Upload
others
View
14
Download
0
Embed Size (px)
Citation preview
MURDOCH RESEARCH REPOSITORY
This is the author’s final version of the work, as accepted for publication following peer review but without the publisher’s layout or pagination.
The definitive version is available at http://dx.doi.org/10.1111/gcb.12444
Beatty, S.J., Morgan, D.L. and Lymbery, A.J. (2014) Implications of climate change for potamodromous fishes. Global Change
Biology, 20 (6). pp. 1794-1807.
http://researchrepository.murdoch.edu.au/22186/
Copyright: © 2013 John Wiley & Sons Ltd.
It is posted here for your personal use. No further distribution is permitted.
Acc
epte
d A
rtic
le
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/gcb.12444 This article is protected by copyright. All rights reserved.
Received Date: 09-May-2013 Revised Date : 25-Sep-2013 Accepted Date: 03-Oct-2013 Article type : Primary Research Articles Implications of climate change for potamodromous fishes
Running head: Potamodromous fishes and climate change
STEPHEN J. BEATTY*, DAVID L. MORGAN and ALAN J. LYMBERY
Freshwater Fish Group & Fish Health Unit, School of Veterinary and Life Sciences, Murdoch
University, South St Murdoch, Western Australia
*Correspondence: Stephen J. Beatty, tel. +618 9360 2813, fax +618 9360 7512, e-mail:
Keywords: freshwater fishes, Mediterranean climate, surface flow decline, groundwater
reduction, secondary salinisation, south-western Australia, aquatic refuge
Primary Research Article
Abstract
There is little understanding of how climate change will impact potamodromous freshwater
fishes. Since the mid-1970’s, a decline in annual rainfall in south-western Australia (a
globally recognised biodiversity hotspot) has resulted in the rivers of the region undergoing
severe reductions in surface flows (~50%). There is universal agreement amongst Global
Climate Models that rainfall will continue to decline in this region. Limited data are
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
available on the movement patterns of the endemic freshwater fishes of south-western
Australia or on the relationship between their life-histories and hydrology. We used this
region as a model to determine how dramatic hydrological change may impact
potamodromous freshwater fishes. Migration patterns of fishes in the largest river in south-
western Australia were quantified over a four year period and were related to a number of key
environmental variables including discharge, temperature, pH, conductivity and dissolved
oxygen. Most of the endemic freshwater fishes were potamodromous, displaying lateral
seasonal spawning migrations from the main channel into tributaries, and there were
significant temporal differences in movement patterns between species. Using a model
averaging approach, amount of discharge was clearly the best predictor of upstream and
downstream movement for most species. Given past and projected reductions in surface flow
and groundwater, the findings have major implications for future recruitment rates and
population viabilities of potamodromous fishes. Freshwater ecosystems in drying climatic
regions can only be managed effectively if such hydro-ecological relationships are
considered. Proactive management and addressing existing anthropogenic stressors on
aquatic ecosystems associated with the development of surface and groundwater resources
and land use is required to increase the resistance and resilience of potamodromous fishes to
ongoing flow reductions.
Introduction
Freshwater ecosystems are in crisis throughout the world. Aquatic habitats associated with
65% of global river flow are moderately or highly threatened, principally by water resource
development and pollution (Dudgeon et al., 2006; Vörösmarty et al., 2010). Freshwater
fishes are one of the most threatened faunal groups and are expected to be among the most
severely impacted by climate change, due to it adding to a wide range of current and future
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
anthropogenic stressors (Ficke et al., 2007; Palmer et al., 2008; Olden et al., 2010; Comte et
al., 2013). Freshwater fishes in Mediterranean climatic regions are particularly imperilled
and while flow reduction due to water extraction and climate change is recognised as a major
threat (Hermoso & Clavero, 2011; Maceda-Veiga, 2013), its impacts are difficult to quantify
due to the complexity of abiotic and biotic interactions (Clavero et al., 2010).
The challenge of conserving critical habitat for freshwater species is exacerbated by life-
cycles that often involve movement between habitats for different life-history stages
(Magoulick & Kobza, 2003) and the linear nature of rivers, whereby upstream activities
influence downstream habitat quality and availability (e.g. Gorman & Karr, 1978; Allan et
al., 1997). Seasonal migrations in potamodromous fishes are critical for completing life-
cycles; particularly in accessing spawning and nursery habitats and refugia (Reynolds, 1983;
O’Connor et al., 2005; Munz & Higgins, 2013). Understanding refuge use and species
mobility is important for enhancing conservation efforts; particularly under drought or flow
reduction scenarios (Magoulick & Kobza, 2003; Morán-López et al., 2012; Chessman, 2013).
In comparison to the amount of research on diadromous fishes (e.g. Crozier et al., 2008;
Lassalle & Rochard, 2009; Finstad & Hein, 2012; Piou & Prévost, 2013), there is little
information on how flow reductions due to climate change will impact potamodromous
species (but see Buisson et al., 2008; Clews et al., 2010). The vast majority of research
examining the impacts of climate change on freshwater fishes has focussed on salmonids and
other cold water species, those with commercial or recreational value, and those in the
Northern Hemisphere (e.g. Comte et al., 2013). Broadening our understanding of these
impacts to other regions and to non-diadromous fishes with little commercial or recreational
value is needed for a more complete understanding of the vulnerability of freshwater
ecosystems to climate change. As predicted globally (e.g. Thieme et al., 2010), climate
change is expected to exacerbate the current anthropogenic stressors on Australian freshwater
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
fishes (Morrongiello et al., 2011a, b) and cause major range contractions of several species
(Balcombe et al. 2011; Bond et al. 2011).
South-western Australia, a global biodiversity hotspot (Myers et al., 2000), has a
Mediterranean climate that is isolated from high rainfall temperate and tropical regions by
extensive arid zones. Although the region has only 11 native freshwater fish species, nine are
regionally endemic (Morgan et al., 1998; Allen et al., 2002). These species are highly
imperilled, principally due to the synergistic impacts of a number of stressors such as
secondary salinisation (Morgan et al., 2003; Beatty et al., 2011), instream barriers (Morgan &
Beatty, 2006; Beatty et al., 2007), and the impacts of introduced species (Morgan et al.,
2004; Marr et al., 2010; Beatty & Morgan, 2013).
Since the mid-1970’s, a considerable climatic shift has occurred in south-western
Australia, resulting in a reduction of ~16% in annual rainfall corresponding to an ~50%
reduction in the annual volume of stream flow (Suppiah et al., 2007; Petrone et al., 2010;
Silberstein et al., 2012). There is an unusually uniform consensus across global climatic
models (GCMs) for this region as they all project this drying trend to continue, with median
reductions of an additional ~8% in annual rainfall and ~25% (range of 10-42%) in annual
stream flow expected by 2030 (Suppiah et al., 2007; Silberstein et al., 2012). Furthermore,
reductions in fresh groundwater levels of >10m are predicted in many areas in the region by
2030 (Barron et al., 2012). Quantifying the potential impacts of this dramatic hydrological
change on freshwater fishes of the region is currently hampered by a lack of ecological data
on these species (Beatty et al., 2010, Morrongiello et al., 2011a).
The two non-endemic native freshwater fishes in south-western Australia are the
galaxiids Galaxias maculatus and Galaxias truttaceus that may have diadromous (marine
larval phase) or land-locked populations; the latter utilising coastal lacustrine environments
for larval development (Morgan 2003; Chapman et al., 2006). While almost nothing is
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
known on the migration patterns of the nine endemic freshwater fishes, they are not
diadromous and are known to breed during the major rainfall and surface flow periods (Pen
& Potter, 1991; Morgan et al., 1998). Given the timing of their breeding period, they may be
highly reliant on seasonal surface discharge and could be potamodromous in order to breed
within intermittent lotic systems that are common to the region.
Secondary salinisation is a major global threat that is expected to be amplified by climate
change (Cañedo-Argüelles et al., 2013). This process is severe in south-western Australia,
with most of the major rivers salinised due to the mobilisation of saline groundwater from
large scale land-clearing in their upper catchments (Mayer et al., 2005). While rising saline
groundwater is a major cause of secondary salinisation (Cañedo-Argüelles et al., 2013), fresh
groundwater aquifers can have a dilution effect in salinised rivers. The fresh south-west
Yarragadee Aquifer (SWYA) underlies a considerable area of south-western Australia and
can contribute to between 30-100% of the baseflow riverine discharge of the Blackwood
River (the largest of the region) during the prolonged dry-season (Strategen, 2006). This
contribution significantly dilutes the main channel during baseflow, when most tributaries fail
to flow or dry completely (Del Borrello, 2008; Beatty et al., 2010).
