long-term experiment on fish health
Sutherland & Meyer 2007). (run later in the season), overall fish lengths ranged
We determined the effects of suspended sedi- from 55 to 90 mm, with an average of 72 ± 2 mm.
ments/turbidity on the health and foraging success There was no significant difference in average initial
of juvenile snapper Pagrus auratus F. Sparidae. fish length between turbidity treatments for either
Snapper form the basis of a major coastal fishery in the short (F = 1.35, p = 0.269) or long-term experi-
warm temperate northern New Zealand (Francis ments (F = 1.35, p = 0.289).
1994), and estuaries are an important nursery for Mysids were used for the feeding experiments as
juveniles, particularly when they contain biogenic they are a common prey item for juvenile snapper
habitats such as seagrass beds (Lowe 2013, Morri- (Usmar 2012). The mysids were collected from the
son et al. 2014, Parsons et al. 2014). New Zealand nearby Whangateau Harbour, held in a flow-through
estuaries have catchments ranging from near pris- glass aquarium (0.6 × 0.35 × 0.3 m), and fed daily
tine to heavily modified, with corresponding varia- with frozen brine shrimp pellets and organic detritus
tions in estuarine sediment input and turbidity off shells collected from the harbour.
(Swales et al. 2002, Morrison et al. 2009), which Experiments were conducted in 15 rectangular
potentially affects the value of these estuaries as plastic aquaria (0.5 × 0.26 × 0.28 m) located in a
nurseries. Short-term impacts of turbidity on snap- building with a translucent roof, which allowed a
per health and foraging success were measured by natural photoperiod. Aquaria had 10 mm mesh lids to
experimentally manipulating turbidity levels rang- prevent fish escaping. All experiments were con-
ing from very low to those occurring in the field ducted during late afternoon (15:00 to 17:00 h) to
after heavy rainfall. To provide context for the standardise for diel activity patterns in snapper or
results of these lab experiments, we also quantified mysids (e.g. Minello et al. 1987, Macia et al. 2003). In
the abundance, diet and health of juvenile snapper all experiments, there were 3 aquaria per turbidity
in 7 estuaries varying greatly in turbidity. level and 3 fish per aquarium. Each aquarium was a
 Lowe et al.: Effects of turbidity on estuarine fish 243

closed system containing a submersible pump (Hi- (20, 40 and 80 NTU), but turbidity in the 160 NTU
tech 3500, 3.6 l min−1) and aerator. treatments declined by up to 10% in some trials.
A total of 5 turbidity levels were used in the exper- To determine the effect of suspended sediment/tur-
iments: ≤10 (control), 20, 40, 80 and 160 nephelomet- bidity on health, fish were maintained in aquaria for
ric turbidity units (NTU), encompassing the natural 1 mo. Fish and treatments were randomly assigned to
range experienced by juvenile snapper in the nearby individual tanks. The weight and length of each fish
Mahurangi Harbour over an 11 mo period (Oldman & were measured at the start and end of the experi-
Swales 1999). During the experiments, turbidity was ment. Sediments were kept in suspension by circu-
measured with a Hach 2100 portable turbidity sen- lating the water within each tank through 3 inde-
sor. Due to natural variation in incoming ambient pendently-tipping 200 ml cups mounted above the
seawater, the control tank was defined as ≤10 NTU, tank (adapted from Barr 2007). This provided inter-
as per Cyrus & Blaber (1987a,b). Turbidity was pro- mittent turbulence to the bottom of the tank. Sea-
duced by means of fine surficial estuarine sediments water was delivered via adjustable flow nozzles
(< 63 µm) collected from the subtidal zone of the above the plastic cups. The cups were equipped with
Mahurangi Estuary. Replicate samples of sediment 13 mm plastic sleeves (to act as a bearings) inserted
were collected to encompass potential spatial vari- through a pivot point, slightly forward of the geomet-
ability in distributions of grain sizes. The sediment ric center of gravity in the side of the cup. The 3 cups
was initially mixed with surface seawater, and al- were positioned equidistantly, on a 10 mm galva-
lowed to settle for 1 h. The container was then stirred nized pipe above the tank. This pipe acted as a pivot,
and left for 6 min, before the top 10 cm of water and allowing the cups to tip when full and then return to
suspended sediment was siphoned off and poured an upright position (see Fig. 3.3a in Lowe 2013).
through a 63 µm sieve. This process was repeated Each day, a 1 l jug of water was scooped from each
until enough stock solution was obtained. Sediment tank and poured back into it in order to re-suspend a
was collected twice a week and kept in a well- small amount of sediment that settled along the rear
aerated container. Turbidity levels were maintained wall of the tank where the turbulence from the tip-
in the tanks over time by the addition of varying ping cups did not reach. Turbidity levels were meas-
amounts of stock solution to the ambient incoming ured each day from random locations within the tank
seawater, until the intended turbidity levels (i.e. 20, and additional sediment from the stock solution was
40, 80 or 160 NTU) were achieved. To maintain water added as required to maintain turbidity levels within
depth, a matching volume of water was removed 20% of the target level. Approximately 25% of the
after the addition of stock solution. To avoid any sub- tank water was changed daily to allow addition of
strate effects for the short- and long-term turbidity stock solution and reduce waste build-up. Water and
experiments, tank bottoms were left clear without sediment were completely replaced every 7 d. Fish
any sediment cover. were removed and placed into holding tanks of
The effect of turbidity on foraging success was de- ambient seawater for ~10 min while tanks, pumps
termined as follows. Juvenile snapper were accli- and connecting pipes were cleaned.
mated to experimental tanks for 2 h and starved for Initially, fish were placed into the experimental
24 h prior to each trial. The water pump was switched tanks for 48 h with ambient seawater. The tipping
off 2 h prior to commencement of the short-term cups were then turned on and off over a period of
feeding experiments. At the start of the experiments, 48 h to allow the fish to acclimate to the apparatus
60 mysids were released into the opposing end of and ensure that they were feeding. Sediment was
each tank, and were separated from the fish by a Per- added gradually over 48 h for the higher NTU treat-
spex divider for 30 min to acclimate, after which time ments. Tipping cups were switched off daily
the divider was removed and fish left to feed. After (between 15:00 and 17:00 h) for 20 min after fish
30 min of feeding, fish were removed, the tank was were provided ~120 live mysids tank−1. Data on
completely drained through a 250 µm sieve, and the daily ambient seawater temperatures were taken
remaining mysids counted. All mysids missing from from the Leigh Marine Laboratory’s Climate Data
the tanks were considered to have been eaten. Pre- Archives (data not shown). During the final week of
liminary trials revealed that all shrimps were recov- the experiment, the ventilation rate of the fish (6
ered from tanks containing no fish. Turbidity levels fish treatment−1) was assessed and expressed as the
were re-measured at the conclusion of the feeding rate of gill operculum opening per 15 s. The short
trials (~3 h after adding the sediment). There was a time frame was necessary to accommodate the
negligible change in the lower turbidity treatments reduced visibility of fish in the higher turbidity
244 Mar Ecol Prog Ser 539: 241–254, 2015

randomly chosen entire gill filaments.

Parasites were also identified and
counted on the 3 randomly selected gill
filaments. Where fusion of lamellae had
occurred, the number of individual
lamellae that were joined together was
counted as in Tricklebank (1997).

