SEARCH

SEARCH BY CITATION

Keywords:

  • climate change;
  • lateral connectivity;
  • longitudinal connectivity;
  • Marxan;
  • refuge;
  • resilience

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  1. Addressing spatial connectivity in conservation planning is important to ensure the maintenance of patterns and processes needed to support the persistence of biodiversity. In freshwater ecosystems, spatial connectivity is constrained by the presence of water, which exhibits marked temporal changes in regions with wet–dry seasonal climates. Previous studies have focused on spatial connectivity and overlooked the temporal component, which is required for the functionality of spatial connections (because of temporal changes in water availability).
  2. We identify priority areas for the conservation of freshwater fish, waterbirds and turtles in the Mitchell River catchment in the wet–dry tropics of northern Australia. We demonstrate how adequacy of freshwater conservation can be enhanced by integrating an estimate of water residency time (WRT) into the prioritization process. WRT reflects refugial potential and connectivity in freshwater ecosystems and was quantified using Moderate Resolution Imaging Spectroradiometer (MODIS) flood and post-flood Landsat satellite imagery. We compare the spatial allocation of priority areas and the spatial and temporal connectivity under two alternative scenarios: (i) accounting only for spatial connectivity and (ii) integrating spatial and temporal connectivity.
  3. Priority areas identified under the spatial and temporal connectivity scenario showed a 40% increase in WRT values with respect to the traditional spatial connectivity scenario. This was achieved at no additional cost in terms of total protected area and maintaining the same spatial connectivity.
  4. Despite priority areas identified under the two alternative scenarios showing intermediate spatial overlap (64%), the selection process was more efficiently biased towards planning units with high WRT values. WRT in planning units that were only selected under the temporal connectivity scenario was 2·5 times higher than in planning units that only appeared in the traditional connectivity scenario. This reveals the importance of accounting for WRT when identifying freshwater priority areas in wet–dry seasonal environments.
  5. Synthesis and applications: Considering the temporal connectivity in conservation prioritization as we propose here helps to assess periods of longest spatial connections, thereby maximizing the refugial role of freshwater priority areas during dry periods. Using publicly available satellite imagery data and software, our approach allows improved management of aquatic resources and biodiversity during periods of water scarcity, which may increase in incidence and duration with climate change.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Connectivity is an important consideration in conservation planning to ensure the maintenance of the patterns and processes needed to support the long-term persistence of biodiversity (Pressey et al. 2007). Hydrological connectivity is a crucial determinant of ecosystem structure and functioning in freshwater habitats (Ward 1989; Pringle 2001; Hughes, Schmidt & Finn 2009) and drives linkages among ecosystem elements in space and time (Fullerton et al. 2010). Hydrological connectivity is featured in three spatial dimensions (Ward 1989): longitudinal (headwater–mouth), lateral (riverine–floodplain) and vertical (surface–ground water). This is critical for maintaining food web dynamics and ecosystem function at a range of spatial scales and the persistence of freshwater populations (Pringle 2001; Bunn & Arthington 2002; Fausch et al. 2002). For example, longitudinal fluxes of organic material are a distinctive feature of riverine food webs, with fluxes occurring between headwater streams, main channel reaches and estuaries. Hydrological connectivity is also vital for the persistence of some species, such as diadromous fish that migrate long distances to complete their life cycles, or to allow re-colonization in seasonal systems after periodic low- or high-flow disturbances (Albanese, Angermeier & Peterson 2009).