Freshwater fishes of the region are stenohaline and have undergone major downstream
range contractions away from the most salinised inland habitats towards those regions
underlain by fresh aquifers (Morgan et al., 2003; Beatty et al., 2011). Fresh groundwater
discharge maintains habitat connectivity for at least one species (Beatty et al., 2010) and is
likely to play a key role in maintaining fresh baseflow refuge for other species. As the
groundwater level and surface discharge during the annual dry season are positively related to
the previous year’s rainfall in the Blackwood River (Golder & Associates, 2008), both will be
affected by projected rainfall reductions (Barron et al., 2012; Silberstein et al., 2012).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Understanding the movement patterns and habitat utilisation of freshwater fishes and
their relationships to hydrology and physicochemical variables will be vital in understanding
how hydrological change will affect future population viabilities. The current study aimed to
elucidate the spatial and temporal movement patterns of freshwater fishes, their life-histories,
and their relationships with environmental variables in a drying climatic region, using the
Blackwood River as a model system. The study hypothesised that: 1) south-western
Australian freshwater fishes are potamodromous, based on existing information on their
reproductive biology; 2) both fresh surface water and groundwater discharge are important in
maintaining remnant populations; and 3) ongoing reductions in rainfall and discharge due to
climate change will negatively impact potamodromous fishes.
Materials and methods
Sampling regime
Spatial and temporal patterns in the movement and demographics of native freshwater fishes
were determined in the Blackwood River main channel and four of its tributaries between
October 2005 and December 2009 (Fig. 1). Three main channel sites included one
downstream of the major zone of groundwater discharge from the SWYA (Denny Rd), one at
the upstream margin of the SWYA discharge zone (Milyeannup Pool) and one upstream of
the SWYA discharge zone (Jalbarragup) (Fig. 1). The tributary sites included one
perennially flowing system (Milyeannup Brook) that is maintained by groundwater from the
SWYA, and three seasonally intermittent systems (Layman Brook, Rosa Brook and McAtee
Brook) (Fig. 1).
The seven sites were sampled on a regular basis between October 2005 and December
2009. In each sampling event (n = 30 within the 50 month period), all sites that had flowing
water at the time were sampled. All seven sites were sampled during the major flow period
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
between August and December (i.e. late winter through to early summer). However, as three
of the four tributaries naturally dry during late summer through autumn (i.e. February to
May), sampling did not occur in those systems during that period due to a lack of flow. For
each sampling event, two fyke nets (each net having a width of 11.2 m, including two 5 m
wings and a 1.2 m wide mouth, depth of 0.8 m, length of 5 m with two funnels all comprised
of 2 mm woven mesh) were simultaneously set for three consecutive 24 hour periods. One
net faced upstream, to determine downstream movements of fishes, and the other faced
downstream, to determine upstream movements of fishes (following Beatty et al., 2010).
As it was not always possible to set the nets across the entire wetted channel, the proportion
of the total stream cross-section that was covered by each fyke was determined in order to
standardise captures (see statistical analysis). All fish captured were identified to species,
enumerated, measured to the nearest 1-mm total length (TL) and released either upstream or
downstream of the capture point, depending on their original movement direction.
Environmental variables
One of the most important steps in modelling species distributions is careful variable
selection, particularly when using predictors when there is a degree of a priori expectation of
ecological relevance (Elith & Leathwick, 2009; Burnham & Anderson, 2010). Many
physical, geographical and ecological characteristics of the study streams were very similar.
The catchments were all effectively completely forested by natural vegetation and the
streamlines had qualitatively similar, near-pristine riparian zones (CENRM, 2005; Strategen,
2006). Furthermore, all streams were highly consistent in terms of overall channel
morphology, slope, altitude, dissection and soil type (CENRM, 2005; Strategen, 2006).
Finally, all four study sites were situated within 14 km of each other and each was located at
a similar distance from the main channel of the Blackwood River with no artificial instream
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
barriers existing in the streams (Figure 1). Therefore, given the considerable spatial and
temporal sampling effort undertaken, we focussed on other environmental variables that are
commonly associated with freshwater fish distributions and movement. These were
temperature, conductivity, pH, dissolved oxygen, and discharge (Matthews, 1998). In order
to determine the spatial and temporal patterns in water temperature and their relationship to
air temperature, data loggers (TinytagTM Gemini Data Loggers, Chichester, West Sussex,
UK) were placed in situ in all the tributary sites and two main channel sites and were
programmed to log water temperature every three hours. Temperature loggers were
downloaded periodically and the water temperature regimes of the various sites were
graphically compared to maximum daily air temperature obtained from the nearest Australian
Bureau of Meteorology weather station (Bridgetown). On each sampling event at each site,
instantaneous temperature, conductivity, pH, and dissolved oxygen were measured using a
handheld water quality meter (Oakton PCD650) in the middle of the water column at three
locations and a mean (± 1 SE) determined.
Flow velocity was measured using a hand held flow meter (Global F101) and
instantaneous discharge measurements were determined in all tributary sites during each
sampling event by undertaking cross-sectional depth-flow profiles (measured every 0.5m)
and applying the formula: Dinst.=VxDxW where Dinst. = instantaneous discharge m3.sec-1, V =
mean water velocity (m.sec-1) over the cross-section, D = mean depth (m) over the cross-
section, W = width (m) of the cross-section.
Statistical analysis
In order to determine the spatial and temporal patterns of movements and demographics of
the five most commonly captured native freshwater fishes in the tributary sites, i.e. the
western minnow Galaxias occidentalis, western mud minnow Galaxiella munda, western
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
pygmy perch Nannoperca vittata, Balston’s pygmy perch Nannatherina balstoni, and
nightfish Bostockia porosa (see Results), the mean (±1 SE) numbers of each species captured
moving upstream and downstream over the three 24 hour periods (numbers were standardised
from the proportion of cross-section of the stream covered by each fyke to assume 100%
coverage) in each sampling event at each site were determined. To elucidate population
structure of each species and best illustrate movement patterns in relation to life-history stage,
the monthly length-frequency distribution (separated on movement direction) of the above
freshwater fishes throughout the study were plotted separately for Milyeannup Brook (G.
occidentalis, N. balstoni, N. vittata, and G. munda), Rosa Brook (G. occidentalis, N. vittata,
and B. porosa) and Layman Brook (B. porosa).
To test the effects of each species captured and the sampling month on fish captures, we
used general linear models (GLMs) with a repeated measures design. The log10(x+1)
transformed number of fish captured in the fykes moving in either an upstream or
downstream direction at each sampling event was used as a response variable, with species as
a between-subject fixed effect, months as a within-subject fixed effect and tributary nested
within species as a random effect.
The relationships between the upstream and downstream movement of the five native
species in tributaries and the prevailing environmental variables in those systems were
explored by first determining the mean of each environmental variable and the overall mean
upstream and downstream movements of each species in each system over the major flow
period in each year (i.e. the period that all four tributaries were flowing between August to
December). As excessive flooding of the tributary sampling sites occurred in August 2009,
they were excluded from the above analysis.
Bivariate correlations (using Pearson’s correlation coefficient) were calculated for each
pair of environmental variables (all except pH log10 transformed) to examine possible inter-
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
relationships. Mean velocity was consistently highly correlated with discharge and was
therefore excluded from subsequent analyses; no other variable pairs showed correlations >
0.6 (Table S1). To determine which environmental variables explained most of the variation
in upstream and downstream migrations of each fish species during the major flow period, we
used GLMs, with a multimodel inference approach (Burnham & Anderson, 2010). For each
fish species and for each movement direction, a global model of number of fish captured
(log10(x+1) transformed) against all environmental predictor variables was fitted using the R
program (R Development Core Team, 2013). The global model was then used to generate a
set of models containing all combinations of predictor variables, using the R package MuMIn
(Bartoñ, 2013). Models were ranked by the Akaike Information Criterion, corrected for small
sample size (AICc) and model averaging performed across those models that were within four
AICc values of the best model (Burnham & Anderson, 2010), using MuMIn. The relative
importance of each environmental predictor variable was determined by summing the Akaike
likelihood weights across all models within the top-ranked set; this gives the selection
probability that a given variable will appear in the AIC-best model (Burnham & Anderson,
2010). For each fish species and each movement direction, we report the relative importance
of all of the environmental predictor variables, along with the model-averaged parameter
estimates and their standard errors.