Field survey

We attempted to collect at least 10

juvenile snapper (fork length 50 to
100 mm) from each of 7 northern North
Island estuaries covering a spectrum of
environmental degradation (sedimenta-
tion and associated water turbidity;
Fig. 1). Fish were caught using opera
traps, which were baited with pilchards
and set over sandy mud between 14:00
and 18:00 h NZST on an incoming tide.
Fish were euthanized by the iki-jime
method and the first right gill arch re-
moved and processed as described for
the laboratory experiments. Fish were
then placed into an ice slurry for trans-
portation back to the lab.
At each site, a 0.25 m diameter black-
and-white Secchi disc was used to meas-
ure water clarity. A Horiba U10 multi
Fig. 1. Estuaries sampled for juvenile snapper Pagrus auratus in northern
probe was used to measure turbidity (in
New Zealand
NTUs), temperature, salinity, pH and
dissolved oxygen at ~1 m above the sea
treatments moving away from the forward wall of floor. Absence of a power source in the field pre-
the tank. cluded using the more accurate Hach Turbidimeter
Upon termination of the trial, all surviving fish for measurements of turbidity as in the laboratory
were euthanized by the iki-jime method (Robb & experiments. To quantify the concentration of total
Kestin 2002), and the first gill arch on the right side of suspended sediments (TSS), a single 1 l water sample
each fish was removed and immediately fixed in was collected from ~1 m above the seafloor at each
Bouin’s fluid for 24 h, then transferred to 70% estuary, using a Van Dorn sampler. Water samples
ethanol. Samples were then dehydrated in graded were later filtered through acid-washed, dried and
ethanol concentrations and embedded in paraffin pre-weighed 0.45 µm polycarbonate membranes
wax. Sagittal sections (4 to 7 µm thick) were cut and using plastic, acid-washed, vacuum filtration equip-
mounted on glass slides. Sections were de-paraf- ment. After filtration, the membrane was dried to a
finized in xylene, hydrated in ethanol and stained constant weight at 60°C and re-weighed.
with hematoxylin-eosin. The gill epithelia were ex- To quantify among-estuary differences in fish
amined for evidence of any pathological changes that growth and nutritional condition, a condition index
may have compromised respiratory function. This (CI) was calculated. In the laboratory, the fork length
included epithelial hyperplasia of the pillar system of each snapper was measured (±1 mm) and its total
(increased proliferation of cells, particularly at the wet weight determined (± 0.001 g). The liver and
base and tips of the lamellae), fusion of the secondary digestive system (i.e. stomach and intestine) were re-
lamellae and the presence of parasites. Counts of the moved and weighed separately, and subtracted from
number of affected lamellae were quantified along 3 the total weight to yield a carcass weight (CW). A rel-
 Lowe et al.: Effects of turbidity on estuarine fish 245

ative CI was calculated for each individual as per During the month-long experiment, fish main-
Francis (1997), as the measured CW divided by the tained at all turbidity levels lost weight on average,
expected carcass weight (ECW) for an ‘average’ fish but the weight loss was significantly higher at tur-
of the same length. The ECW was generated using bidities ≥40 NTU (p = 0.02; Fig. 2B). Weight loss for
the power curve ECW = aLengthb, where a and b the ≤10 and 20 NTU treatments averaged ~7%, dou-
were estimated from a linear regression of log10CW bling to ~14% for the higher treatments.
on log10Length for fish from all sites. Sites were Gills of healthy fish typically have equally spaced
pooled after initial testing to ensure they had homo- secondary lamellae, and intact cellular layers with
genous regression slopes (Francis 1997). no signs of fusion between the lamellae (Trickle-
Foreguts were then preserved in 10% formalin and bank 1997). In this experiment, clear histopathologi-
the contents identified to the lowest possible taxo- cal changes in gill epithelia occurred with increas-
nomic level under a dissecting microscope. To esti- ing turbidity levels (Figs. 2C−E & 3). The occurrence
mate biomass, animals were allocated to log size- of epithelial hyperplasia (causing dilation of the
classes by eye using a graticule in the microscope lamellae) increased significantly from 20 filament−1
and a reference collection consisting of a mixture of in the control (≤10 NTU) to a peak of 60 at 80 NTU
species retained by different-sized sieve meshes fol- (p = 0.01; Fig 2C). Similarly, the incidence of lamel-
lowing Edgar (1994). The equation of Edgar (1990) lar fusion increased significantly as turbidity in-
was then used to estimate individual body mass from creased, from 10 filament−1 at ≤10 NTU to 27 fila-
sieve mesh size. Ash-free dry weight values utilized ment−1 at 160 NTU (p = 0.003; Fig. 2D). However,
for estimation of plankton biomass were calculated there was no evidence of mechanical abrasion or
from Newcombe (2009). Prey were classified as ben- lodging of sediments into gill epithelia of the lamel-
thic (amphipods, shrimps, isopods, decapods, bi- lae (M. L. Lowe pers. obs.). Epitheliocystis, a bacter-
valves, polychaetes and mysids) or pelagic (zoo- ial condition affecting the gills of fish, also increased
plankton, gastropod veligers) and biomasses were significantly with higher turbidities (p = 0.02;
summed across these 2 categories. Mysids were con- Fig. 2E). In total, 88% of the fish in the 160 NTU
sidered benthic in estuaries with TSS/turbidity val- tanks and 67% of fish in the 80 NTU tanks devel-
ues above 20 mg l−1 or 10 NTU respectively, due to oped lesions, but only 0.25% in the 20 NTU treat-
their propensity to school within clear/shallow ment. No infection was present in the control
waters and move independently close to the seafloor (≤10 NTU). Similarly, rates of infection per gill fila-
in turbid waters (Mauchline 1980). ment ranged from a peak of 3.0 ± 1.18 fish−1 for 160
NTU to 0.25 ± 0.16 fish−1 for the 20 NTU treatments.
Quantitative behavioural observations of fish at the
Analysis higher turbidity levels (80 and 160 NTU) were diffi-
cult due to poor visibility. However, the behaviour of
Ordinary least squares regression was used to the fish that could be observed revealed increased
model fish health, foraging success and abundance gill flaring, coughing and gulping at the surface
as a function of turbidity (Cottingham et al. 2005), along with decreased activity levels, particularly for
using Sigma Plot 11 run on treatment or estuary the 80 and 160 NTU treatments. Fish in these treat-
means. Results were considered significant at p < ments tended to settle on the bottom, were lethargic
0.05. Fish that died during the experiments were and interacted little with other fish. Increased venti-
excluded from analyses. lation rates occurred at the higher turbidity levels,
although marginally insignificant (p = 0.07; Fig. 2F).
These ranged from 80 ± 5.98 min−1 for ≤10 NTU to
RESULTS 110 ± 1.36 min−1 at 160 NTU.
Only 5 fish died during the experiment. In the
Laboratory experiments lower turbidity treatments (≤10 and 20 NTU), 2 died
within the first week of the trial, associated with
The proportion of mysids eaten by juvenile snapper aggressive attacks by other individuals. In the higher
during feeding trials decreased significantly as tur- turbidity treatments (80 and 160 NTU), 3 died in the
bidity increased (p < 0.002), with only 8% of mysids final days of the experiment. These fish were ex-
consumed in aquaria with turbidities equating to tremely thin and lethargic, having lost ~25% body
storm conditions (160 NTU) compared to 77% con- weight, and were very pale with evidence of fin rot
sumed in the control (≤10 NTU; Fig. 2A). from fungal infections.
 246 Mar Ecol Prog Ser 539: 241–254, 2015

