Search

In dryland environments where rivers have been modified by water resource developments, the use of environmental water provides a management tool to restore or maintain riparian vegetation conditions. Relevant management actions include watering to promote the recruitment of highly valued woody species. However, for many riparian trees, little is known about the watering regimes needed to achieve successful recruitment. We conducted a glasshouse experiment to investigate the early growth of 5-week old seedlings of two Australian dryland riparian tree species, coolibah (Eucalyptus coolabah) and river red gum (Eucalyptus camaldulensis), in floodplain soil cores to provide insights into their requirements for successful establishment in dryland habitats. Both species exhibited similar growth patterns for their above-ground components. E. coolabah root penetration through soil cores was consistent over the 12-week experiment. In contrast, E. camaldulensis root penetration appeared to be considerably slower during the first 6 weeks, but then became more rapid over the following 6 weeks, with similar rooting depths after 12 weeks observed to those of E. coolabah. These findings suggest that coolibah seedlings are more likely to establish under rapidly drying conditions than river red gum seedlings. Such differences in early growth for these iconic species should be considered when determining the provision of environmental water for their management.

Studies that associate environmental parameters with aquatic organisms in man-made lakes remain limited by accessibility and interest particularly in many Asian countries. With missed opportunities to monitor environmental transitions at Lake Kenyir, our knowledge of lake transition is restricted to the non-mixing shallow waters only. Triplicate monthly benthic invertebrate samples were collected concurrently with various environmental parameters at three locations (zones A-C) of Kenyir Lake, Malaysia. Our results affirmed that the northeast part of Lake Kenyir is oligotrophic. Abundance of phytoplankton, total suspended solids, phosphate, nitrite and nitrate drive the abundance of various groups of benthic invertebrates. All of these extrinsic variables (except phosphate) negatively influenced the density of Trichoptera and positively influenced (P

Most predictions for the future climate in temperate ecosystems show longer periods of drought followed by more extreme precipitation events. These changes in flow regimes can strongly affect riverine fish populations; therefore, assessing the recovery of populations following these episodic events provides important information to help guide species management. This study investigated the dynamics of 5 spatially distinct riverine populations of the nationally endangered Macquarie perch Macquaria australasica in southeastern Australia from 2007-2018, a period that included the tail end of a decadal long drought followed by extreme fluctuations in river flow. Modelling catch rates and size trends, we found that 4 of the 5 populations showed some synchrony and underwent a slow recovery that was undetectable for several years. The largest increases in population size were driven by the survival of new recruits. Despite this general trend, we could only find a weak association between annual flow extremes and annual population change. This suggests intermediate variations in river flows may have a negligible influence on annual population change or are idiosyncratic in their effect within each waterway or on different life stages. The one population which did not follow this trend occurred in the largest waterway and was subject to legal recreational fishing, suggesting the magnitude of impacts of disturbance on populations occupying larger systems are reduced, and/or that recreational fishing hinders recovery rates following episodic disturbance events such as drought. Our results suggest that forecasting and management of long-lived freshwater fish must incorporate multi-year planning to include factors such as the maintenance of refuges, connectivity and increased protection of mature fish to aid recovery.

Wetlands of Melaleuca spp. in Australia form large forests that are highly threatened by deforestation and degradation. In America, Melaleuca has invaded large areas of native wetlands causing extensive damage. Despite their status as an endangered native ecosystem and as a highly invasive one, little is known about their C and N dynamics. In this study, we sampled five Melaleuca wetlands and measured their C and N ecosystem stocks (aboveground biomass and soil), tree accumulation rates, sedimentation rates, and soil stability. Melaleuca wetlands were highly heterogeneous, but most have large ecosystem C [mean ± SE (range); 360 ± 100 (80–670) Mg C ha−1] and N [8100 ± 1900 (1600–13,000) kg N ha−1] stocks. Tree accumulation rates were 5.0 ± 2.1 Mg C y−1 and 26 ± 14 kg N y−1, and surface soil accumulation rates were 0.6 ± 0.2 Mg C ha−1 y−1 and 39 ± 1 kg N y−1. We found evidence of long-term C and N accumulation in the soil, but also of some level of organic decomposition. Overall, we found that Melaleuca wetlands store and accumulate large amounts of C, especially in their trees, and large amounts of N in their soils, suggesting an important role in coastal biogeochemical cycles.

The intrapopulation variability in the size and age structure of the spawning stock and migration of the threatened Macquarie perch Macquaria australasica in Lake Dartmouth was investigated between 2008 and 2016. Sampling centred on the core reproductive period (October-December) when mature fish migrate from the lake into riverine habitat to spawn. Spawning fish were predominantly large, spanning a broad age structure, with a high proportion of fish (25%) aged 15-30 years. The overall median size of spawning fish did not change for males or females during the 9 year study period. The size of the smaller mature male fish did change in some years suggesting a small proportion of male M. australasica matured at age 1+ and 2+. Acoustic telemetry employed over 3 years showed that M. australasica were most likely to be in the spawning reach from October to mid-December, migrated to this reach annually and moved large distances throughout the lake all year, with no evidence for any spatial structuring. Mature fish sometimes occupied the spawning reach for several months after the core reproductive period, which increased their vulnerability to recreational fishing. Males tagged in the lake were seldom recorded in the spawning reach, presumably because a high proportion had not yet entered the spawning stock despite their size suggesting maturity. Maintaining a broad age and size-structure of the spawning stock of long-lived iteroparous fish species is crucial for recruitment stability and population persistence. Overexploitation of the spawning stock has probably contributed to previous population declines in the lake as well as the collapse of other M. australasica populations in south-eastern Australia.

Socio-economic and environmental drivers are causing large scale abandonment of agricultural land worldwide. Simultaneously, there is growing recognition that urgent and widespread revegetation is required in our agricultural landscapes to restore biodiversity, ecosystem services and promote carbon sequestration. The design of effective revegetation strategies, however, can be limited by a lack of knowledge regarding patterns and processes of regeneration. Here, we examine naturally regenerating woody vegetation on 12 old-fields and compare this to paired remnant vegetation across four regions of a semi-arid agricultural area in eastern Australia. We found that naturally regenerating vegetation on old-fields exhibited similarities to remnant vegetation in terms of composition but varied substantially with respect to key structural attributes. Woody vegetation tends to regenerate in these old-fields with relatively high density given sufficient time (~ 30–40 years), but prior to this, very limited structural recovery occurs at all. The composition and richness of tree seedling and shrub assemblages did not differ between remnant and old-field areas. However, tree assemblages in remnant areas differed in composition and had higher species richness. Old-fields had extremely low numbers of exotic trees and shrubs. Simultaneously, most dominant native species present in the remnant vegetation were also observed in regenerating old-fields, contributing to a low degree of compositional difference between old-fields and remnant areas overall. The abundance of paddock trees, time since abandonment and mean annual rainfall are key drivers influencing natural regeneration of these old-fields. We recommend that natural regeneration is a viable form of restoration of old-fields in the northern Murray-Darling Basin given sufficient time.