As estimates of parameters and variable importance from model averaging can be
adversely affected by collinearity among predictors (Murray & Conner, 2009; Freckleton,
2011), we attempted to account for this in three ways. First, only predictor variables with
bivariate correlations ≤ 0.6 with all other predictors were used in the analysis (see Table S1).
Second, for each predictor we calculated the variance inflation factor (VIF), which represents
the overall correlation with all other predictors, using the R package HH (Heiberger, 2013).
As a general rule of thumb, VIF’s > 10 are thought to indicate excessive collinearity (Myers,
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
1990; although see O’Brien, 2007 for a contrary view). Finally, we also used a hierarchical
partitioning approach to determine variable importance, as this has been shown in some
simulations to outperform model averaging when predictor variables are correlated (Murray
& Conner, 2009; although see Smith et al., 2009). R2 values were hierarchically partitioned
to determine the proportion of variance explained independently and jointly by each variable
(Chevan & Sutherland, 1991; Mac Nally, 2002), and the independent effect of each predictor
was calculated as the percentage of total independent contributions, using the R package
hier.part (Walsh & Mac Nally, 2013). The significance of independent effects was assessed
with Z scores from a comparison of observed effects with those obtained after 500
randomisations of the data, using hier.part.
Results
Environmental variables
Water and air temperatures were highly associated at all sites (Fig. 2). Temperatures in the
main channel sites were much higher than tributary sites during summer and autumn, with the
greatest differences occurring between main channel sites and the perennial (groundwater
maintained) Milyeannup Brook, which experienced a less variable temperature regime than
other tributaries. For example, between late December and March (i.e. summer to early
autumn) in 2006 and 2008, Milyeannup Brook was generally >1°C cooler than the non-
perennial Rosa Brook and McAtee Brook, and >6°C cooler than the main channel sites.
Between May and August (i.e. late autumn to late winter) 2006, Milyeannup Brook was ~2°C
warmer than the non-perennial systems and was similar in temperature to the main channel
sites (Fig. 2).
Mean conductivity in the main channel sites increased in an upstream direction, ranging
from 3564 to 4955 µS.cm-1 at the Denny Rd and Jalbarragup sites, respectively (Table S2).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
In contrast, the tributary sites were always fresher with mean values that did not exceed 500
µS.cm-1 (Table S2). Mean dissolved oxygen at all sites was >6.00 mg.l-1 and pH in the
tributary sites was slightly acidic (means from 6.04-6.67), whereas pH was near neutral in the
main channel sites (means from 7.18-7.33) (Table S2).
Species captures and spatial distributions
Six of the nine endemic freshwater fishes of south-western Australia, along with four
estuarine fishes, three introduced fishes and one native anadromous agnathan were captured
during the study. The six endemic freshwater fish species were the galaxiids G. occidentalis
and G. munda, the percichthyids N. vittata, N. balstoni and B. porosa and the plotosid
Tandanus bostocki (Table S3).
Galaxias occidentalis was the most commonly recorded species, present in 92.2% of all
sites sampled in all months, with N. vittata, B. porosa, and T. bostocki recorded at similar
frequencies (~66-67%), N. balstoni at 38.5%, and G. munda (16.5%) being the least
encountered (Table S3). Stark contrasts existed in captures of the various species between
main channel and tributary sites, with five of the six endemic freshwater species more
commonly encountered in the tributary sites (Table S3). The most difference in habitat usage
was recorded for G. munda (32.4 cf 2.7% in tributary and main channel samples,
respectively) (Table S3). Galaxiella munda was recorded in tributary sites that were cooler,
fresher, and had lower mean discharge than habitats occupied by all other species (Table S3).
Galaxias occidentalis was recorded in similar frequency between the two habitat types (i.e.
95.2 and 90.3%, in tributary and main channel sites, respectively) whereas T. bostocki was
much more frequently recorded in main channel (99.0%) compared with tributary (30.5%)
sites (Table S3).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Patterns of species movements
There were clear spatial and temporal patterns of movements of the five endemic freshwater
fish species that were most commonly encountered in the tributaries (i.e. G. occidentalis, G.
munda, N. balstoni, N. vittata and B. porosa). Upstream fish movements were significantly
affected by species (F = 8.25, P = 0.0003) and month of sampling (F = 6.14, P < 0.0001),
with a significant interaction between species and month (F = 2.51, P < 0.0001).
Downstream movement did not quite differ significantly among species (F = 2.79, P = 0.05),
but was significantly affected by the month of sampling (F = 8.63, P < 0.0001), with a
significant interaction between species and month (F = 3.27, P < 0.0001).
Details of the habitat utilisation, spatial and temporal movements and population
structure of the five fish species in each sampling month at each site are shown in the
supporting information (Table S3, Figs S1-S13). Schematic diagrams of the movement
patterns of adults and juveniles of the two species of galaxiids and three species of
percichthyids (Figs 3 and 4) are based on the data presented in the Supporting Information.
While interspecific variation existed in migration patterns (spatially and temporally), all but
G. munda moved laterally between the main channel of the Blackwood River and its
tributaries (Figs 3 and 4).
Downstream movements of G. occidentalis occurred predominately within tributary sites
between August and December (i.e. late winter to early summer), with an upstream
directional movement occurring in main channel sites between June and August (i.e. winter)
(Figs 3 and S1). Downstream movement of juveniles (<40 mm TL) generally occurred
annually between October to December, with a notable one to two month earlier movement
occurring in the perennial Milyeannup Brook compared to Rosa Brook (Figs 3, S2 and S3).
Larger individuals (>120 mm TL) were also much more prevalent in Milyeannup Brook
compared with the seasonally flowing systems. Unlike all other species, there were no strong
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
movement patterns associated with G. munda within the tributaries (Figs 3 and S4).
However, downstream movement of juveniles (<35 mm TL) were recorded during late spring
(October and November) in most years in Milyeannup Brook (Fig. S5). Notably, G. munda
was the only species found to be effectively absent from the main channel (Fig. 3 and Table
S3).
Nannatherina balstoni only utilised Milyeannup Brook as a breeding and nursery area
and was recorded very infrequently in all other tributaries (Figs 4 and S6). A clear annual
downstream movement of juveniles (<50 mm TL) occurred in that system between
November and December (i.e. late spring to early summer) (Fig. S7). Although this species
was not abundant in the main channel sites (Fig. 4 and Table S3), it was recorded
occasionally at the Denny Rd and Milyeannup Pool sites; both located downstream of the
confluence of Milyeannup Brook (Figs 1 and S6). Nannoperca vittata migrated within all
tributaries (aside from Layman Brook), with both upstream and downstream directional
movements of the species occurring between August and December (Figs 4 and S8). The
species was also recorded to lesser degree moving within the main channel sites, with the vast
majority moving in a downstream direction (Fig. S8). Juvenile (<30 mm TL) N. vittata were
recorded moving downstream within tributaries and the timing of that movement varied
between systems. For example, it occurred mostly in March (with some November and
December movement of <20 mm TL fish) in the perennial Milyeannup Brook, compared
with August to November in the ephemerally flowing Rosa Brook (Figs S9 and S10). Large
downstream movements of juvenile (<30 mm TL) B. porosa occurred in a number of
tributaries between November and December (Figs 4 and S11) and were most prevalent in
Milyeannup, Rosa and Layman Brooks, where it was recorded in all years aside from 2009 in
Layman Brook (Figs S11, S12, S13).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Relationships between environmental variables and movement patterns
The influence of environmental variables on the average upstream and downstream
movements of the five fish species over the major flow periods in the four tributaries between
2005 and 2009 are shown in Tables 1 and 2. Variance inflation factors were relatively low
for all variables (temperature 3.11; conductivity 3.31; pH 1.95; dissolved oxygen 3.29;
discharge 1.83) and both model averaging and hierarchical partitioning analyses gave very
similar indications of variable importance (Tables 1 and 2). This suggests that these
measures were not biased by collinearity among predictors. Discharge ranked as the most
important predictor variable in four of the five species for both upstream and downstream
movement, being included in between 43-100% of the best fitting models (Tables 1 and 2).