100 18
A. Foraging success B. Weight loss

% Weight loss (mean ± SE)

Percentage mysids consumed

16
80 14
12
(mean ± SE)

60
10
ƒ = 89.579*exp(-0.014*x)
r2 = 0.96 8
40 F = 90.13
p = 0.002 6 ƒ = 13.980*(1-exp(-0.070*x))
2
r = 0.84
20 4 F = 16.09
p = 0.02
2
0 0
10 20 40 80 160 10 20 40 80 160
70 35
C. Gill hyperplasia D. Gill lamella fusion
60 30

Number per filament

Number per filament

50 25

(mean ± SE)
(mean ± SE)

40 20

30 15
ƒ = 55.83*(1-exp(-0.037*x))
20 r2 = 0.92 ƒ = 26.750*(1-exp(-0.021*x))
10
F = 32.96 r2 = 0.95
p = 0.01 F = 71.82
10 5 p = 0.003

0 0
10 20 40 80 160 10 20 40 80 160
5 30
E. Gill epitheliocystis lesions F. Ventilation rate
Number of lesions per filament

28
Ventilation rate per 15 sec.

26
(mean ± SE)

(mean ± SE)

3
24
2
22
ƒ = 3.014/(1+exp(-(x-54.43)/11.295)) ƒ = 17.44+9.72*(1-0.96^x))
1 2
r = 0.99 r2 = 0.92
F = 461
20
F = 13.02
p = 0.002 p = 0.07
0 18
10 20 40 80 160 10 20 40 80 160
Turbidity (NTU) Turbidity (NTU)
Fig. 2. Effects of turbidity on juvenile snapper Pagrus auratus, in experiments lasting (A) 30 min and (B−F) 30 d

Field survey related (TSS−NTU: r2 = 0.93; TSS−Secchi: r2 = 0.96;

NTU−Secchi: r2 = 0.86), so TSS values are presented
The 7 surveyed estuaries had TSS/turbidities rang- along with NTU values to facilitate comparison with
ing from 4 mg l−1 (~1 to 2 NTU) in the Rangaunu estu- the experimental results. Other measured environ-
ary, a pristine isolated area with a relatively undis- mental variables differed little among estuaries, with
turbed catchment, to 37 mg l−1 (17 NTU) in the temperature ranging from 14 to 17°C, pH from 7.9 to
Waitemata, which has a highly modified catchment 8.3, salinity from 29.4 to 35.6 ‰ and dissolved oxygen
that includes a large part of New Zealand’s largest from 7.5 to 10.5 mg l−1.
city Auckland (Fig. 4). Measures of turbidity (i.e. The capture rate of juveniles decreased signifi-
NTU), Secchi depth and TSS levels were highly cor- cantly as TSS/turbidity increased (p = 0.002; Fig. 4A),
 Lowe et al.: Effects of turbidity on estuarine fish 247

with only 3 individuals collected from the most tur-

bid estuary, Waitemata Harbour. The average
length of juvenile snapper was not significantly cor-
related with turbidity (r2 = 0.04; data not shown).
A significant negative relationship was found be-
tween the CI of fish and increasing TSS/turbidity
(p = 0.01), with fish from Rangaunu Harbour weigh-
ing 20% more for their length than fish from the
highly turbid Waitemata Harbour (Fig. 4B). The rate
of gill epithelial hyperplasia (causing dilation of the
lamellae) was lowest in fish from the 2 least turbid
estuaries (12 to 20 filament−1) and was significantly
higher in the 5 more turbid estuaries (30 to 45 fila-
ment−1) (p = 0.03; Fig. 4C). Similarly, the rate of
lamellar fusion per filament increased significantly
with TSS/turbidity, from 1.4 filament−1 in the least
turbid estuary to 23.7 in the most turbid (p = 0.03;
Fig. 4D). Other histopathological lesions included
hypertrophy (swelling) and shortening of the lamel-
lae (Fig. 3). The occurrence of affected lamellae was
proportionately greater in the southern harbours,
particularly for Mahurangi (85%) and Manukau
(80%). Conversely, this condition was largely absent
from the northernmost harbours, Rangaunu (0%)
and Whangateau (2%). Between these, Kaipara
recorded 20% and Waitemata and Tamaki recorded
30% respectively. Lesions caused by epitheliocystis,
a bacterial condition affecting fish gills, were pres-
ent on the gills of fish from 4 sites, but the abun-
dance of lesions showed no clear relationship with
turbidity (p = 0.14; Fig. 4E). The Manukau had the
highest proportion of affected fish (67%), followed
by Mahurangi (38%), Waitemata (33%) and Tamaki
(14%).
Mysid and caridean shrimps, and copepods domi-
nated the diets of fish in all 7 estuaries. The relative
contribution of the 2 prey categories (benthic, pe-
lagic) varied significantly with increasing suspended
sediments (p = 0.01; Fig. 4F). In estuaries with rela-
tively clear water (Rangaunu, Mahurangi, Whanga-
teau, Tamaki), pelagic prey dominated (98% total
numbers), especially calanoid copepods such as
Paracalanus indicus and the cladoceran Penilia avi-
rostris, while fish in the more turbid estuaries (Manu-
kau, Kaipara, Waitemata) exclusively ate benthic
prey comprising mysid and caridean shrimps, includ-
Fig. 3. Gills of juvenile snapper Pagrus auratus subject to
different turbidities. (A) Normal gill filament (f) in fish from ing Tenagomysis sp. and juvenile Palaemon affinis.
low-turbidity Rangaunu Harbour. l: lamellae. (B) Hyperpla- Modest numbers of benthic copepods Hemicyclops
sia (cell proliferation) of gill filaments especially at the base sp. and amphipods were also consumed. In addition,
and tips (clubbing; c) of the lamellae, and lamellar fusion (lf) the average size of prey items increased with ele-
in fish maintained in highly turbid (160 NTU) water in an
aquarium for 30 d. Note also the bacterial lesion (bl). (C)
vated turbidity levels: 91% of the pelagic prey was
Hypertrophy (thickening) and shortening of a gill filament ≤0.71 mm, while ~70% of benthic prey items con-
in fish from the moderately turbid Manukau Harbour sumed were ≥0.71 mm.
 248 Mar Ecol Prog Ser 539: 241–254, 2015

(20) 1.15
Rangaunu
A. Abundance Rangaunu B. Condition

Condition index (mean ± SE)

8
1.10
Catch rate (ind./pot/h)

ƒ = 12.09*exp(-0.115*x)
r2 = 0.86 1.05 Tamaki
6 F = 31.97 Whangateau
p = 0.002 Manukau Kaipara
1.00 Mahurangi

4 0.95
Waitemata
(10) (13)
Whangateau Manukau
(10)
Mahurangi 0.90 ƒ = 1.069 - 0.004*x
2 (10) 2
Tamaki (7) r = 0.73
Kaipara (3) 0.85 F = 13.84
Waitemata p = 0.01
0 0.80
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
50 30
C. Gill hyperplasia D. Gill lamella fusion
Number per filament (mean ± SE)