Freshwater ecosystems and their associated biota are under increasing threats from multiple stressors including climate and land-use change. The conservation of these ecosystems must be based on an integration of data including species physiological tolerances, the biotic and abiotic drivers of the distribution of populations, and demographic processes, to provide the comprehensive ecological information necessary for management. This study used a Bayesian belief network (BBN) to synthesize research on northern river blackfish, a threatened species in the upper Condamine River, Australia, into a probabilistic framework capable of predicting the complex relationships that exist between environmental conditions and population success. This study tested how predicted air temperature scenarios for the years 2050 and 2080, and catchment restoration scenarios, would be expected to affect three indices of population success: adult abundance, juvenile abundance, and juvenile recruitment. Compared with current climatic conditions, climate warming scenarios reduced the probability of future population success by between 0.4% and 1.6%. These shifts were almost completely offset, and even improved, when riparian zones were restored at the catchment scale, where changes ranged from an overall decrease of 0.2% to an increase of 1%. To achieve the highest probability of population success, the impacts of warming stream temperatures and the degradation of riparian zones must be mitigated. However, the model showed that there is still a possibility of complete population failure under a wide range of conditions, even when conditions appear to be suitable. To maximize the future population success of river blackfish we recommend targeting the restoration of hydrologically active catchment areas where grazing strongly influences stream biota. The use of a BBN allowed the combination of multiple sources of information to solve complex ecological problems, including how multiple stressors may affect threatened freshwater species.

Our lack of knowledge on the spatiotemporal drivers of the distribution of many freshwater fishes, particularly as they differ among life-history stages, is a challenge to conservation of these species. We used 2-stage hurdle models to investigate drivers of occurrence and abundance of locally threatened adult and juvenile Northern River Blackfish in the upper Condamine River, Queensland, Australia. Different processes drive occurrence and abundance between the 2 life-history stages. Both adult and juvenile occurrences were negatively associated with high-magnitude, extended warming events, suggesting Northern River Blackfish are thermally restricted to cooler headwaters. Juveniles had greater sensitivity than adults to high stream temperatures. In contrast, drivers of abundance differed between life-history stages. Adult fish were negatively associated with increased fine-sediment loads, whereas juveniles were negatively associated with a hydrologically active inverse-distance-weighted grazing metric that accounted for the greater influence of grazed land close to the stream or in areas of high overland flow. Teasing apart environmental drivers affecting multiple life-history stages of a locally threatened headwater species enabled us to provide direct management recommendations for conserving this species and ecologically similar headwater fishes and their associated habitats.

Upland freshwater habitats support populations that are especially susceptible to anthropogenic change. Furthermore, their isolation from other suitable habitats, and the fragmented, dendritic structure of headwaters make dispersal an unlikely response to change. We investigated genetic structure and variation in the northern-most population of Gadopsis marmoratus, which is isolated in a tiny area in the headwaters of the Condamine River catchment, in the Murray–Darling Basin, Australia. Strong genetic structure was detected among subpopulations based on microsatellites (FST = 0.173, p < 0.0001) and mitochondrial (mt)DNA (FST = 0.369, p < 0.05). Effective population size was low, ranging between 18.8 and 48.2, depending on the estimation method used. Bayesian clustering revealed 3 genetic clusters, but they were not congruent with drainage patterns, suggesting a complex history of dispersal among headwaters that are isolated by waterfalls. Overall, these results suggest that G. marmoratus is unlik...

The effects of interspecific competition on grazing between two important aquaculture species, mrigal carp Cirrhinus cirrhosis and orangefin labeo Labeo calbasu, in single and in dual combinations were observed in experimental tanks. This study demonstrated that the presence of a competitor did not cause C. cirrhosis to shift its diel feeding patterns. That said, both total food intake and food preference were negatively affected in C. cirrhosis by the presence of a superior competitor, L. calbasu. The feeding patterns of L. calbasu became diurnal in the presence of C. cirrhosus, suggesting highly complex competitive interactions between the two species. That L. calbasu was specifically able to shift circadian feeding patterns to maximize energy intake in the presence of a competitor would suggest that it would be a suitable species to stock in a mixed species aquaculture system.

Freshwater stream systems are under immense pressure from various anthropogenic impacts, including climate change. Stream systems are increasingly being altered by changes to the magnitude, timing, frequency, and duration of their thermal regimes, which will have profound impacts on the life-history dynamics of resident biota within their home range. Although temperature regimes have a significant influence on the biology of instream fauna, large spatio-temporal temperature datasets are often reduced to a single metric at discrete locations and used to describe the thermal regime of a system; potentially leading to a significant loss of information crucial to stream management. Models are often used to extrapolate these metrics to unsampled locations, but it is unclear whether predicting actual daily temperatures or an aggregated metric of the temperature regime best describes the complexity of the thermal regime. We fit spatial statistical stream-network models (SSNMs), random forest and non-spatial linear models to stream temperature data from the Upper Condamine River in QLD, Australia and used them to semi-continuously predict metrics describing the magnitude, duration, and frequency of the thermal regime through space and time. We compared both daily and aggregated temperature metrics and found that SSNMs always had more predictive ability than the random forest models, but both models outperformed the non-spatial linear model. For metrics describing thermal magnitude and duration, aggregated predictions were most accurate, while metrics describing the frequency of heating events were better represented by metrics based on daily predictions generated using a SSNM. A more comprehensive representation of the spatio-temporal thermal regime allows researchers to explore new spatio-temporally explicit questions about the thermal regime. It also provides the information needed to generate a suite of ecologically meaningful metrics capturing multiple aspects of the thermal regime, which will increase our scientific understanding of how organisms respond to thermal cues and provide much-needed information for more effective management actions.

The basic tenet of optimal foraging theory is that animals will select habitats that provide the greatest energy return (of food resources) per unit effort spent searching to optimize their fitness. Fish living in variable environments may use specific cues that indicate optimal feeding patches relative to poorer quality patches. To examine whether carp gudgeons (Hypseleotris spp.) preferentially select habitat patches based on food resource availability, we examined their habitat use in emergent macrophyte stands (the giant rush, Juncus ingens). We tested the hypothesis that carp gudgeons selected patches based upon food availability in a floodplain billabong and experimental billabongs. Under field conditions, fine-scale associations between carp gudgeons and giant rush demonstrated patch-specific differences in their prey (chironomid) density that were positively associated with fish abundance. Furthermore, the amount of biofilm within a patch was positively associated with the density and biomass of chironomids. Experimental evidence demonstrated that carp gudgeon patch selection was independent of stem density and based upon prey density. Carp gudgeons track their food resources through time, thus maximizing their feeding returns in both natural and experimental systems.