The connected nature of freshwater ecosystems poses special challenges for effective conservation management, as human-caused disturbances in one area can seriously threaten freshwater biodiversity far away (e.g. upstream or downstream). For this reason, connectivity has been a focus of attention for freshwater conservation planning (Linke, Turak & Nel 2011). Different approaches have been proposed to address longitudinal (Linke et al. 2007; Moilanen, Leathwick & Elith 2008; Hermoso et al. 2011), lateral (Thieme et al. 2007; Ausseil et al. 2011; Hermoso, Kennard & Linke 2012) and vertical (Nel et al. 2011) components of connectivity. Despite these efforts, some aspects of connectivity have not been adequately accounted for in conservation planning approaches. In freshwater ecosystems, spatial connectivity, in all of its dimensions, is limited by the presence of water, which is strongly influenced by temporal dynamics. In this way, functional connectivity (facilitation of movement of organisms through the landscape, sensu Taylor, Fahrig & With 2006) is constrained by the presence of a minimum amount of water in the landscape (Fig. 1). This applies not only to obligate freshwater species restricted to movements within river networks, but also to organisms with aerial or terrestrial movements (e.g. spatial proximity of floodplain lakes, important for some birds and turtle species; Haig, Melhman & Oring 1998; Rea, Brinton & Georges 2009). In climatic regions with pronounced wet to dry seasonality, such as the wet–dry tropics (Hamilton, Sippel & Melack 2002), or the Mediterranean (Gasith & Resh 1999), dry periods constrain the presence of water and the connections between the different elements of the system. Rivers and streams that are hydrologically connected during the wet season may cease to flow during the dry season and become reduced to a set of isolated pools. Important dry season refugia such as floodplain wetlands and lakes are also affected, becoming completely dry or constrained to a reduced area (Fig. 1). Water availability does not only fluctuate within annual hydrological cycles but also inter-annually, and extended dry periods (droughts) may periodically occur (Gasith & Resh 1999; Bond, Lake & Arthington 2008). The frequency, duration and intensity of these events are expected to increase under the effects of climate change and some freshwater ecosystems that are currently perennial are likely to undergo more pronounced wet and dry cycles in the future (Bates et al. 2008).

image

Figure 1. Change in water level and connectivity potential across a hydrological cycle for a perennial and a seasonal (dotted line) water body. Perennial water bodies offer refuge to freshwater biota for longer periods of time (t′) than seasonal waterbodies (t) and allow movements for longer periods (the minimum water level to allow connections remains for most of the hydrological cycle). Lines on top of the figure represent the extent in time in the two water bodies.

Download figure to PowerPoint

Temporal changes in spatial connections can have important implications for population dynamics and community structure (Taylor, Fahrig & With 2006), and ultimately for population viability (Magalhães et al. 2007). For example, connectivity can affect reproduction and mortality by allowing or limiting access to potential breeding sites (Taylor, Fahrig & With 2006). Incorporating the temporal dimension of connectivity (Pringle 2001) in freshwater conservation planning is therefore a crucial challenge to adequately address the persistence of freshwater biodiversity (Nel et al. 2009).

Here, we demonstrate how to integrate a measure of water residency time (WRT) in the landscape into previous considerations of spatial connectivity in conservation planning methods. We consider two different components of spatial connectivity: longitudinal connectivity within a catchment, and lateral connectivity between different subcatchments. These types of connectivity were applied to address the spatial movement requirements of fish, turtles and waterbirds in the Mitchell River catchment (northern Queensland, Australia). In this way, we aimed to find a set of planning units that maintained water (i.e. to maximize the refugial role of freshwater priority areas) and allowed connections to be effective over the longest possible period of the hydrological cycle to enable movement and access to key aquatic resources (functional connectivity). This novel approach to incorporating WRT in conservation planning methods would result in increased adequacy (capacity to maintain biodiversity over the long term) of identified freshwater priority areas and higher likelihood of future persistence of freshwater-dependent biota.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Study area

The Mitchell River catchment (71 630 km2) is located in northern Queensland, Australia (Fig. 2). The wet–dry tropical climate of the region is largely controlled by the equatorial southern monsoon. It is strongly seasonal with >80% of the annual rainfall occurring between the wet season months of December to March (summer). Mean annual rainfall increases from around 600 mm in the south to over 1200 mm in the northeast and northwest. High mean annual evapotranspiration leads to annual water deficits across the catchment except in the very wettest of years (Ward et al. 2011).

image

Figure 2. Location of the Mitchell River catchment in northern Australia and water residency time (WRT) for each of the 2316 planning units used in this work. WRT was calculated as shown in Fig. 3.