Number per filament (mean ± SE)

40 25

20
30
15
20
ƒ = 38.547*(1-exp(-0.122*x))
2
10 ƒ = 2.192+0.537*x
r = 0.64 2
r = 0.99
F = 8.9
10 p = 0.03
F = 8.5
5 p = 0.03

0 0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
Pelagic prey biomass (%) (mean ± SE)

4
Number per filament (mean ± SE)

E. Gill lesions 100 F. Prey type

3 80
2
r = 0.48
F = 1.8 ƒ = 61.1/(1+exp(-(x-21.714)/-0.114))
p = 0.14 60 2
r = 0.89
2 F = 17.85
p = 0.01
40

1
20

0
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
Suspended sediments (g m–3) Suspended sediments (g m–3)

0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18
NTU NTU
Fig. 4. Relationships between the turbidity of 7 estuaries in northern New Zealand and the abundance, health, and diet of
juvenile snapper Pagrus auratus. Numbers in parentheses in (A) represent sample sizes. Data are plotted against measured
total suspended sediment (TSS) values, which provide higher accuracy. Nephelometric turbidity unit (NTU) values are plotted
as an extra axis to facilitate comparison to experimental results, based on the strong correlation between NTU and TSS from
the field data (r2 = 0.93)

DISCUSSION exposure to short-term (30 min) turbidity pulses.

Chronic exposure (30 d) at turbidity levels equating to
This study clearly demonstrates the impacts of in- storm conditions caused acute effects on fish growth
creasing sedimentation/turbidity on juvenile snapper, and health, including increased weight loss, mortality,
with foraging success declining markedly following presence of gill lesions and behaviours associated
 Lowe et al.: Effects of turbidity on estuarine fish 249

with hypoxia (i.e. gulping at the surface, lethargy and A forced shift from small zooplankton to large ben-
increased ventilation). A survey of 7 northern estuaries thic prey is likely to be detrimental to juvenile snap-
revealed a decline in abundance and health (i.e. sig- per. Newly-settled fish are limited in their ability to
nificantly lower CI and higher incidence of gill eat larger prey due to their small gape size and lim-
lesions) in the more impacted estuaries characterized ited ability to digest larger prey because of their
by increasing sedimentation, lower water clarities and poorly differentiated guts (Morrison 1990, Gillanders
increasing urbanisation. This is consistent with direct, 1997, Sudo & Azeta 2001). Juvenile fish prefer
chronic effects of sediments on fish. smaller, easily digestible prey such as copepods
(Mills et al. 1984, Lankford & Targett 1997, Gning et
al. 2010) because of their high caloric and protein
Impacts of turbidity on feeding content, which facilitates rapid growth to minimize
size-dependent predation risk (Volk et al. 1984).
Turbid water can reduce the feeding ability of However, increased TSS can also negatively affect
visual predators such as juvenile fish, by reducing the abundance, nutritional value and composition of
both reactive distance and foraging efficiency (Kerr zooplankton and other prey species (Forbes et al.
1995, Bash et al. 2001). The results from our survey 1981, David et al. 2005). Thus, increasing TSS/
are consistent with prior research (e.g. Vinyard & turbidity may reduce the condition and growth of
O’Brien 1976, Rowe & Dean 1998, Wenger et al. 2012, juvenile snapper by reducing their ability to visually
Manning 2013), with foraging success of juvenile choose zooplankton at optimal prey sizes, and/or by
snapper on mysids declining by 44 to 89% under reducing zooplankton densities. Accordingly, both
experimental exposure to short-term pulses of tur- the magnitude and timing of sediment pulses may
bidity ≥40 NTU. Even greater reductions in foraging have profound and long-term effects on survival and
success could be expected if our experiment had recruitment of larval or juvenile fish, particularly
used smaller, highly mobile planktonic prey, which where diet is restricted due to ontogeny (Partridge &
are a major dietary item for newly-settled snapper Michael 2010, Murphy et al. 2012, Manning 2013,
(Usmar 2009, Johansen & Jones 2013). This may be of Wenger et al. 2014).
particular concern given the increasing frequency of
sudden downpours due to climate change, and the
short retention times of flood waters in catchments Impacts of turbidity on health
(Willis et al. 2007).
In the field, the diet of juvenile snapper changed There are significant thresholds of susceptibility to
abruptly from highly mobile pelagic prey (zooplank- suspended sediment/turbidity in many sensitive spe-
ton) to larger, slower-moving benthic prey (mysids, cies and life stages (e.g. Sigler et al. 1984, Wilber &
shrimps and amphipods) when TSS/turbidity ex- Clarke 2001, Partridge & Michael 2010). The results
ceeded 23 mg l−1 (~12 to 17 NTU). The extent to from this study have revealed a more gradual
which this diet difference was due to differences in response to increased TSS/turbidity, with signifi-
prey abundance or to turbidity levels in the estuaries cantly higher rates of sublethal stress above 40 NTU
is not known since prey densities were not sampled for chronic exposure (30 d) in the laboratory. In the
concurrently. Simultaneous spatiotemporal assess- lab experiments, weight loss reached a threshold
ments of prey abundance merits further study. How- response (~14%) at 40 NTU, levelling off for the 80
ever, assuming both prey types were present at all and 160 NTU treatments. Nominal weight losses
field sites, this suggests a change in foraging tac- recorded for the control group (≤10 NTU) could be
tics — which has been documented in other studies, due to stress from confinement and handling and/or
often with a ‘turbidity’ threshold of around 40 NTU, ambient sea surface temperatures being the lowest
above which declining vision results in more oppor- on record for 38 yr during this trial, which was con-
tunistic, ambush-type predation (Hecht & van der ducted during autumn and winter (Leigh Marine
Lingen 1992, Macia et al. 2003, De Robertis et al. Laboratory’s Climate Data Archives, data not
2003, Helenius et al. 2013). Strategy shifts of this shown). Growth rates for 0+ age juvenile snapper are
nature are likely cost-effective given the higher ener- correlated with temperature, and slow dramatically
getic costs associated with increased foraging time, over the first winter (Francis 1994). Our results are
attendant declines in attack success (Johansen & consistent with prior research of freshwater fish
Jones 2013), and increased exposure to predation showing reduced growth rates, with escalation of ill
(Meager et al. 2005, Engström-Öst & Mattila 2008). effects following chronic exposure to increasing sus-
250 Mar Ecol Prog Ser 539: 241–254, 2015