In many intermittent, dryland rivers, fish are confined to isolated waterholes for much of the year. It is only during brief flow events, which typify the hydrology of these systems, that fish are able to move between waterholes and explore surrounding habitat. Because most of the river channel will dry afterwards, there is a strong advantage for selection of persistent waterholes. Two hundred and fifteen individual fish of three common large-bodied species were tagged in two isolated waterholes in the Moonie River (Queensland, Australia) over 3years. Their movements were monitored to identify the flow events that trigger fish movement between waterholes, differences in response among species and size classes and refuge selection preferences. Some individuals of all species moved during flow events and others remained within the same waterhole. There was no clear upstream or downstream preference, and most individuals used a reach of up to 20km, although some individuals ranged over more than 70km in only several days. Above a threshold flow of 2m above commence-to-flow level, timing of flow was more important than magnitude, with most movement occurring in response to the first post-winter flow event, independent of its magnitude and duration. Many of the fish that moved displayed philopatry and subsequently returned to their starting waterhole either by the end of a flow event or on subsequent events, suggesting ability to navigate and a preference for more permanent refuge pools. Maximising survival in a highly variable environment provides a plausible mechanism for maintaining these behaviours. Modifications to both flow regime and hydrological connectivity may reduce movement opportunities for fish in intermittent rivers. Our findings show that fish in intermittent systems use networks of waterholes and that management and conservation strategies should aim to maintain movement opportunities at large spatial scales to preserve population resilience.

Forecasting population persistence in environments subjected to periodic disturbances represents a general challenge for ecologists. In arid and semiarid regions, climate change and human water use pose significant threats to the future persistence of aquatic biota whose populations typically depend on permanent refuge waterholes for their viability. As such, habitats are increasingly being lost as a result of decreasing runoff and increasing water extraction. We constructed a spatially explicit population model for golden perch Macquaria ambigua (Richardson), a native freshwater fish in the Murray-Darling Basin in eastern Australia. We then used the model to examine the effects of increased aridity, increased drought frequency, and localized human water extraction on population viability. Consistent with current observations, the model predicted golden perch population persistence under the current climate and levels of water use. Modeled increases in local water extraction greatly increased the risk of population decline, while scenarios of increasing aridity and drought frequency were associated with only minor increases in this risk. We conclude that natural variability in abundances and high turnover rates (extinction/recolonization) of local populations dictate the importance of spatial connectivity and periodic cycles of population growth. Our study also demonstrates an effective way to examine population persistence in intermittent and ephemeral river systems by integrating spatial and temporal dynamics of waterhole persistence with demographic processes (survival, recruitment, and dispersal) within a stochastic modeling framework. The approach can be used to help understand the impacts of natural and anthropogenic drivers, including water resource development, on the viability of biota inhabiting highly dynamic environments.

Dryland river foodwebs are thought to be largely driven by autochthonous production derived from littoral/benthic algae, particularly during extended periods of zero flow, with allochthonous production taking primacy after flooding as floodplain carbon is distributed into waterholes during flood recession. This study tested whether we could detect any influence of autochthonous production (littoral gross primary production) on fish biomass after flooding in Cooper Creek, a dryland river in the Lake Eyre Basin, Australia. It was found that the majority of fish biomass, mostly large carnivores, was not supported by benthic algal production. However, the biomass of juvenile bony bream Nematalosa erebi, the highly abundant benthivore of many Australian dryland rivers, was indeed driven by benthic algal productivity throughout the post-flood period, a finding also evident in post-flood waterholes in the nearby Warrego River (Murray-Darling Basin). With a consistently high abundance turnover of bony bream juveniles and their significant importance to dryland river food-webs this study demonstrates not only the need to protecting intact littoral zones for N. erebi, but by association, fish assemblages in general.

All communities may re-assemble after disturbance. Predictions for re-assembly outcomes are, however, rare. Here we model how fish communities in an extremely variable Australian desert river re-assemble following episodic floods and drying. We apply information entropy to quantify variability in re-assembly and the dichotomy between stochastic and deterministic community states. Species traits were the prime driver of community state: poor oxygen tolerance, low dispersal ability, and high fecundity constrain variation in re-assembly, shifting assemblages towards more stochastic states. In contrast, greater connectivity, while less influential than the measured traits, results in more deterministic states. Ecology has long recognised both the stochastic nature of some re-assembly trajectories and the role of evolutionary and bio-geographic processes. Our models explicitly test the addition of species traits and landscape linkages to improve predictions of community re-assembly, and will be useful in a range of different ecosystems.

The Murray–Darling Basin (MDB) is Australia’s food bowl, contributing 40% of agricultural production and supporting a population of over 4 million people. Historically, the MDB supported a unique native fish community with significant cultural, subsistence, recreational, commercial and ecological values. Approximately one-quarter of the MDB’s native species are endemic. Changes to river flows and habitats have led to a >90% decline in native fish populations over the past 150 years, with almost half the species now of conservation concern. Commercial fisheries have collapsed, and important traditional cultural practices of First Nations People have been weakened. The past 20 years have seen significant advances in the scientific understanding of native fish ecology, the effects of human-related activities and the recovery measures needed. The science is well established, and some robust restoration-enabling policies have been initiated to underpin actions. What is now required is the political vision and commitment to support investment to drive long-term recovery. We present a summary of 30 priority activities urgently needed to restore MDB native fishes.

Many freshwater fishes are imperilled globally, and there is a need for easily accessible, contemporary ecological knowledge to guide management. This compendium contains knowledge collated from over 600 publications and 27 expert workshops to support the restoration of 9 priority native freshwater fish species, representative of the range of life-history strategies and values in south-eastern Australia’s Murray–Darling Basin. To help prioritise future research investment and restoration actions, ecological knowledge and threats were assessed for each species and life stage. There is considerable new knowledge (80% of publications used were from the past 20 years), but this varied among species and life stages, with most known about adults, then egg, juvenile and larval stages (in that order). The biggest knowledge gaps concerned early life stage requirements, survival, recruitment, growth rates, condition and movements. Key threats include reduced longitudinal and lateral connectivity, altered flows, loss of refugia, reductions in both flowing (lotic) and slackwater riverine habitats, degradation of wetland habitats, alien species interactions and loss of aquatic vegetation. Examples and case studies illustrating the application of this knowledge to underpin effective restoration management are provided. This extensive ecological evidence base for multiple species is presented in a tabular format to assist a range of readers.

The distribution and population structure of organisms is governed by a broad suite of biotic and abiotic variables, interacting across multiple scales. Recruitment is a key demographic process critical to the maintenance of successful populations. Isolating and quantifying the multiscale environmental drivers of recruitment is vital for species conservation, especially for those species with traits that increase their susceptibility to local extirpation. We developed a Bayesian hierarchical model to quantify the relationship between the rate of recruitment in a locally threatened river blackfish (Gadopsis marmoratus) and environmental predictors across two spatial scales. There was a 0.99 probability that increased broad-scale stream temperature negatively affected juvenile recruitment rate. We also found that there was a 0.97 probability that the fine-scale relationship between recruitment rate and riparian foliage cover was dependent on temperature. This suggests that broad-scale thermal conditions provide the template upon which at least one local environmental variable influences recruitment rate. Understanding drivers influencing key population processes and the spatial scales at which they operate is critical to gaining an insight into likely changes in population persistence for potentially imperilled species, along with the potential future effects of habitat degradation and climate warming on freshwater fishes in general.