Download figure to PowerPoint

Biodiversity data

The spatial distribution of fish, waterbirds and turtle species was sourced from Kennard (2010). This database contained predictions of spatial distributions for 104 species of freshwater fish, 106 waterbird species and 13 turtle species across northern Australia derived from Multivariate Adaptive Regression Splines models (Leathwick et al. 2005) at a fine scale (average area of predictive units was 3·6 km2). The predictive model was built on a data set of 1609 presence-only sites plus 115 presence–absence sites and validated in an independent data set of 604 presence–absence sites for fish, and built and validated on 2109 and 350 presence-only sites for birds and turtles respectively (see Kennard 2010 and Hermoso, Kennard & Linke 2012 for more details on predictive models).

Identification of priority areas

We used the software marxan (Ball, Possingham & Watts 2009) for identifying priority areas for conservation of freshwater fish, waterbirds and turtles. marxan tries to find an optimum set of planning units that achieve the requested representation for each species (e.g. 100 km2, hereafter termed conservation target) at the minimum cost. We used subcatchments as planning units rather than equal-sized grid cells as generally used in terrestrial and marine conservation assessments. We delineated 2316 planning units from a 9-s digital elevation model using ARC Hydro for arcgis 9·3 (ESRI 2002). We then translated the information from the predictive models for each of the 45 freshwater fish, 106 bird and five turtle species inhabiting the Mitchell River into the planning units by summing the area where each species was predicted to occur within each planning unit. Given the lack of ecological knowledge to guide more objective conservation target setting, we set conservation targets to ensure the adequate representation of the rarest species in the data set (the use of constant% targets for all the species, as more commonly used, could lead to the underrepresentation of these rare species and overrepresentation of common ones). We adapted the target according to each species' prevalence in the data set to represent 100% of the spatial distribution of the rarest species, and then scaled them down exponentially to reach a minimum of 10% for the most common species. In this way, the spatial distribution of rare species would be almost fully covered in solutions. To account for the importance of connectivity in freshwater conservation planning and the different requirements of the biodiversity considered in this study, we incorporated two different types of spatial connections in the analyses. We used the longitudinal connectivity rule proposed in Hermoso et al. (2011) to account for the longitudinal propagation of threats and movements along the river network. Moreover, we incorporated intercatchment connectivity (Hermoso, Kennard & Linke 2012) to account for movements of birds or turtles that are not necessarily restricted to movement via river networks (i.e. can move across drainage divides). We ran marxan 100 times (1·5 M iterations each) to find 100 near-optimal solutions. The number of iterations per run was set after a sensitivity analysis to ensure convergence (measured as the average of planning units in the best solution) of the optimization algorithm. The solution with the lowest value for the objective function was used as the best solution in subsequent analyses. The frequency of occurrence of each planning unit across the 100 near-optimal solutions was used as an estimate of their irreplaceability [the likelihood that an area will be required to meet a given set of targets, Pressey, Johnson & Wilson (1994)].

Addressing temporal connectivity

We incorporated an estimate of water residency time (WRT) within each planning unit to represent spatial connectivity between planning units. In this way, we aimed to maximize not only the spatial connection between planning units but also the time those connections provided continuous aquatic habitat because of the presence of water. We used Moderate Resolution Imaging Spectroradiometer (MODIS) flood imagery (250-m resolution) (http://modis.gsfc.nasa.gov/), captured for February and March 2009, to map seasonal flood inundation extents, and post-flood Landsat TM 5 imagery (30-m resolution) (http://landsat.usgs.gov/), captured for March, June and October 2009, to map seasonal change in waterbody extents. A 20-year Landsat TM 5 dry season time series, captured in the late dry season in any year, was used to map dry season perennial waterbodies defined as being dry in less than one in 10 years (Environmental Protection Agency 2005). This resulted in a total of eight surface water extent snapshots in 2009, comprising four flood event snapshots, three dry season water extent snapshots and one long-term estimate of the extent of perennial water at the end of the dry season. The long-term estimate of the extent of perennial waterbodies was included because cloud cover at the beginning of the wet season prevents image capture of water extent just prior to the onset of flooding. Although there is interannual variability in the magnitude of annual flood inundation in this region, the timing and duration of the wet season inundation, and the dry season are highly regular in the study region (Kennard et al. 2010). We then summarized the information on the seasonal dynamics of surface water extents contained in grid cells for each planning unit. The main issue in doing this was how to capture the variability in WRT between planning units. Halls et al. (2008) propose a flood index based upon the sum of the number of days that a given lake area is flooded. We adapted this index as a measure of WRT, calculated as the surface water area–time integral, which measures the area under the curve of the relationship between surface water area and time it remains for each planning unit (Fig. 3). The measure of WRT used here was estimated using the trapezoidal rule (Atkinson 1989). The resulting WRT unit is area–time (e.g. hectare–days). A consequence of WRT calculated in this way is that the same area–time integral value can be obtained for a small amount of water that resides in a planning unit for a long time as well as for a large amount of water that resides for a short time. To address this issue, the surface water area–time integral used to calculate WRT was weighted by the proportion of time water resides in the planning unit. For each planning unit, the weight is calculated as ti/T, where ti is the time from the beginning of the time series to the ith surface water area measurement, and T is the total length of the time series.