pended sediments (e.g. Sherk et al. 1974, 1975, New- lethal stress and, in combination with suspended
combe & Jensen 1996, Manning 2013), and suggest sediments, may have important interactive effects on
that suspended sediment concentration alone is a marine fishes (Au et al. 2004). Therefore laboratory
poor indicator. Rather, effective predictive modelling results may be conservative in their estimation of sus-
requires both concentration and exposure duration pended sediment impacts under field conditions.
(Newcombe & MacDonald 1991). In sum, these experiments demonstrate that al-
Juvenile snapper showed a progressive increase though turbidity levels in the range investigated here
in the incidence of damage to gill lamellae with in- (10 to 160 NTU) are unlikely to cause immediate high
creasing turbidity in long-term exposure trials. Gill mortality in juvenile snapper, prolonged exposure to
pathological symptoms observed included epithelial turbidities ≥40 NTU can result in adverse growth and
hyperplasia (thickening) of the pillar system and developmental effects from (1) reduced prey capture
fusion of the secondary lamellae. These changes re- success due to impaired vision and/or (b) increased
present sublethal rather than acute effects, and can metabolic costs from physiological stress (e.g. respi-
be considered a defence mechanism by increasing ratory distress/disease). Hence, if exposure continues
the distance across which irritants must diffuse to in the long-term, reduced energy acquisition could
reach the bloodstream (Mallat 1985). However, cell be reflected by lower CIs of fish from turbid estuaries
proliferation of epithelial tissue, and eventual loss of relative to those with higher water clarity, and/or by
surface by clubbing and fusing of lamellae can reduced fish abundance in highly turbid estuaries
impair respiration and ammonia excretion, leading to (Hecht & van der Lingen 1992, Au et al. 2004).
respiratory stress and ammonia intoxication (Goldes Our field survey of 7 northern estuaries similarly
et al. 1988, Kerr 1995, Bergstedt & Bergersen 1997, revealed significant differences among the biological
Au et al. 2004). Bacterial infection (i.e. epithelio- variables measured for juvenile snapper, particularly
cystis) of the gills was also recorded. Epitheliocystis for TSS above 23 mg l−1 (~12 to 17 NTU). The onset of
affects the epithelial and chloride cells of gills and negative health impacts occurred at lower levels in
cause hypertrophy (swelling of individual cells), the field than in the laboratory trials (i.e. ≥40 NTU in
which can result in respiratory distress (Meijer et al. the lab). However, given the variety of environmen-
2006). The increase in ventilation rates, increased gill tal parameters fish are exposed to in the field (e.g.
flaring, coughing and gulping at the surface, in addi- particle size distribution and angularity, pollutants,
tion to reduced activity levels recorded in the higher prey availability and cumulative storm events) it is
turbidity treatments, are all indicative of anoxia, not unexpected to find substantial variation between
which is consistent with published information (Berg the two (Appleby & Scarrett 1989).
& Northcote 1985, McLeay et al. 1987, Servizi & Juvenile snapper had significantly lower CIs (an
Martens 1992). Paler coloration, fin rot and early effective proxy for growth rate and nutritional status)
mortality observed in 17% of fish from the 80 and 160 in the more impacted estuaries characterized by in-
NTU treatments agrees with prior research, which creasing sedimentation, concomitant with lower
collectively suggests that exposure to elevated sedi- water clarities and increasing urbanisation (e.g. Wait-
ment levels decreases tolerance rates to disease and emata, Manukau), whilst fish from Rangaunu Har-
time to death (e.g. Appleby & Scarrett 1989, Redding bour, the most ‘pristine’ estuary, had the highest
et al. 1987, see review in Newcombe & Jensen 1996). average CIs.
However, with laboratory-based studies, experi- Higher levels of gill deformation (hyperplasia/
mental artefacts are unavoidable since factors such fusion) and parasite loads were also recorded, partic-
as prey availability and intra- and interspecific ularly for Manukau and Mahurangi Harbours. More-
encounters are controlled (Bash et al. 2001). In addi- over, observed gill lesions also included shortening
tion, spatial and temporal factors such as distribution, and hypertrophy (swelling of epithelial cells) of sec-
abundance or availability of suitable habitat, time of ondary lamellae, which showed close similarity to
year, frequency, duration and magnitude of prior lesions brought about by elevated levels of other
storm events (with associated increase in current environmental pollutants such as zinc (Bhagwant
velocities) may be more deleterious in the field & Elahee 2002), nickel (Al-Attar 2007), nitrogen
(Bruton 1985). In these experiments, fish were (Schlacher et al. 2007) and phosphate (Omoregie
exposed to sediment unlikely to be contaminated. et al. 2009). The markedly elevated histopathological
However, in coastal environments suspended sedi- alterations recorded for both the Manukau (80%)
ments frequently serve as a sink for contaminants and Mahurangi (85%) harbours may well be a result
(Hack et al. 2007). These chemicals can cause sub- of other environmental pollutants acting alone or
 Lowe et al.: Effects of turbidity on estuarine fish 251

synergistically with TSS. Manukau Harbour, located of fish from a known range of sedimented estuaries
next to New Zealand’s largest city, has historically utilizing otolith chemistry to estimate daily/annual
had high levels of copper (Cu), lead (Pb), zinc (Zn) growth as fish recruit to offshore fisheries would
and polycyclic aromatic hydrocarbons (PAHs) in its enhance our understanding of these processes to bet-
sediments (Hack et al. 2007, Mutoro 2001), whilst ter inform managers of acceptable NTU thresholds.
both harbours have elevated nitrate and phosphate The findings from this study support the increasing
levels (Scarsbrook 2008). However, determining levels of concern being shown over escalating diffuse
cause and effect relationships between specific source inputs of suspended sediments into rivers, es-
chemicals and their effects on juvenile fish is particu- tuaries and the coastal zone. The direct effects of sus-
larly difficult given the myriad of contaminants and pended sediments on fish health, physiology and be-
their potential synergisms (Tricklebank 1997 and ref- haviour, as well as the associated loss of seagrasses
erences therein, Wong et al. 2013). In addition, and other organisms that form important nursery
higher levels of both TSS and pollutants have been habitats, along with reduced numbers of available
shown to predispose fish to opportunistic infections prey for juvenile fish, collectively pose a significant
(Redding et al. 1987, Goldes et al. 1988). Nowak & and increasing threat to estuarine and coastal fish
LaPatra (2006) found more frequent and severe populations. This highlights the need for manage-
epitheliocystis in fish exposed to sewage. This agrees ment to encompass both marine environments and
with findings in this study, with the highest rates of terrestrial catchments for the effective protection and
infection being recorded for the Manukau and sustainability of juvenile fish nurseries.
Mahurangi Harbours respectively.
Whether the gills of estuarine juvenile fish can Acknowledgements. This research was funded by the Auck-
recover from these histopathological changes is land University Doctoral Scholarship, the Stanley Wishart
unknown. However, studies on freshwater species Low Memorial Scholarship in Marine Science, and by the
(e.g. Fukuda 1983, Goldes et al. 1988) showed virtual New Zealand Ministry of Business, Innovation, and Enter-
prise (MBIE) projects (CO1832, CO1X0506, CO1X0907),
complete recovery from severe reactive hyperplasia
and NIWA's Coasts and Oceans Centre. Juvenile snapper
in less than 1 mo when the stimulus was removed were collected under a New Zealand Ministry of Fisheries
and adequate water quality was available. Nonethe- permit. The authors gratefully acknowledge the institutional
less, for estuarine fish, exposure to elevated levels of and technical support received from NIWA and the Univer-
sity of Auckland’s Leigh Marine Laboratory, including the
TSS can occur for extended periods during frequent
use of boats, laboratories and field support staff. We also
storms, and the effects could be cumulative with thank the 2 referees for their thoughtful and constructive
ongoing sediment pulses. Our results suggest that reviews.
physiological stress in fishes in response to increased
TSS/turbidity can decrease immunological compe-
LITERATURE CITED
tence and growth, and agree with prior studies
demonstrating negative relationships between fish ➤ Adams MS, Greeley MS, McHugh J, Noga EJ, Zelikoff JT
condition and water clarity in the field (e.g. Amara et (2003) Application of multiple sublethal stress indicators
al. 2007, Courrat et al. 2009, Zingel & Paaver 2010). to assess the health of fish in Pamlico Sound following
extensive flooding. Estuaries 26:1365−1382
➤ Al-Attar AM (2007) The influences of nickel exposure on
selected physiological parameters and gill structure in
CONCLUSIONS the teleost fish, Oreochromis niloticus. J Biol Sci 7:77−85
➤ Amara R, Meziane T, Gilliers C, Hermel G, Laffargue P
(2007) Growth and condition indices in juvenile sole
This is the first study to demonstrate that increased
Solea solea measured to assess the quality of essential
suspended sediment levels in temperate estuarine fish habitat. Mar Ecol Prog Ser 351:201−208
fish nurseries can have direct negative effects on Appleby JP, Scarrett DJ (1989) Physical effects of suspended
individual fish. Lower growth and nutritional status solids on marine and estuarine fish and shellfish with
special reference to ocean dumping: a literature review.
of juvenile snapper may potentially lead to increased
Can Tech Rep Fish Aquat Sci 1681:1−33
vulnerability to predation, physiological stress and
➤ Au DWT, Pollino CA, Wu RSS, Shin PKS, Lau STF, Tang JY
disease, in addition to lower overwinter survival and (2004) Chronic effects of suspended solids on gill struc-
subsequent recruitment to commercially exploitable ture, osmoregulation, growth, and triiodothyronine in
stocks (Francis 1994, Adams et al. 2003, Amara et al. juvenile green grouper Epinephelus coioides. Mar Ecol
Prog Ser 266:255−264
2007, Manning et al. 2014). Further research on the Barr NG (2007) Aspects of nitrogen metabolism in the green
effects of TSS/turbidity on larvae and juveniles alga Ulva: developing an indicator for seawater nitrogen
< 50 mm fork length, and tracking the development loading. PhD dissertation, University of Auckland
 252 Mar Ecol Prog Ser 539: 241–254, 2015