Rapid increases in native riverine fish populations associated with trophic upsurge immediately following reservoir construction are well documented. Repeated upsurge periods and extended benefits to populations are, however, less understood. We used sclerochronology to investigate fish growth and netting surveys to estimate recruitment and abundance of a lacustrine population of an Australian riverine fish, the Macquarie perch (Macquaria australasica) in Lake Dartmouth. Record low inflows from 1997 to 2008 caused the reservoir to shrink to its lowest volume since construction. Refilling began in 2008, reaching 99% capacity in 2013. We hypothesized that fish growth, recruitment, and abundance would increase in response to the refilling of the lake, reproducing a similar response to the initial filling period. Our findings supported this hypothesis. Macquarie perch growth, recruitment, and abundance were enhanced during the refilling of Lake Dartmouth. Growth, best explained by the effects of dam height, change in dam height, temperature (and their interactions), and recruitment, were highest during the first years of refilling when lake levels and temperatures were low. We propose one or a combination of varying levels of intraspecific competition (low during initial filling and high following population expansion) and improved riverine conditions for reproductive success as the most plausible explanation. Our results suggest extended periods of low lake levels followed by rapid inundation events are likely to enhance recruitment and population growth opportunities for this species. While reservoir construction in general impacts negatively on native fish populations, the potential to offset these impacts for conservation management purposes should be considered.

Restoration of floodplain ecosystems through the reinstatement of floods is often hampered by insufficient water as a result of competing human demands. An emerging alternative approach relies on floodplain infrastructure – such as levees, weirs, regulators, and pumps – to control water levels within floodplains without requiring landscape-scale overbank floods. This technique, albeit water efficient and capable of achieving some ecological targets, does not mimic the hydraulics, hydrodynamics, and lateral connectivity of natural floods. Engineering approaches like this may risk detrimental ecological outcomes, including reductions in biotic connectivity, river–floodplain productivity, and water quality, and thus may fail to support the range of ecological processes required to sustain healthy river–floodplain systems. Here, we review the potential benefits, risks, and mitigation options associated with engineered artificial flooding. Given the growing challenge of equitable water allocation, further research on and monitoring of engineered floods as a tool to sustain floodplain ecosystems are urgently required.

Aim This study aims (1) to quantify the broad-scale patterns of functional diversity and life-history strategies of freshwater fish in relation to environmental variation across Australian river basins and (2) to identify key life-history traits associated with species extinction risk in order to determine how fish communities and extinction-prone species may respond to future environmental change. Location One hundred and twenty-three river basins across eastern Australia. Methods Based on 10 key life-history traits for 194 freshwater fish we used a novel analytical approach to quantify multivariate life-history indices in relation to environmental variation within a spatio-phylogenetic framework. We assessed the utility of our analytical framework by contrasting final models against all candidate models, both with and without an eigenvector filtering procedure, and quantified the degree of autocorrelation in all model residuals. Results Temperature, habitat heterogeneity/availability, flow variability and primary productivity accounted for between 55 and 80% of the variation in life-history indices. Best-performing models were all derived from the addition of spatial and phylogenetic covariates to the analytical framework which consistently produced more parsimonious final models with higher explanatory power and insignificant levels of autocorrelation in the model residuals. Main conclusion The life-history functional diversity of fish assemblages and the composition of life-history strategies across Australian river basins is in part determined by environmental variability, stability and seasonality, highlighting both the importance of environmentally driven community assembly processes and the potential changes to freshwater fish biodiversity in response to climate change. A spatio-phylogenetic analytical framework is a key component in the effective management of autocorrelation in ecological data and the derivation of more rigorous trait–environment relationships.

Dryland rivers in arid and semi-arid regions drain approximately one-third of the Earth's land area, yet basic aspects of their ecology are poorly documented and many are threatened by excessive water use leading to alterations of the flow regime. Management of dryland rivers imperilled by changes to the flow regime requires quantitative hydro-ecological relationships and models to support environmental flow restoration strategies and to define ecological response indicators for assessment of trends in ecological health. Fish living in dryland rivers experience highly variable levels of food resources associated with 'boom and bust' productivity patterns driven by highly irregular episodes of rainfall, runoff and floodplain inundation followed by busts associated with drying of remnant aquatic habitats. To cope with variable food resource levels, fish store energy as fat during productivity booms, enabling them to breed later or to survive through extended periods of limited food resources. This study tracked temporal patterns of body condition and recruitment success of the four most abundant fish species in two Australian dryland river systems with contrasting patterns of flow variability, Cooper Creek and the Moonie River. We found consistent responses to flow magnitude and timing for body condition and/or recruitment success in Cooper Creek but fewer responses in the Moonie River. Results suggest that the fish metrics and relationships to flow established for Cooper Creek may apply in similar dryland rivers exhibiting clear boom and bust patterns, whereas the Moonie River results could be applicable to rivers with less variable flow patterns and fewer spells of zero flows. Our study demonstrates that simple, non-destructive methods based on fish length and weight provide useful biological indicators of response to flow variability provided that care is taken with species selection and that hydro-ecological relationships are calibrated for each contrasting type of hydrologic regime.

The endorheic Lake Eyre Basin drains 1.2 million square kilometres of arid central Australia, yet provides habitat for only 30 species of freshwater fish due to the scarcity of water and extreme climate. The majority are hardy riverine species that are adapted to the unpredictable flow regimes, and capable of massive population booms following heavy rainfall and the restoration of connectivity between isolated waterholes. The remainder are endemic specialists from isolated springs with very restricted ranges, and many are listed under relevant state and national endangered species legislation and also by the International Union for Conservation of Nature (IUCN). For these spring communities, which are sustained by water from the Great Artesian Basin, survival is contingent on suitable habitat persisting alongside extractive mining, agriculture and the imposition of alien species. For the riverine species, which frequently undertake long migrations into ephemeral systems, preservation of the natural flow regime is paramount, as this reinstates riverine connectivity. In this study, fish were sampled from the Bulloo River in the east to the Mulligan River in the west, along a temporal timeframe and using a standard set of sampling gears. Fish presence was influenced by factors such as natural catchment divides, sampling time, ephemerality and the occurrence of connection flows and flooding. Despite the comparatively low diversity of species, the aquatic systems of this isolated region remain in good ecological condition, and as such they offer excellent opportunities to investigate the ecology of arid water systems. However, the presence of both endangered species (in the springs) and invasive and translocated species more widely indicates that active protection and management of this unique area is essential to maintain biodiversity and ecosystem integrity.