In particular, discharge explained a substantial proportion of the upstream movements of the
three most widespread and abundant species, N. vittata, B. porosa and G. occidentalis (Table
1). With the exception of one case (i.e., downstream movement in B. porosa), more fish were
captured with greater discharge levels. Temperature also figured prominently as an important
predictor variable, particularly for N. vittata, with more fish captured at lower temperatures.
Of the other variables, pH was positively associated with upstream movements, and dissolved
oxygen was negatively associated with downstream movements of B. porosa.
Discussion
There has been limited research on potamodromous fishes that migrate laterally between
main channel and low order streams, compared to those that undertake longitudinal
migrations within the main streams of large river networks (e.g. O’Connor et al. 2005; Dugan
et al., 2010) or between lentic and lotic systems (e.g. Barthel et al., 2008; Conallin et al.,
2011). The new recruits of five of the six endemic freshwater fishes found within the study
area were recorded almost exclusively within tributaries, and their downstream movement
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
into main channel habitats occurred for four of the five species during periods of declining
surface flows (i.e., late spring and early summer). As with other climatic regions with
pronounced dry seasons, most lower order tributaries in south-western Australia either dry
completely or consist of separated, often isolated refuge pools during summer and autumn.
Therefore, our findings strongly suggest that most species are potamodromous and undergo
lateral spawning migrations into lower order tributaries during the seasonal flow period.
Discharge was clearly the most important environmental variable related to the upstream and
downstream movement of most species and therefore our findings have considerable
implications in understanding how life-cycles of potamodromous fishes in drying temperate
regions may be impacted by climate change.
Influence of hydrology on life-histories
Hydrologic regime has been shown to significantly influence the life-history strategies of
freshwater fishes in lotic systems over broad biogeographic scales (Mims & Olden, 2012;
Chessman, 2013; Sternberg & Kennard, 2013). Due to the highly seasonal and predictable
flow regimes of aquatic systems in south-western Australia, its freshwater fishes would be
expected to have periodic life history strategies (Mims & Olden, 2012). Consistent with this
expectation is that most species have annual spawning periods that coincide with peak flow in
winter and spring (e.g. Pen & Potter, 1991; Morgan et al., 1998; this study). Periodic species
are also characterised by having large body size, late maturation, and high fecundity. Most of
the freshwater fishes of south-western Australia, however, have many traits that would
classify them as opportunistic (r strategists), such as small body sizes (<190 mm TL) and
early maturation (most in the first or second year of life) (Morgan et al., 1998). This may be
due to these species having evolved in relatively small, seasonally intermittent, naturally
unproductive aquatic systems typical of this region (Bunn & Davies, 1990), which would
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
limit their maximum sizes and reproductive investment. Despite having several opportunistic
life-history traits, the movements of most species were undoubtedly associated with spawning
and recruitment and they would be therefore vulnerable to reductions in amount of surface
discharge.
Rivers that are free-flowing in relatively undeveloped basins are predicted to be resilient
to climate change compared with those that are more highly regulated (Palmer et al., 2008).
Artificial instream barriers can have profound, wide-ranging impacts on aquatic ecosystems,
particularly on migratory fishes (Bunn & Arthington, 2002; Katopodis & Aadland, 2006).
Most other studies relating to anthropogenic impacts on the migrations of potamodromous
fishes have focussed on the impacts of barriers and river connectivity (e.g. Baumgartner &
Harris, 2007; Mallen-Cooper & Brand, 2007; Branco et al., 2012). While our study systems
are free from artificial barriers, our findings suggest that the numerous natural instream
barriers that exist elsewhere throughout the region may also be disrupting the life-cycles of
these potamodromous fishes by restricting access to upstream spawning habitats exacerbating
the impacts of past and projected declines in surface discharge.
Several species in the current study are known to also occupy natural and artificial lentic
systems (Morgan et al., 1998; Beatty et al. 2013) and this suggests they have a degree of
habitat flexibility. This could help at least some populations survive the expected ongoing
reductions in flow period and surface discharge due to climate change by utilising main
channel lotic habitats or artificial lentic habitats for spawning and recruitment. Nonetheless,
these projected hydrological changes will undoubtedly reduce recruitment to some degree
through reductions or loss of spawning habitats in lower order streams.
Galaxiids are generally known to migrate for spawning purposes (McDowall &
Frankenberg, 1981; Baker & Montgomery, 2001; Baker & Hicks, 2003; Morgan, 2003) and
G. occidentalis in the current study showed the strongest movement patterns; migrating
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
laterally into all tributaries to spawn, with this movement positively associated with amount
of stream discharge during the annual flow period. As opposed to galaxiids, pygmy perches
are generally not known to be strong migrators (e.g. Hammer et al., 2009), although
movement between water bodies has been linked to flooding events (Morgan et al., 1995;
Hughes et al., 1999; Knight & Arthington, 2008). We demonstrated that the two pygmy
perches were also potamodromous and that the upstream spawning movements of N. vittata
and the other percichthyid, B. porosa, were positively associated with amount of surface
discharge. Furthermore, hydrological seasonality would explain why N. balstoni, the most
threatened species in the study, only breeds within the groundwater maintained perennial
Milyeannup Brook, despite other tributaries appearing to provide equally suitable habitat
when they are flowing. The onset of breeding by N. balstoni is earlier than others in this
study (i.e. from early winter cf late winter and spring (Morgan et al., 1995, 1998)), requiring
its spawning habitat to be available at a time when seasonal tributaries are yet to flow.
Ongoing changes in both the amount and timing of discharge may therefore have serious
implications for this and other potamodromous species.
Declines in refuge habitats due to climate change
Understanding refuge habitat requirements is crucial in assessing impacts of climate change
on freshwater fishes in seasonal systems (Hodges & Magoulick, 2011; Morán-López et al.,
2012; Davis et al., 2013). System size and the volume and periodicity of discharge, along
with a number of other abiotic characteristics, such as anthropogenic pollution, have been
demonstrated to influence refuge availability and use by freshwater fishes (e.g. Magoulick &
Kobza, 2003; Hodges & Magoulick, 2011; Morán-López et al., 2012). Groundwater
maintained aquatic refuges can be thought of as ‘evolutionary refugia’ and are extremely
important in maintaining endemic aquatic species (Davis et al., 2013). Our study identified
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
that the key refuge habitats for freshwater fishes during the dry season in the Blackwood
River were the secondarily salinised main channel and a fresh perennial tributary; both reliant
on fresh groundwater.
Pollution caused by human activities is one of the major global stressors on riverine
ecosystems (Vörösmarty et al., 2010). Secondary salinisation of rivers is a major threat to
aquatic ecosystem health and climate change is expected to amplify the problem, particularly
in climates that experience prolonged periods of drought (Cañedo-Argüelles et al., 2013).
Salinisation in south-western Australia has caused fundamental ecosystem change, including
widespread range contractions of many stenohaline freshwater fishes and concomitant
colonisation by estuarine euryhaline species (Morgan et al., 2003; Mayer et al., 2005; Beatty
et al., 2011). Reduced rainfall is projected to cause severe reductions in surface flow and
fresh groundwater levels in many catchments (Barron et al., 2012; Silberstein et al., 2012),
which may cause remnant fresh perennial tributaries to be lost and further increase the
salinity in main channel refuges to levels that may exceed the tolerances of several species
(Beatty et al., 2011). In order to model future distributions and identify viable refuges in this
region and elsewhere, robust analyses of physicochemical thresholds of freshwater fishes are
required (e.g. Segurado et al., 2011).
The effect of interactions between pollution and other stressors on aquatic ecosystems
can be complex and difficult to predict (Cañedo-Argüelles et al., 2013). For example,
although reductions in rainfall and fresh surface flow may reduce the dilution capacity of
rivers in south-western Australia, this may be partially offset by concomitant reductions in
saline groundwater levels (Mayer et al., 2005). Specifically, the disconnection of saline
groundwater with low order streams due to rainfall decline results in a disproportionally
greater reduction in streamflow and also a reduction in the stream salinity (Kinal &
Stoneman, 2012). Given the vital importance of dry-season refuge in the life-cycles of
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
potamodromous fishes in temperate environments, the impact of climate change on its
availability and quality warrants much greater research attention.