Bash J, Berman C, Bolton S (2001) Effects of turbidity and habitat modification from coal ash effluent on stream
suspended solids on salmonids. Center for Streamside macrobenthos: a synthesis. In: Krumholz LA (ed) The
Studies, University of Washington, Seattle, WA warmwater streams symposium: a national symposium
Beck MW, Heck KL Jr., Able KW, Childers DL and others on fisheries aspects of warmwater streams. Proc Am Fish
(2001) The identification, conservation, and manage- Soc Symp, 9−11 March 1980, Knoxville, TN. American
ment of estuarine and marine nurseries for fish and Fisheries Society, Southern Division, Lawrence, KS,
invertebrates. BioScience 51:633–641 p 241−249
➤ Berg L, Northcote TG (1985) Changes in territorial, gill flar- ➤ Francis MP (1994) Growth of juvenile snapper, Pagrus aura-
ing, and feeding behaviour in juvenile Coho salmon tus. N Z J Mar Freshw Res 28:201−218
(Oncorhynchus kisutch) following short-term pulses of Francis MP (1997) Condition cycles in juvenile Pagrus aura-
suspended sediment. Can J Fish Aquat Sci 42:1410−1417 tus. J Fish Biol 51:583−600
➤ Bergstedt LC, Bergersen EP (1997) Health and movements Fukuda Y (1983) Specific reactions of goldfish gills exposed
of fish in response to sediment sluicing in the Wind River, to linear alkylbenzene sulphonate. Jpn J Ichthyol 30:
Wyoming. Can J Fish Aquat Sci 54:312−319 268−274
Berry W, Rubinstein N, Melzian B, Hill B (2003) The bio- ➤ Gillanders BM (1997) Comparison of growth rates between
logical effects of suspended and bedded sediment estuarine and coastal reef populations of Achoerodus
(SABS) in aquatic systems: a review. Internal report, US viridis (Pisces: Labridae). Mar Ecol Prog Ser 146:283−287
Environmental Protection Agency, Narragansett, RI ➤ Gning N, Le Loc’h F, Thiawc OT, Aliaume C, Vidy G (2010)
Bhagwant S, Elahee KB (2002) Pathologic gill lesions in two Estuarine resources use by juvenile flagfin mojarra
edible lagoon fish species, Mulloidichthys flavolineatus (Eucinostomus melanopterus) in an inverse tropical estu-
and Mugil cephalus, from the Bay of Poudre d’Or, Mau- ary (Sine Saloum, Senegal). Estuar Coast Shelf Sci 86:
ritius. Western Indian Ocean. J Mar Sci 1:35−42 683−691
➤ Bruton MN (1985) The effects of suspensoids on fish. Hydro- ➤ Goldes SA, Ferguson HW, Moccia RD, Daoust PY (1988)
biologia 125:221−241 Histological effects of the inert suspended clay kaolin on
➤ Cottingham KL, Lennon JT, Brown BL (2005) Knowing when the gills of juvenile rainbow trout, Salmo gairdneri
to draw the line: designing more informative ecological Richardson. J Fish Dis 11:23−33
experiments. Front Ecol Environ 3:145−152 ➤ Grech A, Chartrand-Miller K, Erftemeijer P, Fonseca M and
➤ Courrat A, Lobry J, Nicolas D, Laffargue P and others (2009) others (2012) A comparison of threats, vulnerabilities and
Anthropogenic disturbance on nursery function of estu- management approaches in global seagrass bioregions.
arine areas for marine species. Estuar Coast Shelf Sci 81: Environ Res Lett 7:024006
179−190 Green EP, Short FT (2003) World atlas of seagrasses. Univer-
➤ Cyrus DP, Blaber SJM (1987a) The influence of turbidity on sity of California Press, Berkeley, CA
juvenile marine fishes in estuaries. Part 1. Field studies at ➤ Hack LA, Tremblay LA, Wratten SD, Lister A, Keesing V
Lake St. Lucia on the southeastern coast of Africa. J Exp (2007) Benthic meiofauna community composition at pol-
Mar Biol Ecol 109:53−70 luted and non-polluted sites in New Zealand intertidal
➤ Cyrus DP, Blaber SJM (1987b) The influence of turbidity on environments. Mar Pollut Bull 54:1801−1812
juvenile marine fishes in estuaries. Part 2. Laboratory Hecht T, van der Lingen CD (1992) Turbidity-induced
studies, comparisons with field data and conclusions. changes in feeding strategies of fish in estuaries. S Afr J
J Exp Mar Biol Ecol 109:71−91 Zool 27:95−107
➤ David V, Sautour B, Chardy P, Leconte M (2005) Long-term ➤ Helenius LK, Borg JPG, Nurminen L, Leskinen E, Lehtonen
changes of the zooplankton variability in a turbid envi- H (2013) The effects of turbidity on prey consumption
ronment: the Gironde estuary (France). Estuar Coast and selection of zooplanktivorous Gasterosteus aculea-
Shelf Sci 64:171−184 tus L. Aquat Ecol 47:349−356
➤ De Robertis A, Ryer CH, Veloza A, Brodeur RD (2003) Differ- Houde ED (1987) Fish early life dynamics and recruitment
ential effects of turbidity on prey consumption of pisci- and variability. Am Fish Soc Symp 2:17−29
vorous and planktivorous fish. Can J Fish Aquat Sci 60: ➤ Johansen JL, Jones GP (2013) Sediment-induced turbidity
1517−1526 impairs foraging performance and prey choice of plank-
➤ Duarte C (2002) The future of seagrass meadows. Environ tivorous coral reef fishes. Ecol Appl 23:1504−1517
Conserv 29:192−206 Kerr SJ (1995) Silt, turbidity and suspended sediments in the
➤ Edgar GJ (1990) The use of the size structure of benthic aquatic environment: an annotated bibliography and lit-
macrofaunal communities to estimate faunal biomass erature review. Southern Region Science & Technology
and secondary production. J Exp Mar Biol Ecol 137: Transfer Unit Technical Report TR-008, Ontario Ministry
195−214 of Natural Resources, North Bay
➤ Edgar GJ (1994) Observations on the size-structure of ➤ Lankford TE Jr, Targett TE (1997) Selective predation by
macrofaunal assemblages. J Exp Mar Biol Ecol 176: juvenile weakfish: post-consumptive constraints on
227−243 energy maximization and growth. Ecology 78:1049−1061
➤ Ellis J, Nicholls P, Craggs R, Hofstra D, Hewitt J (2004) ➤ Lohrer AM, Thrush SF, Hewitt JE, Berkenbusch K, Ahrens
Effects of terrigenous sedimentation on mangrove M, Cummings VJ (2004) Terrestrially derived sediment:
physiology and associated macrobenthic communities. response of marine macrobenthic communities to thin
Mar Ecol Prog Ser 270:71−82 terrigenous deposits. Mar Ecol Prog Ser 273:121−138
➤ Engström-Öst J, Mattila J (2008) Foraging, growth and habi- Lowe ML (2013) Factors affecting the habitat usage of estu-
tat choice in turbid water: an experimental study with arine juvenile fish in northern New Zealand. PhD disser-
fish larvae in the Baltic Sea. Mar Ecol Prog Ser 359: tation, University of Auckland
275−281 ➤ Macia A, Abrantes KGS, Paula J (2003) Thorn fish Terapon
Forbes AM, Magnuson JJ, Harrell DM (1981) Effects of jarbua (Forskal) predation on juvenile white shrimp
 Lowe et al.: Effects of turbidity on estuarine fish 253