Amphidromy, the dispersal of larvae to the estuarine or marine environment with juveniles recolonising fresh waters to complete development, is common on tropical islands. This has led to the suggestion that amphidromy is an adaptation to hydrologically unpredictable environments, allowing recolonisation after local extinction (dispersal-limitation hypothesis). Alternatively, amphidromy may be more common in streams with stronger flow, as high tractive forces flush larvae into the estuarine/marine environment, forcing them to recolonise (passive dispersal hypothesis). We tested these hypotheses in a continental gobiid fish, the Roman Nose Goby (Awaous acritosus) that inhabits hydrologically and physically diverse catchments on the east coast of Australia. We measured 87Sr/86Sr along transects of the otoliths of 92 adult fish from 12 catchments, identified migratory life histories from these data and correlated the proportion of migratory fish with key environmental variables. This information was augmented with data from mtDNA (n = 276) and six microsatellite loci (n = 429) from 19 catchments, to explore gene flow among catchments and to assess whether local adaptation was likely in the face of high or restricted gene flow. Estimates of 87Sr/86Sr along transects revealed high variation in life history, with 63% of individuals showing no evidence of marine/estuarine residence. The slope of catchments was the only significant variable associated with the proportion of larvae retained in catchments, with steep catchments having a greater proportion of adults exhibiting a marine/estuarine phase during early life. Total panmixia was detected with mtDNA and six microsatellite loci, suggesting that gene flow is high for this species. These results support the passive dispersal hypothesis. Steep slopes are likely to have higher tractive forces in their flow, leading to a greater proportion of larvae being flushed into the estuary or sea. In addition, high gene flow would inhibit local adaptation, making the dispersal-limitation hypothesis unlikely. This study provides a framework for understanding how local adaptation may be constrained by gene flow across the landscape so that the adaptive or passive mechanism underlying facultative amphidromy can be explored.

Arid zone catchments experience extreme hydrological variability and some rivers are entirely ephemeral, replenished only by intermittent flooding. The ecological roles of ephemeral systems are rarely studied. This paper describes movement patterns of fish in the Mulligan River, an ephemeral system in the Lake Eyre Basin, central Australia. Several sites were sampled along a temporal gradient encompassing floods and dry periods. After a single major flood in 2007 up to seven fish species were found at sites up to 300 km from the closest permanent waterhole. Following a series of floods (when waterholes were replenished and remained wet between 2009 and 2011) a further five species were recorded including the first records for the Lake Eyre hardyhead, Craterocephalus eyresii, from the rivers of far western Queensland. The presence of all species known from the parent catchment (the Georgina, where permanent waterholes occur) in the ephemeral catchment (the Mulligan) suggests that many fish species present in the river systems of central Australia are capable of dispersing long distances following the opening of movement pathways during flooding. However, two distinct groups of species were identified: extreme dispersing species, that move as far as possible into intermittently wetted habitats, and conservative dispersing species, that do not move as far, tending to inhabit deeper waterholes within mid-reaches of the river that are more likely to hold water for longer. Preservation of the natural flow regime of Australia's arid-zone rivers is important for maintaining these fish communities and facilitating study of their adaptations to ephemerality.

In fluctuating aquatic environments such as intermittent streams, fish condition is often highly variable due to the associated fluctuating levels of food resources. Having the physiological capacity to both metabolise and catabolise lipids, fish can survive through droughts and rapidly gain condition during flows and floods. Dryland rivers continuously cycle through periods of boom and bust due to their intermittent patterns of rainfall and flow. To help gain an understanding of how fish respond physiologically to varying antecedent flow conditions, we examined body condition measured by percent tissue water content of two generalist fish species in an Australian dryland river. We predicted that fish would be in better condition following recent flows and poorer condition when there had been no recent flows. Our expectations were met for both species with low water tissue content after high flows and high tissue water content after significant drying. It was also found that strong juvenile recruitment was also evident when body condition was high, indicating that when there are food resource spikes driven by flow pulses fish can utilise resources both to produce offspring and to store as lipid for future survival.

The ecology of dryland rivers is driven by their highly variable hydrology, particularly flooding regimes, whereby intermittent floods typically generate ‘booms’ of primary and secondary productivity, including massive fish production. We tested these concepts in the Moonie River, Australia, using the percichthyid, Macquaria ambigua, a dryland river species known to display pronounced ‘boom and bust’ abundance patterns in response to floodplain inundation followed by extended periods of low to no channel flow. We expected that body condition (as measured by whole body lipid content) and biomass of M. ambigua would be related to prey biomass, and that these factors would all ‘spike’ following widespread flooding. Instead we found more subtle responses. There were ‘booms’ in biomass of Macrobrachium and zooplankton, two important food items, whereas M. ambigua maintained relatively low but sustained lipid and biomass levels following flooding. It appears that instead of a ‘boom’ in fish biomass, abundant invertebrate food resources and sustained lipid levels contributed to high survivorship of this species during the ‘bust’ period over cool dry months.

Floodplain rivers in arid and semi-arid regions may be the most threatened of all river systems because water resource developments typically dampen their most distinctive characteristics-extreme flow variability and 'boom and bust' ecological dynamics. This article shows how one of the world's most variable arid-zone river systems-Cooper Creek in Australia's Lake Eyre Basin-functions and how it supports its unique fish assemblage and productive fisheries. The ecological roles of drought refugia, channel flows and flooding are reviewed in relation to fish persistence and losses, life history strategies, movement potential, food web processes and production levels. Comparisons are drawn with other floodplain rivers and fisheries to draw out common understandings and universal principles for conservation and management of arid-zone rivers and their fish resources. Ecological implications of hydrologic alterations and land-based activities are presented to highlight the importance of maintaining the hydrologic, geomorphic, sedimentary and biogeochemical processes of arid-zone river systems. Preservation or restoration of natural flow intermittency, sequential flood pulses, complex habitat mosaics, connectivity and identification of the environmental flow requirements for highly valued species and processes are key scientific principles for the management of arid-zone floodplain rivers.

Hydrological regime, physical habitat structure and water chemistry are interacting drivers of fish assemblage structure in floodplain rivers throughout the world. In rivers with altered flow regimes, understanding fish assemblage responses to flow and physico-chemical conditions is important in setting priorities for environmental flow allocations and other river management strategies. To this end we examined fish assemblage patterns across a simple gradient of flow regulation in the upper Murray-Darling Basin, Australia. We found clear separation of three fish assemblage groups that were spatially differentiated in November 2002, at the end of the winter dry season. Fish assemblage patterns were concordant with differences in water chemistry, but not with the geomorphological attributes of channel and floodplain waterholes. After the summer-flow period, when all in-channel river sites received flow, some floodplain sites were lost to drying and one increased in volume, fish assemblages were less clearly differentiated. The fish assemblages of river sites did not increase in richness or abundance in response to channel flow and the associated potential for increased fish recruitment and movement associated with flow connectivity. Instead, the more regulated river's fish assemblages appeared to be under stress, most likely from historical flow regulation. These findings have clear implications for the management of hydrological regimes and the provision of environmental flows in regulated rivers of the upper Murray-Darling Basin.

Excess nitrogen (N) leading to the eutrophication of water and impacts on ecosystems is a serious environmental challenge. Wetlands can remove significant amounts of N from the water, primarily through the process of denitrification. Most of our knowledge on wetland denitrification is from temperate climates; studies in natural tropical wetlands are very scarce. We measured denitrification rates during a dry and a wet season in five floodplain forests dominated by Melaleuca spp., a coastal freshwater wetland of tropical Australia. We hypothesised that the denitrification potential of these wetlands would be high throughout the year and would be limited by N and carbon (C) availability. Mean potential denitrification rates (Dt) were 5.0±1.7mgm2h–1, and were within the reported ranges for other tropical and temperate wetlands. The rates of Dt were similar between the dry and the wet seasons. From the total unamended denitrification rates (Dw, 3.1±1.7mgm2h–1), 64% was derived from NO3– of the water column and the rest from coupled nitrification–denitrification. The factor most closely associated with denitrification was background water NO3–-N concentrations. Improved management and protection of wetlands could play an important role in improving water quality in tropical catchments.