image

Figure 3. Dynamics of Water Residence Time (WRT) in planning units with short and long WRTs (left and right figures respectively). WRT is calculated as the area under the curve of the relationship between surface water area and the time it remains for each planning unit.

Download figure to PowerPoint

We then used the WRT to weight the spatial (longitudinal and inter-subcatchment) connections in marxan. The longitudinal connectivity penalty accounts for connections along the river network using a distance-decay function, so for example, the penalty for not including a planning unit located far upstream of a focal area would be lower than if not including the most immediate upstream planning unit (Hermoso et al. 2011). The inter-subcatchment connectivity penalty incorporates connections not included in the upstream–downstream topology. This additional component of connectivity was recently incorporated into freshwater conservation planning to account for the ecological needs of freshwater taxa that are not strictly water dependent (e.g. waterbirds and some turtles that can move across the landscape; Hermoso, Kennard & Linke 2012). We multiplied the connectivity penalty described in Hermoso et al. (2011), Hermoso, Kennard & Linke (2012) by the minimum WRT for the pair of planning units involved in every connection (Fig. 4). In this way, if two planning units were spatially close but retained water only for a few weeks a year, a low penalty would apply (the spatial connection is very ephemeral and is therefore not considered as important). On the other hand, close planning units that retained water most of the year offered a high temporal-spatial connection. Given the difference in magnitude between the spatial penalties and WRT (Fig. 2), we used a log-transformed WRT to weight the spatial connectivity penalty. In this way, we ensured that we optimized for both, spatial and temporal connectivity, at the same time.

image

Figure 4. Examples of how the connectivity penalty (indicated by arrow thickness) was calculated by combining information on water residency time (WRT, indicated by shading on the river network) and spatial connectivity (longitudinal and inter-subcatchment combined). High connectivity penalties would apply if two close planning units that maintain water for most of the year are not selected at the same time. This penalty decays with distance between planning units (either longitudinal or inter-subcatchment) and the time that units do not maintain water.

Download figure to PowerPoint

To test the effect of WRT, we identified priority areas for conservation of freshwater biodiversity in the Mitchell River catchment under two different scenarios, with and without WRT-weighted spatial connectivity. We used each planning unit's area as a surrogate of cost in the analyses.