Panaeus indicus H. Milne Edwards and brown shrimp diversity Report No. 137. Ministry for Primary Industries,
Metapenaeus monoceros (Fabricus): the effect of turbid- Wellington
ity, prey density, substrate type and pneumataphore ➤ Murphy HM, Jenkins GP, Hamer PA, Swearer SE (2012)
density. J Exp Mar Biol Ecol 291:29−56 Interannual survival of larval snapper (Chrysophrys
Mallatt J (1985) Fish gill structural changes induced by tox- auratus, Sparidae) is linked to diet breadth and prey
icants and other irritants: a statistical review. Can J Fish availability. Can J Fish Aquat Sci 69:1340−1351
Aquat Sci 42:630–648 Mutoro DB (2001) Life of a flatfish, the yellowbelly flounder,
➤ Manderson JP, Phelan BA, Meise LL, Stehlik AJ and others Rhombosolea leporina Günther, 1873, in Auckland’s
(2002) Spatial dynamics of habitat suitability for the sheltered waters. PhD dissertation, University of Auck-
growth of newly settled winter flounder Psudopleuro- land
nectes americanus in an estuarine nursery. Mar Ecol Newcombe E (2009) The nature and implications of varia-
Prog Ser 228:227−239 tion in a seaweed-epifauna-fish food chain. PhD disserta-
Manning NF (2013) Modelling the effects of turbidity on tion, University of Auckland
age-0 yellow perch (Perca flavescens) in the western ➤ Newcombe CP, Jensen JOT (1996) Channel suspended sed-
basin of Lake Erie. PhD dissertation, University of Tole- iment and fisheries: a synthesis for quantitative assess-
do, Toledo, OH ment of risk and impact. N Am J Fish Manage 16:
➤ Manning NF, Bossenbroek JM, Mayer CM, Bunnell DB, 693−727
Tyson JT, Rudstam LG, Jackson JR (2014) Modeling tur- ➤ Newcombe CP, MacDonald DD (1991) Effects of suspended
bidity type and intensity effects on the growth and star- sediments on aquatic ecosystems. N Am J Fish Manage
vation mortality of yellow perch. Can J Fish Aquat Sci 71: 11:72−82
1544−1553 ➤ Norkko A, Thrush SF, Hewitt JE, Cummings VJ and others
Mauchline J (1980) The biology of mysids and euphausiids. (2002) Smothering of estuarine sandflats by terrigenous
Adv Mar Biol 18:1−677 clay: the role of wind-wave disturbance and bioturbation
➤ McLeay DJ, Birtwell IK, Hartman GF, Ennis GL (1987) in site-dependent macrofaunal recovery. Mar Ecol Prog
Responses of Artic grayling, Thymallus articus, to acute Ser 234:23−41
prolonged exposure to Yukon placer mining sediment. Nowak BF, LaPatra SE (2006) Epitheliocystis in fish. J Fish
Can J Fish Aquat Sci 44:658−673 Dis 29:573–588
➤ Meager JJ, Solbakken T, Utne-Palm C, Oen T (2005) Effects Oldman JW, Swales A (1999) Mangemangeroa estuary
of turbidity on the reactive distance, search time, and for- numerical modelling and sedimentation. NIWA Client
aging success of juvenile Atlantic cod (Gadus morhua). Report ARC70224, National Institute of Water and
Can J Fish Aquat Sci 62:1978−1984 Atmospheric Research, Auckland
➤ Meijer A, Roholl PJM, Ossewaarde JM, Jones B, Nowak BF ➤ Omoregie E, Ajima MN, Keke RI, Wi ski K (2009) Effect of
(2006) Molecular evidence for association of Chlamy- single superphosphate fertilizer on survival and respira-
diales bacteria with epitheliocystis in leafy seadragon tory dynamics of Nile tilapia, Oreochromis niloticus
(Phycodurus eques), silver perch (Bidyanus bidyanus), (Actinopterygii: Perciformes: Cichlidae). Acta Ichthyol
and barramundi (Lates calcarifer). Appl Environ Micro- Piscat 39:103−110
biol 72:284−290 ➤ Orth RJ, Carruthers TJB, Dennison WB, Duarte CM and
➤ Mills EL, Ready RC, Confer JL (1984) Prey selection by others (2006) A global crisis for seagrass ecosystems.
young yellow perch: the influence of capture success, Bioscience 56:987−996
visual acuity, and prey choice. Trans Am Fish Soc 113: ➤ Parsons D, Sim-Smith C, Cryer M, Francis M and others
579−587 (2014) Snapper (Chrysophrys auratus): a review of life
Minello TJ, Zimmerman RJ, Martinee EX (1987) Fish preda- history and key vulnerabilities in New Zealand. N Z J
tion on juvenile brown shrimp, Penaeus aztecus Ives: Mar Freshw Res 48:256−283
effects of turbidity and substratum on predation rates. ➤ Partridge GJ, Michael RJ (2010) Direct and indirect effects
Fish Bull 85:59−70 of simulated calcareous dredge material on eggs and
Morrison MA (1990) Ontogenetic shifts in the ecology of the larvae of pink snapper Pagrus auratus. J Fish Biol 77:
parore, Girella triscupidata. MSc dissertation, University 227−240
of Auckland ➤ Redding JM, Schreck CB, Everest FH (1987) Physiological
➤ Morrison MA, Carbines G (2006) Estimating the abundance effects on Coho salmon and steelhead of exposure to sus-
and size structure of an estuarine population of the pended solids. Trans Am Fish Soc 116:737−744
sparid Pagrus auratus, using a towed camera during noc- Robb DHF, Kestin SC (2002) Methods used to kill fish: field
turnal periods of inactivity, and comparisons with con- observations and literature reviewed. Anim Welf 11:
ventional sampling techniques. Fish Res 82:150−161 269−282
➤ Morrison MA, Francis MP, Hartill BW, Parkinson DM (2002) ➤ Rowe DK, Dean TL (1998) Effects of turbidity on the feed-
Diurnal and tidal variation in the abundance of the fish ing ability of the juvenile migrant stage of six New
fauna of a temperate tidal mudflat. Estuar Coast Shelf Sci Zealand freshwater fish species. N Z J Mar Freshw Res
54:793−807 32:21−29
Morrison MA, Lowe ML, Parsons DM, Usmar NR, McLeod Scarsbrook M (2008) Saline water quality state and trends in
IM (2009) A review of land-based effects on coastal fish- the Auckland region. NIWA Client Report ARC 2008/
eries and supporting biodiversity in New Zealand. New 005, National Institute of Water and Atmospheric
Zealand Aquatic Environment Biodiversity Report No. Research, Auckland
37, Ministry for Primary Industries, Wellington ➤ Schlacher TA, Mondon JA, Connolly RM (2007) Estuarine
Morrison MA, Lowe ML, Grant CM, Reed J and others fish health assessment: evidence of wastewater impacts
(2014) Seagrass meadows as biodiversity and productiv- based on nitrogen isotopes and histopathology. Mar Pol-
ity hotspots. New Zealand Aquatic Environment Bio- lut Bull 54:1762−1776
 254 Mar Ecol Prog Ser 539: 241–254, 2015