We used customized fish tanks as model fish ponds to observe grazing, swimming, and conspecific social behavior of common carp (Cyprinus carpio) under variable food-resource conditions to assess alterations in feeding niche. Different food and feeding situations were created by using only pond water or pond water plus pond bottom sediment or pond water plus pond bottom sediment and artificial feeding. All tanks were fertilized twice, prior to stocking and 2 weeks later after starting the experiment to stimulate natural food production. Common carp preferred artificial feed over benthic macroinvertebrates, followed by zooplankton. Common carp did not prefer any group of phytoplankton in any treatment. Common carp was mainly benthic in habitat choice, feeding on benthic macroinvertebrates when only plankton and benthic macroinvertebrates were available in the system. In the absence of benthic macroinvertebrates, their feeding niche shifted from near the bottom of the tanks to the water column where they spent 85% of the total time and fed principally on zooplankton. Common carp readily switched to artificial feed when available, which led to better growth. Common carp preferred to graze individually. Behavioral observations of common carp in tanks yielded new information that assists our understanding of their ecological niche. This knowledge could be potentially used to further the development of common carp aquaculture.

Although interest in the ecological impacts of urbanisation has increased, very little is known about its impacts on freshwater turtles in Australia. This study investigated the abundance and diversity of turtles in lakes in an urbanised subtropical landscape, Brisbane, Australia. It was found that turtles were abundant throughout the lakes surveyed, with four native species and one introduced species being detected. A total of 371 individuals were captured, 77.7% of which were Brisbane River turtle (Emydura macquarii signata). The lakes surveyed were estimated to support a population of 63-269 individuals for all species combined with most lake populations containing less than 100. Turtle demographics may be classed into three stages of population growth: recovery or nascent; intermediate; and climax. Turtle reproductive success is a major concern for the survival of turtles in urban areas. Continued monitoring is needed to establish if population declines are occurring. This study suggests that while turtles are persisting in this urban environment of Australia the pressures of urbanisation, such as habitat loss, increased predation on both nests and juveniles and pollution of waterways may pose on-going risks to their survival.

Summary 1. Floodplain inundation provides many benefits to fish assemblages of floodplain river systems, particularly those with a predictable annual flood pulse that drives yearly peaks in fish production. In arid-zone rivers, hydrological patterns are highly variable and the influence of irregular floods on fish production and floodplain energy subsidies may be less clear-cut. To investigate the importance of floodplain inundation to a dryland river fish assemblage, we sampled fish life stages on the floodplain of Cooper Creek, an Australian arid-zone river. Sampling was focused around Windorah during a major flood in January 2004 and in isolated waterholes in March 2004 following flood drawdown. 2. Of the 12 native species known to occur in this region, 11 were present on the floodplain, and all were represented by at least two of three life-stages – larvae, juveniles or adult fish. Late stage larvae of six fish species were found on the floodplain. There were site-specific differences in larval species assemblages, individual species abundances and larval distribution patterns among floodplain sites. 3. Significant growth was evident on the floodplain, particularly by larval and juvenile fish, reflecting the combination of high water temperatures and shallow, food rich habitats provided by the relatively flat floodplain. 4. Low variation in biomass, species richness and presence/absence of juvenile and adult fish across four floodplain sites indicates consistently high fish productivity across an extensive area. 5. Similarities and differences in fish biomass between the floodplain and isolated post-flood waterholes suggest high rates of biomass transfer (involving the most abundant species) into local waterholes and, potentially, biomass transfer by some species to other waterholes in the catchment during floodplain inundation and after floods recede. 6. The high concentration of fish on this shallow floodplain suggests it could be a key area of high fish production that drives a significant proportion of waterhole productivity in the vicinity. The Windorah floodplain provides favourable conditions necessary for the spawning of some species and juvenile recruitment of the majority of species. It is also appears to be a significant conduit for the movements of fish that underpin high genetic similarity, hence population mixing, of many species throughout the Cooper Creek catchment. The high floodplain fish production in turn provides a significant energy subsidy to waterholes after floodwaters recede. 7. The identification of key sites of high fish production, such as the Windorah floodplain, may be important from a conservation perspective. Key management principles should be: maintenance of the natural flooding regime; identification of the most productive floodplain areas; and maintenance of their connectivity to anastomosing river channels and the remnant aquatic habitats that ultimately sustain this fish assemblage through long-term dry/drought and flood cycles.

The abundance of epiphytic invertebrates living on both submerged and emergent macrophytes in lentic systems is influenced by a range of environmental features at both spatial and temporal scales. Such features include water chemistry, habitat complexity, hydrology and external influences such as climatic cycles. Dugays 2 billabong is a floodplain lake on the highly regulated, mid-upper Murray River in southeastern Australia. Giant rush (Juncus ingens N.A. Wakefield) is the dominant habitat feature in this billabong. This plant is a simple, single-stemmed, densely packed emergent macrophyte that fringes the littoral margins. The macrophyte habitat available to epiphytic invertebrates in this billabong is highly variable through time largely due to the variable water levels imposed by the regulated river channel for downstream irrigation needs, particularly through summer. To investigate both spatial and temporal patterns in the epiphytic invertebrates of Dugays 2 billabong, invertebrates were sampled within giant rush stands on five occasions over one year. Spatial variability in epifaunal density was generally low, while temporal densities were highly variable through time. Spatial patterns of invertebrate abundance revealed few associations with habitat structure or water quality, apart from those associated with water depth, particularly when sampling had been preceded by fluctuating water levels. High temporal variability in assemblage structure of the epifauna reflects the high background environmental variation that occurs in this system. Copyright © 2007 John Wiley & Sons, Ltd.

To examine temporal variation in diet, two size classes of western carp gudgeon Hypseleotris klunzingeri were collected from macrophyte habitats in Dugays 2 billabong, a floodplain lake on the River Murray, Victoria, Australia, for 1 year. Fish were collected bimonthly during the day and at night between October 1995 and September 1996. Western carp gudgeon consumed a range of prey, including caddisflies, odonates, shrimps and detritus. The bulk of their diet volume, however, consisted of chironomids and zooplankton. The diets of small and large western carp gudgeon were similar and did not change from day to night. There was, however, high variability across sampling trips in fish diets, both at the community level and among selected prey groups, particularly chironomids, zooplankton and detritus. Community analysis revealed that diets of fish collected from the billabong following stable antecedent water levels were similar and showed little variation, while those collected following fluctuating antecedent water levels showed much more variation. These patterns were mirrored in the volumes of individual diet components.