Impacts of climate change on migrations
Globally, reductions in the discharge from rivers due to water extraction and climate change
is predicted to cause a loss of up to 75% of freshwater fish diversity by 2070 (Xenopoulos et
al., 2005; Xenopoulos & Lodge, 2006). However, the threat of climate change to freshwater
ecosystems on the Australian continent, particularly freshwater fishes, has only recently
begun to be investigated (Koehn, 2011; Morrongiello et al., 2011a,b; Chessman, 2013).
There is universal agreement amongst GCMs that rainfall will continue to decline in south-
western Australia (Suppiah et al., 2007; Silberstein et al., 2012). The dramatic past and
projected climatic shift that has occurred in this region renders it a ‘living experiment’ in
assessing the ecological impacts of climate change in temperate regions.
We found that spawning migrations of potamodromous fishes in the current study were
almost always positively related to discharge. Our study streams were remarkably similar in
most geophysical and many ecological variables and we therefore examined only a modest
number of potential predictors of migrations. A study with a wider geographical range may
reveal other environmental variables that also influence migrations in these potamodromous
fishes. Nevertheless, the projected reductions in annual surface water discharge and stream
flow periods of the region (Silberstein et al. 2012) will reduce spawning habitat availability
and connectivity for freshwater fishes on both a temporal and spatial scale. The amount of
river discharge has previously been shown to be positively associated with migration,
reproduction and recruitment of fishes elsewhere (e.g. Reynolds, 1983; O’Connor et al.,
2005; Beatty et al., 2010; Crook et al., 2010; Walsh et al., 2012; Munz & Higgins, 2013).
Our findings suggest that potamodromous fishes may be particularly impacted by flow
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
reductions via reduced levels of recruitment and increased resource competition, with
potentially serious implications for population dynamics and viabilities (e.g. Morán-López et
al., 2012).
Along with causing flow reductions, climate change is projected to increase mean air
temperatures and the number of extreme hot days (>35°C) in south-western Australia
(Suppiah et al., 2007). As expected, water temperatures in the current study were closely
linked to air temperatures. As temperature affects the aerobic scope of fishes, which dictates
their swimming performance (Farrell et al., 2008), these increases may further restrict the
migratory capacity and population viability of potamodromous species in this region. Hague
et al. (2011) modelled the effects of climate change on sockeye salmon populations and
concluded that increasing temperatures would result in most being forced to migrate at sub-
optimal aerobic scopes. The geographical location of south-western Australia (i.e. the
southern edge of the Australian continent) means that aquatic species including freshwater
fishes cannot migrate southward in response to increasing temperatures and this greatly
increases their vulnerability to climate change (Davies, 2010).
In a recent assessment of how 39 species of south-eastern Australian fishes fared under a
‘Millennium’ drought, the most impacted species had a small maximum size, were
invertivorous (cf omnivorous), had a low age at maturity, spawned in low temperatures, had a
prolonged spawning period, low fecundity, demersal eggs and a low upper thermal limit
(Chessman, 2013). The species in the current study have almost all of these vulnerable traits
(Morgan et al. 1998). There is a paucity of research into how Australian freshwater fishes
affect aquatic food webs (Stoffels, 2013). The lack of any comprehensive food-web studies
that have incorporated freshwater fishes in south-western Australia precludes predicting what
impact their decline will have on the prevailing aquatic ecosystems. However, the loss of
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
such highly abundant top-order consumers (Morgan et al., 1998) may cause considerable
ecological change through shifts in prey populations and trophic cascades.
Protecting freshwater biodiversity, particularly migrating fauna, is extremely challenging
as freshwater ecosystems are impacted by activities that occur on terrestrial and aquatic
landscapes and there is a fundamental ‘thirst’ by humans for fresh water (Dudgeon et al.
2006). Altered flow regimes caused by climatically induced reductions in runoff and water
abstraction coupled with pollution, particularly salinisation, are acting synergistically in
south-western Australia. Setting thresholds of declines in discharge in order to guide
management intervention has been undertaken globally but can also be undertaken on a more
local scale to increase sensitivity (Palmer et al., 2008). Our study suggests that spawning
migrations of potamodromous fishes will inevitably decline, leading to reduced recruitment
and threatening the viability of populations and species. Such hydro-ecological research is
urgently required elsewhere to better understand the resistance and resilience of
potamodromous fishes to climate change (see Moilanen et al., 2008; Williams et al., 2011).
Given the recent climatic shift and consensus among GCMs of continued severe drying
in south-western Australia, there is little excuse not to undertake proactive management
strategies and adapt our responses to buffer the impacts on aquatic ecosystems (see Palmer et
al., 2009). Remediation strategies to re-connect aquatic habitats may help mitigate the
impact of flow declines (Palmer et al., 2008, 2009; Beatty et al., 2013). Addressing existing
stressors on freshwater ecosystems in drying climatic regions at their source, such as
undertaking more sustainable land use practices and approaches to both surface water and
groundwater resource development (Xenopoulos et al., 2005; Xenopoulos & Lodge, 2006;
Palmer et al., 2008, 2009; Vörösmarty et al., 2010; Davis et al., 2013) must also be a priority
if the current and projected impacts of climate change are to be reduced.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Acknowledgements
This project was jointly funded by the Department of Water Government, of Western
Australia and the Australian Government under its Water for the Future Plan. Special thanks
to Adrian Goodreid, Natasha Del Borrello and Rob Donohue (Department of Water,
Government of Western Australia) for project management. Many thanks also to the
Department of Water Bunbury staff including Ash Ramsay, Richard Pickett and Andrew
Bland for hydrological data. We are grateful for the field assistance that was provided by
Fiona and Andrew Rowland, Travis Fazeldean, Simon Visser, Mark Allen, David John,
Joshua Johnston and Trine Beatty. Thanks also to Adrian Gleiss for reviewing earlier drafts
of the manuscript, and to three anonymous referees and Rhea Kressman, who made many
constructive suggestions which substantially improved the manuscript. The study complied
with permits provided by the Murdoch University Animal Ethics Committee, the Department
of Environment and Conservation and the Department of Fisheries, Government of Western
Australia.
References Allan JD, Erickson DL, Fay J (1997) The influence of catchment land use on stream integrity
across multiple spatial scales. Freshwater Biology, 37, 149-161.
Allen GR, Midgley SH, Allen M (2002) Field Guide to the Freshwater Fishes of Australia.
Western Australian Museum, Perth, Western Australia, 394 pp.
Baker CF, Hicks BJ (2003) Attraction of migratory inanga (Galaxias maculatus) and koaro
(Galaxias brevipinnis) juveniles to adult galaxiid odours. New Zealand Journal of
Marine and Freshwater Research, 37, 291-299.
Baker CF, Montgomery JC (2001) Species-specific attraction of migratory banded kokopu
juveniles to adult pheromones. Journal of Fish Biology, 58, 1221-1229.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Balcombe SR, Sheldon F, Capon SJ, Bond NR, Hadwen WL, Marsh N, Bernays SJ (2011)
Climate-change threats to native fish in degraded rivers and floodplains of the Murray
Darling Basin, Australia. Marine and Freshwater Research, 62, 1099-1114.
Barron O, Silberstein R, Ali R et al. (2012) Climate change effects on water-dependent
ecosystems in south-western Australia. Journal of Hydrology, 434-435, 95-109.
Barthel BL, Cooke SJ, Svec JH, Suski CD, Bunt CM, Phelan FJS, Philipp DP (2008)
Divergent life histories among smallmouth bass Micropterus dolomieu inhabiting a
connected river-lake system. Journal of Fish Biology, 73, 829-852.
Bartoñ K (2013) MuMIn: multi-model inference. Version 1.9.5. Available at: <cran.r-
project.org/web/packages/MuMIn/>
Baumgartner LJ, Harris JH (2007) Passage of non-salmonid fish through a deelder lock on a
lowland river. River Research and Applications, 23, 1058-1069.
Beatty S, Allen M, Lymbery A, Storer T, White G, Morgan D, Ryan T (2013) Novel methods
for managing freshwater refuges against climate change in southern Australia:
Evaluating small barrier removal to improve refuge connectivity: A global review of
barrier decommissioning and a process for southern Australia in a drying climate.
National Climate Change Adaptation Research Facility, Gold Coast, 73 pp. Available
at: <http://www.nccarf.edu.au/publications/evaluating-small-barrier-removal-improve-
refuge-connectivity>
Beatty SJ, Morgan DL (2013) Introduced freshwater fishes in a global endemic hotspot and
implications of habitat and climatic change. BioInvasions Records, 2, 1-9.