We measured the total length (km) of connected planning units within the best solution and the average WRT of planning units within and outside the best solution in both scenarios (with and without WRT) and for 100 best solutions randomly generated (random allocation of the same number of planning units across the study area). We also estimated the potential overall benefit of WRT to species represented in the best solution by measuring the WRT in selected planning units where each species occurred. In this way, we had an estimate of WRT in priority areas where each species was expected to be present. We then checked for significant differences in the average WRT of planning units in the best solution and the species' WRT benefit (with vs. without WRT) by means of a one-way anova. A conceptual diagram of the analyses performed can be found in Fig. S1 (Supporting information).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The spatial allocation of best solutions and highly irreplaceable areas in the Mitchell River catchment was influenced by the incorporation of WRT (Fig. 5). Both solutions provide reasonable identification of priority areas, as conservation targets were fully achieved in both cases. There was an intermediate overlap between both solutions with 64% of planning units (n = 165) selected in both scenarios. Despite this coincidence, the use of temporal connectivity had a significant effect in the selection of planning units. Planning units that were selected only under the temporal scenario (n = 52) showed WRT values 2·5 higher (average WRT = 4200) than planning units only selected under the traditional connectivity scenario (n = 40 planning units, average WRT = 1712). This was also reflected in the change of spatial patterns of irreplaceability (Fig. 5). High irreplaceable areas shifted from intermediate reaches to lowland reaches where WRT showed higher values on average (Fig. 2). There was a substantial difference in temporal connectivity between both scenarios (Fig. 6). The average WRT value for planning units included in the best solution when using WRT was higher (40% increase on average) than when it was not used (Fig. 6). The difference in spatial allocation of priority areas when using WRT did not affect the degree of spatial aggregation of priority areas, the total area required, or the achievement of conservation goals (all the species achieved the required target in all cases). In both cases, the average fully spatially connected river length within priority areas was around 322 km for each patch of clustered planning units, and the total area required was around 8000 km2 (Fig. 5, Table S1, Supporting information). Note that the total number of planning units included in the best solutions was similar with and without WRT (n = 205 and n = 216, respectively) and therefore unlikely to influence our assessment of the benefits of incorporating WRT. The preferential selection of planning units with high WRT in the best solution also translated into an increase in WRT where each species occurred within best solutions (Fig. 6). The average species WRT benefit was significantly higher when using WRT (anova, F1,304 = 48·3, < 0·001). See Table S1 (Supporting information) for a summary of results of comparisons between scenarios.

image

Figure 5. Best solutions and irreplaceability values for two planning scenarios: with and without water residency times (WRT) as weight to spatial connection.

Download figure to PowerPoint

image

Figure 6. (a) Mean (±SE) water residency time (WRT) within high priority areas (planning units selected in best solution) and low priority areas (not selected) for three different scenarios: planning with no WRT considerations, using WRT to weight spatial connectivity, and random selection of planning units (100 solutions containing 200 planning units). (b) Water residency time (WRT) for planning units where each species occurred within the best solution (mean ± SE across species) under the two scenarios tested in this work (with and without WRT as a weight to spatial connections).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

There is increasing concern among conservation practitioners to quantify and manage processes that support the persistence and functioning of ecosystems and their biodiversity (Meir, Andelman & Possingham 2004; Possingham et al. 2005; Pressey et al. 2007). Here, we have demonstrated how to integrate an estimate of WRT in the identification of priority areas for conservation to enhance the management of biodiversity. We have addressed two important features of aquatic ecosystems to improve adequacy of freshwater priority areas, namely the seasonality of spatial connectivity and the presence of dry season refuges. This will help maintain some key ecological processes such as persistence and movements of biota over the dry season or recolonization when water returns, which are essential factors to secure the long-term maintenance of freshwater biodiversity.

Native biodiversity in dynamic wet–dry seasonal freshwater environments have developed strategies (desiccation resistant life-history stages or retreating to refuge habitats, fast recolonization and recruitment after disturbance) to cope with extended dry periods (Bond, Lake & Arthington 2008). For example, Magalhães et al. (2007) found that fish assemblages in Mediterranean streams recovered quickly from short-term fluctuations in water level. However, they also warned that exacerbated disruptions in longitudinal connectivity caused by low water levels could negatively influence sensitive species. This is of particular concern under the expected effects of climate change on water availability (Bates et al. 2008). Global-scaled predictions include a two to threefold increase in the frequency of extreme low flows in many areas (Arnell 2003) and a reduction in mean annual discharge. Under these conditions, it is likely that some currently perennial freshwater ecosystems will become non-perennial and that the duration and extent of water scarcity in already wet–dry seasonal ecosystems will increase. This represents a challenge for conservation of freshwater biodiversity, because the new conditions will affect ecosystem functioning and might surpass some species' ability to adjust or adapt to these changes (Morrongiello et al. 2011).