The fish assemblages of an arid zone floodplain river, Cooper Creek, Queensland, Australia, were sampled during two dry periods in isolated waterholes and on the inundated floodplain during the early and late phase of a major flood event. Diets were described for nine native species and compared within and between dry and flood periods. In the dry season, when fishes were restricted to waterholes, diets were characteristically simple with narrow diet breadths. Movement onto the floodplain during flooding clearly increased feeding opportunities, with greater diet breadths evident in all species. Despite obvious potential for terrestrial inputs, diets tended to be dominated by aquatic resources in both the waterholes and on the floodplain. Stomach fullness, however, varied little between dry season waterhole and floodplain samples. Fish appear to feed on potentially lower value resources such as detritus and calanoid copepods during the dry season, when waterholes are isolated and food resources were limited. They were then able to capitalize on the "boom" of aquatic production and more diverse food resources associated with episodic flood events.

Highly variable recruitment is common for many riverine fish species, governed by a wide range of biotic and abiotic drivers that operate at local and regional scales. The dynamics and drivers of recruitment for many Australian freshwater fish species, particularly those that are rare and long-lived, are relatively undescribed. This study describes the recruitment dynamics of an endangered riverine fish, Macquarie perch Macquaria australasica, across five isolated populations from south-eastern Australia, and relates these dynamics to drivers that vary at local and regional scales. We hypothesised large flow events occurring during the core egg and larval period would be negatively associated with recruitment strength and that recruitment patterns across populations would fluctuate in synchrony in response to extremes in river discharge resulting from regional-scale climatic patterns. Discharge during the core egg and larval period, which was highly correlated across the region; and a local scale variable, spawning stock abundance, were the covariates most important in explaining recruitment strength. We also observed synchronised patterns in recruitment across our populations, thus conforming to predictions of the Moran effect (environmental synchrony). The findings suggest most remnant populations of Macquarie perch, which are now predominantly isolated within small tributary systems characterised by highly variable flows, face a heightened risk of poor recruitment periods, particularly under climate change predictions. The synchronised patterns in recruitment suggests threatened freshwater fishes like Macquarie perch with highly fragmented isolated populations, have an increased risk of the regional population becoming imperilled, thus, the need for a coordinated multijurisdictional conservation approach.

Hydrology is widely considered to be the dominant driver of understorey vegetation dynamics in arid and semiarid riparian ecosystems. Consequently, environmental watering is often a key approach to restoring and promoting riparian vegetation diversity. Few studies however have considered the role of local factors, for example, shading or litter accumulation, on vegetation responses to flow and how these may influence the outcomes of watering actions. Knowledge concerning modifying effects of local canopy factors on vegetation responses to watering is needed to guide environmental water planning (e.g., setting objectives and targets) and delivery (e.g., timing and duration of managed flows). We conducted a greenhouse experiment to investigate the effects of shade and litter, and their interaction, on plant communities establishing from riparian soil seed banks, collected from the northern Murray–Darling Basin, under a range of hydrological conditions. We found that shade had a positive influence on the abundance, diversity, and reproduction of establishing plant communities under dry conditions but a negative, or null effect, under wet conditions. Litter exerted strong negative effects on plant community metrics under all hydrological conditions. Some understorey species emerged as hardy generalists capable of establishment under a range of conditions whereas others had more specific requirements. Our results highlight the importance of canopy structure for riparian understorey vegetation diversity and dynamics at both local and landscape scales. Land and water management practices seeking to protect or restore vegetation diversity in these systems must therefore take such local factors into account when planning and evaluating interventions.

Eleotrid fish (gudgeons) commonly dominate the small-fish fauna throughout much of the Indo-Pacific region. Eleotrids frequently occupy habitats comprised of dense emergent macrophytes, making estimation of fish abundance difficult. In this study, the suitability of unbaited fish traps to measure carp gudgeon (Hypseleotris spp.) abundance in giant rush habitats was studied in a River Murray floodplain billabong using depletion trapping. Traps were repeatedly set at ten locations within giant rush beds over a two-day period. There was no significant variation in carp gudgeon catch rate over time within locations, but variation between locations was significant. It was not possible to deplete fish from any location over two days of trapping, suggesting rapid replacement of trapped fish in sampled locations. Patch specific variation in the gudgeon catch rate among different traps in macrophyte beds may reflect small-scale variation in habitat quality.

Two killing methods were compared on the clupeid, bony bream Nematolosa erebi and it was found that ice-slurry immersion was more humane than benzocaine overdose. The use of ice-slurry for killing N. erebi should be accepted as a standard humane method and considered similarly for other warm-water species.

Summary The recent description of complex food webs that contain many species abundant omnivores, relatively long food chains and have a high number of links per species has cast significant doubt over our understanding of the relationships between food-web stability, interaction strength and overall web complexity. Complex food webs contradict previous theoretical observations that highly complex webs should also be unstable and hence less likely to persist. The description of a high links per species ratio in food webs suggests that generalist predators are a common feature of many systems. Despite the abundance of such predators, there have been few considerations of their role in food-web dynamics. Theoretical and empirical studies of generalist predators suggest that their role in food-web dynamics may range from being highly destabilising to having little impact on food-web dynamics. We argue that a full understanding of the role of generalist predators in food-web dynamics can only be developed when their actions are considered within a broad spatial and temporal framework. In contrast to specialist predators, generalist predators have the capacity to switch to alternative prey as preferred prey decline. Consequently, the predator's population dynamics are not necessarily closely coupled to that of their preferred prey. When, considered over short temporal scales and at small spatial scales, such a predator is clearly destabilising as local extinction of vulnerable prey may occur given that the predator's abundance can be sustained by access to alternative prey. Species loss from a community is, however, a phenomena that is dependent on the spatial and temporal scales at which it is viewed. It cannot continue indefinitely and, once predators have eliminated vulnerable prey, those that remain are likely to be little affected by that predator's actions even though significant numbers of prey individuals may still be consumed by that predator. Remaining prey may persist within the predator's habitat, through protection within refugia or alternatively by maintaining reproduction rates that exceed the rate of predation. The food web persists in a stable, albeit altered form, but is now dominated by weak, effectively donor-control, interactions. We review evidence for this perspective of generalist predation and food-web dynamics. Consideration of the role of generalist predators in food webs suggests that dynamic constraints on food-web structure may only apply in food-webs with few species, that the relative importance of top-down and bottom-up regulation of food-web structure may be strongly linked to the availability of refugia within a habitat, and that the strength and energetic importance of a link may, to some degree, be inversely related.

Fish living in highly variable and unpredictable environments need to possess life-history strategies that enablethemtosurviveenvironmentalextremessuchasfloodsanddrought.Weusedthelength-frequencydistributionsof multiple fish species in multiple seasons and highly variable hydrological conditions to infer antecedent breeding behaviourinriversoffar-westernQueensland,Australia.Hypothesestestedwereasfollows:(1)recruitmentofsomeorall speciesoffishwouldoccurwithinwaterholesduringno-flowperiods;(2)therewouldbeseasonalrecruitmentresponsesin some fish species; (3) recruitment of some species would be enhanced by channel flows and/or flooding. Hydrology and the incidence of flooding were highly variable across the study area during 2006-2008. Flood-influenced recruitment was evident for Hyrtl's tandan, Barcoo grunter and Welch's grunter. Silver tandan, golden goby, Cooper Creek catfish and Australian smelt showed evidence of seasonal recruitment unrelated to antecedent hydrology. However, most species demonstrated continual recruitment in isolated waterholes, irrespective of antecedent flow conditions and season. Continual and seasonal recruitment capabilities have obvious advantages over flood-pulse recruitment in rivers with highly unpredictable flood regimes and underpin the persistence of many fish species in arid and semiarid rivers. Additional keywords: ephemeral rivers, fish life-history variation, flood pulse concept, Lake Eyre Basin, no-flow recruitment hypothesis.