Beatty SJ, Morgan DL, Mcaleer FJ, Ramsay AR (2010) Groundwater contribution to
baseflow maintains habitat connectivity for Tandanus bostocki (Teleostei: Plotosidae)
in a south-western Australian river. Ecology of Freshwater Fish, 19, 595-608.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Beatty SJ, Morgan DL, Rashnavadi M, Lymbery AJ (2011) Salinity tolerances of endemic
freshwater fishes of south-western Australia: Implications for conservation in a
biodiversity hotspot. Marine and Freshwater Research, 62, 91-100.
Beatty SJ, Morgan DL, Torre A (2007) Restoring ecological connectivity in the Margaret
River: Western Australia's first rock-ramp fishways. Ecological Management and
Restoration, 8, 224-228.
Bond N, Thomson J, Reich P, Stein J (2011) Using species distribution models to infer
potential climate change-induced range shifts of freshwater fish in south-eastern
Australia. Marine and Freshwater Research, 62, 1043-1061.
Branco P, Segurado P, Santos JM, Pinheiro P, Ferreira MT (2012) Does longitudinal
connectivity loss affect the distribution of freshwater fish? Ecological Engineering, 48,
70-78.
Buisson L, Thuiller W, Lek S, Lim PUY, Grenouillet G (2008) Climate change hastens the
turnover of stream fish assemblages. Global Change Biology, 14, 2232-2248.
Burnham KP, Anderson DR (2010) Model Selection and Multimodel Inference. Second
Edition. Springer-Verlag, New York, USA.
Bunn SE, Arthington AH (2002) Basic principles and ecological consequences of altered
flow regimes for aquatic biodiversity. Environmental Management, 30, 492-507.
Bunn SE, Davies PM (1990) Why is the stream fauna of south-western Australia so
impoverished? Hydrobiologia, 194, 169-176.
Cañedo-Argüelles M, Kefford BJ, Piscart C, Prat N, Schäfer RB, Schulz CJ (2013)
Salinisation of rivers: An urgent ecological issue. Environmental Pollution, 173, 157-
167.
Centre of Excellence in Natural Resource Management (2005) Ecological Water
Requirements of the Blackwood River and tributaries – Nannup to Hut Pool. Report
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
CENRM 11/04. Centre of Excellence in Natural Resource Management, the University
of Western Australia. Viewed 30 June 2013
<http://www.water.wa.gov.au/PublicationStore/first/82070.pdf>
Chapman A, Morgan DL, Beatty SJ, Gill HS (2006) Variation in life history of land-locked
lacustrine and riverine populations of Galaxias maculatus (Jenyns 1842) in Western
Australia. Environmental Biology of Fishes, 77, 21-37.
Chessman BC (2013) Identifying species at risk from climate change: Traits predict the
drought vulnerability of freshwater fishes. Biological Conservation, 160, 40-49.
Chevan A, Sutherland M (1991) Hierarchical partitioning. The American Statistician, 45, 90-
96.
Clavero M, Hermoso V, Levin N, Kark S (2010) Geographical linkages between threats and
imperilment in freshwater fish in the Mediterranean Basin. Diversity and Distributions,
16, 744-754.
Clews E, Durance I, Vaughan IP, Ormerod SJ (2010) Juvenile salmonid populations in a
temperate river system track synoptic trends in climate. Global Change Biology, 16,
3271-3283.
Comte L, Buisson L, Daufresne M, Grenouillet G (2013) Climate-induced changes in the
distribution of freshwater fish: observed and predicted trends. Freshwater Biology, 58,
625-639.
Conallin AJ, Hillyard KA, Walker KF, Gillanders BM, Smith BB (2011) Offstream
movements of fish during drought in a regulated lowland river. River Research and
Applications, 27, 1237-1252.
Crook DA, Koster WM, Macdonald JI, Nicol SJ et al. (2010) Catadromous migrations by
female tupong (Pseudaphritis urvillii) in coastal streams in Victoria, Australia. Marine
and Freshwater Research, 61, 474-483.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Crozier LG, Zabel RW, Hamlet AF (2008) Predicting differential effects of climate change at
the population level with life-cycle models of spring Chinook salmon. Global Change
Biology, 14, 236-249.
Davies PM (2010) Climate change implications for river restoration in global biodiversity
hotspots. Restoration Ecology, 18, 261-268.
Davis J, Pavlova A, Thompson R, Sunnucks P (2013) Evolutionary refugia and ecological
refuges: key concepts for conserving Australian arid zone freshwater biodiversity under
climate change. Global Change Biology, 19, 1970-1984.
Del Borrello NL (2008) Management Triggers and Responses for Groundwater Dependent
Ecosystems in the South West Groundwater Areas, Water resource allocation planning
report No. 31, Department of Water, Government of Western Australia, Perth. doi:
http://www.water.wa.gov.au/PublicationStore/first/84212.pdf
Dugan PJ, Barlow C, Agostinho AA et al. (2010) Fish migration, dams, and loss of
ecosystem services in the Mekong Basin. Ambio, 39, 344-348.
Dudgeon D, Arthington AH, Gessner MO et al. (2006) Freshwater biodiversity: importance,
threats, status and conservation challenges. Biological Reviews, 81, 163-182.
Elith J, Leathwick JR (2009) Species distribution models: Ecological explanation and
prediction across space and time. Annual Review of Ecology, Evolution, and
Systematics, 40, 677-697.
Farrell AP, Hinch SG, Cooke SJ, Patterson DA, Crossin GT, Lapointe M, Mathes MT (2008)
Pacific salmon in hot water: Applying aerobic scope models and biotelemetry to predict
the success of spawning migrations. Physiological and Biochemical Zoology, 81, 697-
708.
Ficke AD, Myrick CA, Hansen LJ (2007) Potential impacts of global climate change on
freshwater fisheries. Reviews in Fish Biology and Fisheries, 17, 581-613.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Finstad AG, Hein CL (2012) Migrate or stay: terrestrial primary productivity and climate
drive anadromy in Arctic char. Global Change Biology, 18, 2487-2497.
Freckleton RP (2011) Dealing with collinearity in behavioural and ecological data: model
averaging and the problems of measurement error. Behavioural Ecology and
Sociobiology, 65, 91-101.
Golder & Associates (2008) Analysis of the Lower Blackwood River and Tributaries. Flow,
Groundwater and Climate Relationships and Recommendation of Management
Triggers. Report number 077641626001 R Rev1. Report to the Department of Water,
Government of Western Australia.
Gorman OT, Karr JR (1978) Habitat structure and stream fish communities. Ecology, 59,
507-515.
Hague MJ, Ferrari MR, Miller JR, Patterson DA, Russell GL, Farrell AP, Hinch SG (2011)
Modelling the future hydroclimatology of the lower Fraser River and its impacts on the
spawning migration survival of sockeye salmon. Global Change Biology, 17, 87-98.
Hammer M, Unmack P, Adams M, Johnson J, Walker K (2009) Phylogeographic structure in
the threatened Yarra pygmy perch Nannoperca obscura (Teleostei: Percichthyidae) has
major implications for declining populations. Conservation Genetics, 11, 213-23.
Heiberger RM (2013) HH: statistical analysis and data display. Version 2.3-37. Available at:
<cran.r-project.org/web/packages/HH/>
Hermoso V, Clavero M (2011) Threatening processes and conservation management of
endemic freshwater fish in the Mediterranean basin: A review. Marine and Freshwater
Research, 62, 244-254.
Hodges SW, Magoulick DD (2011) Refuge habitats for fishes during seasonal drying in an
intermittent stream: Movement, survival and abundance of three minnow species.
Aquatic Sciences, 73, 513-522.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Hughes J, Ponniah M, Hurwood D, Chenoweth S, Arthington A (1999) Strong genetic
structuring in a habitat specialist, the Oxleyan Pygmy Perch Nannoperca oxleyana.
Heredity, 83, 5-14.
Katopodis C, Aadland LP (2006) Effective dam removal and river channel restoration
approaches, IAHR International Journal of River Basin Management, Special Issue on
Ecohydraulics, 4, 153-168.
Kinal J, Stoneman GL (2012) Disconnection of groundwater from surface water causes a
fundamental change in hydrology in a forested catchment in south-western Australia.
Journal of Hydrology, 472–473, 14-24.