➤ Servizi JA, Martens DW (1992) Sublethal responses of Coho Usmar N (2009) Ontogeny and ecology of snapper (Pagrus
salmon, Oncorhynchus kisutch, to suspended sediments. auratus) within an estuary, the Mahurangi Harbour. PhD
Can J Fish Aquat Sci 49:1389−1395 dissertation, University of Auckland
Sherk JA, O’Connor JM, Nuemann DA, Prince RD, Wood ➤ Usmar N (2012) Ontogenetic diet shifts in snapper (Pagrus
KV (1974) Effects of suspended and deposited sediments auratus: Sparidae) within a New Zealand estuary. N Z J
on estuarine organisms. Phase II. Final report no. 74-20, Mar Freshw Res 46:31−46
University of Maryland, Solomons Natural Resources ➤ Vinyard GL, O’Brien WJ (1976) Effects of light and turbidity
Institute, Solomons, MD on the reactive distance of blugill (Lepomis macro-
Sherk JA, O’Connor JM, Neumann DA (1975) Effects of sus- chirus). J Fish Res Board Can 33:2845−2849
pended sediments on estuarine environments. In: Cronin ➤ Volk EC, Wissmar RC, Simenstad CA, Douglas M, Eggers
LE (ed) Estuarine research, Vol 2: geology and engineer- DM (1984) Relationship between otolith microstructure
ing. Academic Press, New York, NY, p 541−558 and the growth of juvenile chum salmon (Oncorhynchus
➤ Sigler JW, Bjornn TC, Everest FH (1984) Effects of chronic keta) under different prey rations. Can J Fish Aquat Sci
turbidity on density and growth of steelheads and Coho 41:126−133
salmon. Trans Am Fish Soc 113:142−150 ➤ Wenger AS, Johansen JL, Jones GP (2011) Suspended sedi-
➤ Sogard SM (1992) Variability in growth rates of juvenile ment impairs habitat choice and chemosensory discrimi-
fishes in different estuarine habitats. Mar Ecol Prog Ser nation in two coral reef fishes. Coral Reefs 30:879−887
85:35−53 ➤ Wenger AS, Johansen JL, Jones GP (2012) Increasing sus-
➤ Sudo H, Azeta M (2001) The microhabitat and size of gam- pended sediment reduces foraging, growth and condi-
marid species selectively predated by young red sea tion of a planktivorous damselfish. J Exp Mar Biol Ecol
bream Pagrus major. Fish Sci 67:389−400 428:43−48
➤ Sutherland AB, Meyer JL (2007) Effects of increased sus- ➤ Wenger AS, McCormick M, Endo GGK, McLeod IM,
pended sediment on growth rate and gill condition of two Kroon FJ, Jones GP (2014) Suspended sediment prolongs
southern Appalachian minnows. Environ Biol Fishes 80: larval development in a coral reef fish. J Exp Biol 217:
389−403 1122−1128
➤ Swales A, Williamson BR, Van Dam LF, Stroud MJ, McGlone ➤ Wilber DH, Clarke DG (2001) Biological effects of suspen-
MS (2002) Reconstruction of urban stormwater contami- ded sediments: a review of suspended sediment impacts
nation of an estuary using catchment history and sedi- on fish and shellfish with relation to dredging activities
ment profile dating. Estuaries 25:43−56 in estuaries. N Am J Fish Manage 21:855−875
➤ Syvitski JPM, Vörösmarty CJ, Kettner AJ, Green P (2005) Willis TJ, Handley SJ, Chang FH, Law CS and others (2007)
Impact of humans on the flux of terrestrial sediment to Climate change and the New Zealand environment:
the global coastal ocean. Science 308:376−380 NIWA Client Report DOC 08305, National Institute of
➤ Thrush SF, Hewitt JE, Cummings VJ, Ellis JI, Hatton C, Water and Atmospheric Research, Auckland
Lohrer A, Norkko A (2004) Muddy waters: elevating sed- Wong CK, Pak IAP, Liu XJ (2013) Gill damage to juvenile
iment input to coastal and estuarine habitats. Ecol Envi- orange-spotted grouper Epinephelus coioides (Hamilton,
ron 2:299−306 1822) following exposure to suspended sediments.
Tricklebank KA (1997) Organochlorine pesticide residues Aquacult Res 44:1685−1695
and histopathology in the damselfish Parma microlepis, ➤ Zingel P, Paaver T (2010) Effects of turbidity on feeding of
from reefs in New South Wales. PhD dissertation, Uni- the young-of-the-year pikeperch (Sander lucioperca) in
versity of Sydney fishponds. Aquacult Res 41:189−197

Editorial responsibility: Anna Pasternak, Submitted: March 18, 2015; Accepted: September 23, 2015
Moscow, Russian Federation Proofs received from author(s): November 9, 2015