Many aquatic ecosystems have been severely degraded by water-resource development affecting flow regimes and biological connectivity. Freshwater fish have been particularly affected by these changes and climate change will place further stress on them. The Murray–Darling Basin (MDB), Australia, represents a highly affected aquatic system with dramatically modified flow regimes. This has impaired the health of its rivers, and potentially limited the adaptive capacity of its biota to respond to a changing climate. Here, we present our predictions of the potential impacts of climate change on 18 native fish species across their distributional ranges against the back-drop of past and continuing water-resource development (WRD). Because most of these species are found across a wide range of geographical and hydrological settings, we classified the MDB into 10 regions to account for likely variation in climate-change effects, on the basis of latitude, elevation and WRD. Cold water-tolerant species will be under greater stress than are warm water-tolerant species. In some regions, the negative impacts on exotic fish such as trout are likely to improve current conditions for native species. Because the impacts of climate change on any given species are likely to vary from region to region, regional fish assemblages will also be differentially affected. The most affected region is likely to occur in the highly disturbed Lower Murray River region, whereas the dryland rivers that are less affected in the northern MDB are likely to remain largely unchanged. Although climate change is a current and future threat to the MDB fish fauna, the continued over-regulation of water resources will place as much, if not more, stress on the remnant fish species.

Dryland rivers are renowned for their periods of ‘boom’ related to the episodic floods that extend over vast floodplains and fuel incredible production, and periods of ‘bust’ where the extensive channel network is restricted to the permanent refugial waterholes. Many of these river systems are unregulated by dams but are under increasing pressure, especially from water abstraction and overland flow interception for agriculture and mining. Although some aquatic organisms with desiccation-resistant life stages can utilise ephemeral floodplain habitats, the larger river waterholes represent the only permanent aquatic habitat during extended periods of low or no flow. These waterholes act as aquatic refugia in an otherwise terrestrial landscape. Variable patterns of connection and disconnection in space and time are a fundamental driver of diversity and function in these dryland river systems, and are vital for dispersal and the maintenance of diverse populations, generate the spatial and temporal variability in assemblage structure for a range of different organisms and fuel the productivity that sustains higher trophic levels. Changes to natural patterns of connection and disconnection of refugial waterholes, owing to water-resource development or climate change, will threaten their persistence and diminish their functional capacity to act as aquatic refugia.

Many rivers experience intermittent flows naturally or as a consequence of water abstraction. Climate change is likely to exacerbate flow variability such that dry spells may become more common. It is important to understand the ecological consequences of flow intermittency and habitat fragmentation in rivers, and to identify and protect habitat patches that provide refugia for aquatic biota. This paper explores environmental factors influencing dry season fish losses from isolated waterbodies in Cooper Creek, an unregulated arid-zone river in the Lake Eyre Basin, Australia. Multivariate ordination techniques and classification and regression trees (CART) were used to decompose species–environment relationships into a hierarchically structured data set, and to determine factors explaining changes in fish assemblage structure and species losses over a single dry season. Canonical correspondence analysis (CCA) explained 74% of fish assemblage change in terms of waterhole morphology (wetted perimeter, depth), habitat structure (bench development, off-take channels), waterhole quality (eroded banks, gross primary production), the size of surrounding floodplains and the relative isolation of waterholes. Classification trees for endemic and restricted species reaffirmed the importance of these waterhole and floodplain variables as drivers of fish losses. The CCA and CART models offer valuable tools for identification of refugia in Cooper Creek and, possibly, other dryland rivers.

Riverine fish living in unpredictable flow environments tend to be ecological generalists with traits that allow them to persist under highly variable and often harsh conditions associated with hydrological variation. Cooper Creek, an Australian dryland river, is characterised by extreme flow variability, especially in the magnitude, timing and duration of channel flows and floods, which, if they occur, do so mainly in summer. The present study examined the influence of hydrological variability on fish assemblages and abundance in four waterholes in the Windorah reach of Cooper Creek over eight occasions between 2001 and 2004. Antecedent flows had marked influences on fish species richness and assemblage structure. Following high summer flows, all waterholes supported a rich and abundant fish fauna, whereas fewer species and lower numbers were recorded following periods of zero channel flow. Recruitment of three of the four most common and abundant species was enhanced when intermittent flows inundated backwater and floodplain habitats that provide a food-rich environment. Opportunistic responses to rising channel flows and occasional large floods in Cooper Creek help to explain the prominent ‘boom’ patterns of fish production in this arid-zone river, whereas low-level recruitment during periods of low or no flow maintains populations of some species through the ‘bust’.

Dryland rivers are characterised by highly pulsed and unpredictable flow, and support a diverse biota. The present study examined the contribution of floodplain sources to the productivity of a disconnected dryland river; that is a waterhole, after a major overland flood event. Rate measures of productivity were combined with stable isotope and biomass data on the food web in the waterhole and floodplain. The present study estimated that 50% of the fish carbon in the waterhole after flooding was derived from floodplain food sources. In the few months after retraction of the river to isolated waterholes, the large biomass of fish concentrated from the flooding decreased by 80%, most likely as a result of starvation. Based on the development of a carbon budget for the waterhole, mass mortality is hypothesised to be the cause of the high rates of heterotrophic production in the waterhole. The present study suggests that floodplain inputs are important for fuelling short-term production in waterholes, but via an unconventional pathway; that is, fish mortality. The episodic nature of flooding in dryland rivers means that changes in flow regimes, such as water regulation or abstraction, will reduce flooding and hence floodplain subsidies to the river. This is likely to have significant impacts on river productivity.

To examine how food resource availability links with natural variation in primary productivity in the Moonie River, south-west Queensland, the diets of two native Australian fish species (Nematalosa erebi and Macquaria ambigua) were examined from fifteen waterholes in February, May and September 2006. N. erebi diets reflected strong ‘boom and bust’ patterns of food consumption, with high concentrations of benthic (non-filamentous) algae during boom (flow) times, moving to higher concentrations of filamentous algae and detritus during bust (no flow) periods. M. ambigua diets were primarily dominated by aquatic insects in all sampling periods. Although there was no clear ‘boom to bust’ pattern in relation to flow, M. ambigua secondary prey consumption revealed a compensatory switch between high energy prey (crustaceans) during more productive periods with terrestrial insects during less productive periods. The ability of both species of fish to switch from high to low concentrations of food quality under a variable environmental background allows them to persist through both high productive and low productive periods. This interaction between native biota and variable ‘boom’ and ‘bust’ conditions, and how changes to the natural hydrology will affect it is an important consideration of any future water resource development plans.