Knight JT, Arthington AH (2008) Distribution and habitat associations of the endangered
Oxleyan pygmy perch, Nannoperca oxleyana Whitley, in eastern Australia. Aquatic
Conservation: Marine and Freshwater Ecosystems, 18, 1240-1254.
Koehn JD (2011) Climate change and Australian marine and freshwater environments, fishes
and fisheries: introduction. Marine and Freshwater Research, 62, 981-983.
Lassalle G, Rochard E (2009) Impact of twenty-first century climate change on diadromous
fish spread over Europe, North Africa and the Middle East. Global Change Biology, 15,
1072-1089.
Mac Nally R (2002) Multiple regression and inference in ecology and conservation biology:
further comments on identifying important predictor variables. Biodiversity and
Conservation, 11, 1397-1401.
Maceda-Veiga A (2013) Towards the conservation of freshwater fish: Iberian Rivers as an
example of threats and management practices. Reviews in Fish Biology and Fisheries,
23, 1-22.
Magoulick DD, Kobza RM (2003) The role of refugia for fishes during drought: A review
and synthesis. Freshwater Biology, 48, 1186-1198.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Mallen-Cooper M, Brand DA (2007) Non-salmonids in a salmonid fishway: What do 50
years of data tell us about past and future fish passage? Fisheries Management and
Ecology, 14, 319-332.
Marr SM, Marchetti MP, Olden JD et al. (2010) Freshwater fish introductions in
mediterranean-climate regions: Are there commonalities in the conservation problem?
Diversity and Distributions, 16, 606-619.
Matthews WJ (1998) Patterns in Freshwater Fish Ecology. Chapman and Hall, Norwell,
Massachusetts, U.S.A. 756 pp.
Mayer, X, Ruprecht, J, Bari, M (2005) Stream Salinity Status and Trends in South-West
Western Australia. Department of Environment, Government of Western Australia. 188
pp.
McDowall, RM & Frankenberg RS (1981) The galaxiid fishes of Australia (Pisces:
Galaxiidae). Records of the Australian Museum, 33, 443–605.
Mims MC, Olden JD (2012) Life history theory predicts fish assemblage response to
hydrologic regimes. Ecology, 93, 35-45.
Moilanen A, Leathwick J, Elith J (2008) A method for spatial freshwater conservation
prioritization. Freshwater Biology, 53, 577-592.
Morán-López R, Pérez-Bote JL, da Silva E, Casildo ABP (2012) Hierarchical large-scale to
local-scale influence of abiotic factors in summer-fragmented Mediterranean rivers:
Structuring effects on fish distributions, assemblage composition and species richness.
Hydrobiologia, 696, 137-158.
Morgan, DL (2003) Distribution and biology of Galaxias truttaceus (Galaxiidae) in
southwestern Australia, including first evidence of parasitism of fishes in Western
Australia by Ligula intestinalis (Cestoda). Environmental Biology of Fishes 66, 155-
167.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Morgan DL, Beatty SJ (2006) Use of a vertical-slot fishway by galaxiids in Western
Australia. Ecology of Freshwater Fish, 15, 500-509.
Morgan DL, Beatty SJ (2013) Nannoperca pygmaea, a new species of pygmy perch
(Teleostei: Percichthyidae) from Western Australia. Zootaxa, 3637, 401-411.
Morgan DL, Gill HS, Potter IC (1995) Life cycle, growth and diet of Balston's pygmy perch
in its natural habitat of acidic pools in south-western Australia. Journal of Fish Biology,
47, 808-825.
Morgan DL, Gill HS, Potter IC (1998) Distribution, identification and biology of freshwater
fishes in south-western Australia. Records of Western Australian Museum, Supplement
56, 1-97.
Morgan DL, Gill HS, Maddern MG, Beatty SJ (2004) Distribution and impacts of introduced
freshwater fishes in Western Australia. New Zealand Journal of Marine and
Freshwater Research, 38, 511-523.
Morgan DL, Thorburn DC, Gill HS (2003) Salinization of southwestern Western Australian
rivers and the implications for the inland fish fauna - The Blackwood River, a case
study. Pacific Conservation Biology, 9, 161-171.
Morrongiello JR, Beatty SJ, Bennett JC et al. (2011a) Climate change and its implications for
Australia's freshwater fish. Marine and Freshwater Research, 62, 1082-1098.
Morrongiello JR, Crook DA, King AJ, Ramsey DSL, Brown P (2011b) Impacts of drought
and predicted effects of climate change on fish growth in temperate Australian lakes.
Global Change Biology, 17, 745-755.
Munz JT, Higgins CL (2013) The influence of discharge, photoperiod, and temperature on
the reproductive ecology of cyprinids in the Paluxy River, Texas. Aquatic Ecology, 47,
67-74.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Murray K, Conner MM (2009) Methods to quantify variable importance: implications for the
analysis of noisy ecological data. Ecology, 90, 348-355.
Myers R (1990) Classical and Modern Regression with Applications (Second Edition).
Duxbury, Boston, MA, 488 pp.
Myers N, Mittermeler RA, Mittermeler CG, Da Fonseca GaB, Kent J (2000) Biodiversity
hotspots for conservation priorities. Nature, 403, 853-858.
O’Brien RM (2007) A caution regarding rules of thumb for variance inflation factors. Quality
& Quantity, 41, 673-690.
O’Connor JP, O'mahony DJ, O'mahony JM (2005) Movements of Macquaria ambigua, in the
Murray River, south-eastern Australia. Journal of Fish Biology, 66, 392-403.
Olden JD, Kennard MJ, Leprieur F, Tedesco PA, Winemiller KO, Garcã-a-Berthou E (2010)
Conservation biogeography of freshwater fishes: Recent progress and future challenges.
Diversity and Distributions, 16, 496-513.
Palmer MA, Lettenmaier DP, Poff NL, Postel SL, Richter B, Warner R (2009) Climate
change and river ecosystems: protection and adaptation options. Environmental
Management, 44, 1053-1068.
Palmer MA, Liermann CaR, Nilsson C, Florke M, Alcamo J, Lake PS, Bond N (2008)
Climate change and the world's river basins: anticipating management options.
Frontiers in Ecology and the Environment, 6, 81-89.
Pen LJ, Potter IC (1991) Biology of the western pygmy perch, Edelia vittata, and
comparisons with two other teleost species endemic to south-western Australia.
Environmental Biology of Fishes, 31, 365-380.
Petrone KC, Hughes JD, Van Niel TG, Silberstein RP (2010) Streamflow decline in
southwestern Australia, 1950–2008. Geophysical Research Letters, 37, L11401.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Piou C, Prévost E (2013) Contrasting effects of climate change in continental vs. oceanic
environments on population persistence and microevolution of Atlantic salmon. Global
Change Biology, 19, 711-723.
R Development Core Team (2013) R: a language and environment for statistical computing.
Version 3.0.1. R Foundation for Statistical Computing, Vienna. Available at: <http://r-
project.org>
Reynolds LF (1983) Migration patterns of five fish species in the Murray-Darling River
system. Australian Journal of Marine and Freshwater Research, 34, 857-871.
Segurado P, Santos JM, Pont D, Melcher AH, Jalon DG, Hughes RM, Ferreira MT (2011)
Estimating species tolerance to human perturbation: Expert judgment versus empirical
approaches. Ecological Indicators, 11, 1623-1635.
Silberstein RP, Aryal SK, Durrant J et al. (2012) Climate change and runoff in south-western
Australia. Journal of Hydrology, 475, 441-455.
Smith AC, Koper N, Francis CM, Fahrig L (2009) Confronting collinearity: comparing
methods for disentangling the effects of habitat loss and fragmentation. Landscape
Ecology, 24, 1271-1285.
Strategen (2006) South West Yarragadee Water Supply Development: Sustainability
Evaluation/Environmental Review and Management Programme Volume 2. Report for
the Water Corporation, Perth, Western Australia. Viewed 14 December 2012
<http://www.water.wa.gov.au/PublicationStore/first/82411.pdf>
Sternberg D, Kennard MJ (2013) Environmental, spatial and phylogenetic determinants of
fish life-history traits and functional composition of Australian rivers. Freshwater
Biology, 58, 1767-1778.
Stoffels RJ (2013) Trophic ecology. In: Ecology of Australian Freshwater Fishes (eds
Humphries P, Walker K), pp.131-158. CSIRO Publishing, Victoria, Australia.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Suppiah R, Hennessy KJ, Whetton PH, McInnes K, Macadam I, Bathols J, Ricketts J, Page
CM (2007) Australian climate change projections derived from simulations performed
for the IPCC 4th assessment report. Australian Meteorological Magazine, 131, 131–
152.
Thieme ML, Lehner B, Abell R, Matthews J (2010) Exposure of Africa's freshwater
biodiversity to a changing climate. Conservation Letters, 3, 324-331.
Vörösmarty CJ, Mcintyre PB, Gessner MO, et al. (2010) Global threats to human water
security and river biodiversity. Nature, 467, 555-561.
Walsh C, Mac Nally R (2013) hier.part: hierarchical partitioning. Version 1.0-4. Available at:
<cran.r-project.org/web/packages/hier.part/>
Walsh CT, Reinfelds IV, Gray CA, West RJ, Van Der Meulen DE, Craig JR (2012) Seasonal
residency and movement patterns of two co-occurring catadromous percichthyids
within a south-eastern Australian river. Ecology of Freshwater Fish, 21, 145-159.
Williams JE, Williams RN, Thurow RF et al. (2011) Native Fish Conservation Areas: A
Vision for Large-Scale Conservation of Native Fish Communities. Fisheries, 36, 267-
277.
Xenopoulos MA, Lodge DM (2006) Going with the flow: Using species-discharge
relationships to forecast losses in fish biodiversity. Ecology, 87, 1907-1914.
Xenopoulos MA, Lodge DM, Alcamo J, Märker M, Schulze K, Van Vuuren DP (2005)
Scenarios of freshwater fish extinctions from climate change and water withdrawal.
Global Change Biology, 11, 1557-1564.
SUPPORTING INFORMATION Table S1 Matrix of Pearson product-moment correlations among environmental predictor
variables (all except pH log10 transformed).
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Table S2 Water quality at sites sampled in the Blackwood River.
Table S3 Prevalence of species and the water quality at their capture sites in the Blackwood
River.
Fig. S1 Mean number of Galaxias occidentalis moving in the tributary and main channel
sites of the Blackwood River. NF indicates that the system was not flowing and therefore
was not sampled.
Fig. S2 Length-frequency distributions of Galaxias occidentalis in Milyeannup Brook.
Fig. S3 Length-frequency distributions of Galaxias occidentalis in Rosa Brook.
Fig. S4 Mean number of Galaxiella munda moving in the tributary sites of the Blackwood
River. NF indicates that the system was not flowing and therefore was not sampled.
Fig. S5 Length-frequency distributions of Galaxiella munda in Milyeannup Brook.
Fig. S6 Mean number of Nannatherina balstoni moving in the tributary and main channel
sites of the Blackwood River. NF indicates that the system was not flowing and therefore
was not sampled.
Fig. S7 Length-frequency distributions of Nannatherina balstoni in Milyeannup Brook.
Fig. S8 Mean number of Nannoperca vittata moving in the tributary and main channel sites
of the Blackwood River. NF indicates that the system was not flowing and therefore was not
sampled.
Fig. S9 Length-frequency distributions of Nannoperca vittata in Milyeannup Brook.
Fig. S10 Length-frequency distributions of Nannoperca vittata in Rosa Brook.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Fig. S11 Mean number of Bostockia porosa moving in the tributary and main channel sites
of the Blackwood River. NF indicates that the system was not flowing and therefore was not
sampled.
Fig. S12 Length-frequency distributions of Bostockia porosa in Rosa Brook.
Fig. S13 Length-frequency distributions of Bostockia porosa in Layman Brook.
Table 1 Association between environmental variables and number of fish moving upstream for five freshwater fish species over the major flow period in four tributaries of the Blackwood River. All variables except pH were log10 transformed. Variables are ranked by relative importance, with coefficient estimates and standard errors from model averaging. The percentage independent contribution of each environmental variable to total explained variance was determined by hierarchical partitioning of R2 values. Significant predictors (P < 0.05) from model averaging are shown in bold and from hierarchical partitioning with an asterix. Response Environmental
variable Importance Estimate SE Independent
effect (%)
N. balstoni Discharge 0.44 0.44 0.29 38.3
Conductivity 0.39 -2.49 1.70 35.2
pH 0.19 -0.27 0.42 11.7
Dissolved oxygen 0.16 -0.93 2.52 9.1
Temperature 0.15 -1.97 5.10 5.7
N. vittata Discharge 0.98 1.00 0.27 49.8
Temperature 0.81 -12.02 4.48 25.3
Conductivity 0.20 1.90 1.71 14.0
Dissolved oxygen 0.15 -2.19 3.05 5.5
pH 0.10 0.14 0.41 5.4
B porosa Discharge 0.99 0.46 0.12 63.0*
pH 0.57 0.29 0.13 23.9
Conductivity 0.34 -1.13 0.68 4.6
Temperature 0.13 -1.40 1.99 2.2
Dissolved oxygen 0.10 0.02 1.01 6.3
G. occidentalis Discharge 0.86 0.69 0.26 61.8
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Conductivity 0.54 -2.89 1.49 23.5
pH 0.17 --0.23 0.41 5.6
Dissolved oxygen 0.13 0.00 4.86 6.2
Temperature 0.13 -2.45 2.56 2.9
G. munda Temperature 0.62 -1.04 0.51 42.5*
Discharge 0.34 0.07 0.03 43.1*
Conductivity 0.16 0.11 0.19 5.5
Dissolved oxygen 0.16 0.04 0.33 6.8
pH 0.13 0.01 0.04 2.1
Table 2 Association between environmental variables and number of fish moving upstream for five freshwater fish species over the major flow period in four tributaries of the Blackwood River. All variables except pH were log10 transformed. Variables are ranked by relative importance, with coefficient estimates and standard errors from model averaging. The percentage independent contribution of each environmental variable to total explained variance was determined by hierarchical partitioning of R2 values. Significant predictors (P < 0.05) from model averaging are shown in bold and from hierarchical partitioning with an asterix.
Response Environmental variable
Importance Estimate SE Independent effect (%)
N. balstoni Discharge 0.63 0.52 0.55 26.3
Conductivity 0.33 -4.60 3.32 37.9
pH 0.22 -0.77 0.78 20.7
Temperature 0.16 -6.10 9.69 6.3
Dissolved oxygen 0.14 -1.93 4.63 8.8
N. vittata Discharge 1.00 1.15 0.22 51.1*
Temperature 0.98 -14.84 3.81 31.7*
Conductivity 0.15 1.33 1.39 8.8
Dissolved oxygen 0.09 0.62 2.59 5.9
pH 0.08 0.00 0.30 2.6
B porosa Dissolved oxygen 0.89 3.87 1.31 62.7*
Discharge 0.84 -0.19 0.17 21.2
Temperature 0.11 -0.02 4.21 9.9
Conductivity 0.10 0.26 1.15 4.5
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
pH 0.09 0.02 0.23 1.7
G. occidentalis Discharge 0.55 0.59 0.32 38.6*
Conductivity 0.43 -3.35 2.03 22.3
pH 0.41 -0.69 0.43 23.5
Dissolved oxygen 0.16 1.91 2.67 8.8
Temperature 0.15 -4.00 5.72 6.8
G. munda Discharge 0.99 0.13 0.10 38.8
Conductivity 0.42 -1.01 0.58 33.9
pH 0.33 -0.20 0.13 18.7
Temperature 0.17 -1.58 1.60 5.0
Dissolved oxygen 0.11 -0.11 0.82 3.6
FIGURE LEGENDS
Fig. 1 Sites sampled in the Blackwood River and its tributaries to determine the movement
patterns of south-western Australian freshwater fishes.
Fig. 2 Water temperatures at the sites sampled in the Blackwood River and its tributaries.
Note the close association between the water temperatures in all sites and the maximum daily
air temperature.
Fig. 3 Schematic diagram of the temporal and spatial movement patterns of two species of
Galaxiidae in the main channel of the Blackwood River and in a seasonal and a perennial
tributary, south-western Australia. Direction of movement (upstream or downstream) is
indicated by the orientation of fish. New recruits are indicated by smaller sized fish.
Fig. 4 Schematic diagram of the temporal and spatial movement patterns of three species of
Percichthyidae in the main channel of the Blackwood River and in a seasonal and a perennial
tributary, south-western Australia. Direction of movement (upstream or downstream) is
indicated by the orientation of fish. New recruits are indicated by smaller sized fish.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rtic
le
This article is protected by copyright. All rights reserved.