Focusing conservation efforts in those areas that retain water and allow connections for long periods of time is an appropriate strategy to improve long-term persistence of freshwater-dependent biota. Here, we demonstrated a substantial increase in WRT within priority areas (40%) with no additional increase in the total area selected and maintaining the spatial connectivity constant (both scenarios showed similar number of planning units and length of river connected in the best solution). In this way, we focused the selection of priority areas in more perennial areas where freshwater biodiversity has a higher chance of resisting the dry period. There was a shift in the spatial allocation of irreplaceable areas when introducing WRT in the selection process. When using WRT to weight spatial connections, the selection of priority areas was focused on those zones that enabled movements along the river network (longitudinal connectivity, Hermoso et al. 2011) and between floodplain wetlands and lakes (inter-subcatchments connectivity, Hermoso et al. 2011) for a longer period of time. This result should satisfy seasonal connectivity requirements for a range of different species, such as waterbirds and turtles, between wetlands and lakes that maintained water or fish within river reaches where water levels provided functional connections.

By incorporating WRT in the identification of priority areas, we focus attention on areas that help not only to enhance biodiversity persistence by increasing the connectivity potential for biota within priority areas, it also provides relevant information regarding assisted translocations of sensitive biodiversity elements that otherwise face impassable migration barriers within and across catchments. This in turn, can promote resilience of freshwater biodiversity elements that are sensitive to increased frequencies and durations of droughts associated with a changing climate by focusing conservation efforts on potential refuge areas. This is an important feature in wet–dry seasonal ecosystems like those in the Mitchell River catchment in northern Australia. This catchment features a high mean annual rainfall and high evapotranspiration, which often leads to annual water deficits in some parts of the catchment. For example, a post-flood surface water contraction of 89% in area has been recorded between March and October in 2009 (D. Ward unpublished data). Seasonality in water availability was spatially heterogeneous with a pattern of increasing WRTs towards lower reaches of the Mitchell floodplain. Perennial aquatic habitats in these parts of the catchment function as biological refuges during dry periods and as re-colonization sources at the onset of flooding at the beginning of the next wet season (Magoulick & Kobza 2003). Priority areas identified in our study hold water for most of the dry period allowing species to have access to key resources when they are scarce (e.g. some waterbirds or turtles; Haig, Melhman & Oring 1998; Rea, Brinton & Georges 2009) and offer refugial opportunity (e.g. freshwater fish; Arthington et al. 2005). This should help address the ecological needs of some species and sustain populations that will eventually serve as sources of colonization when water returns water returns. The potential recolonization from refuge areas might be compromised by the presence of barrier to the movement, especially for water strict species (e.g. dams). Although this was not a major issue in our study area, we acknowledge that it should be further considered when estimating the potential benefit of a given refuge area in other catchments where barriers constraint biota movements.

The method proposed here is transferable and could be used to address adequacy in freshwater conservation planning in regions of the world subject to strong seasonality in water availability. Satellite imagery and the software marxan used in this work are publicly available, which facilitates the application of the method we demonstrate here.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank J. Stein for the help with the environmental characterization of planning units and B. Pusey, D. Burrows, P. Bayliss and J. Boyden for assistance with compilation of fish and waterbird data. We acknowledge the Australian Government Department of Sustainability, Environment, Water, Population and Communities, the National Water Commission, the Tropical Rivers and Coastal Knowledge (TRaCK) Research Hub, the National Environmental Research Program, and the Australian Rivers Institute, Griffith University, for funding this study.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Albanese, B., Angermeier, P.L. & Peterson, J.T. (2009) Does mobility explain variation in colonization and population recovery among stream fishes? Freshwater Biology, 54, 14441460.
  • Arnell, N.W. (2003) Effects of IPCC SRES emission scenarios on river runoff: a global perspective. Hydrology and Earth Systems Science, 7, 619641.
  • Arthington, A.H., Balcombe, S.R., Wilson, G.A., Thoms, M.C. & Marshall, J. (2005) Spatial and temporal variation in fish-assemblage structure in isolated waterholes during the 2001 dry season of an arid-zone floodplain river, Cooper Creek, Australia. Marine and Freshwater Research, 56, 2535.
  • Atkinson, K. (1989) An Introduction to Numerical Analysis, 2nd edn. John Wiley & Sons, New York.
  • Ausseil, A.G.E., Chadderton, W.L., Gerbeaux, P., Stephens, R.T.T. & Leathwick, J.R. (2011) Applying systematic conservation planning principles to palustrine and inland saline wetlands of New Zealand. Freshwater Biology, 56, 142161.
  • Ball, I.R., Possingham, H.P. & Watts, M. (2009) Marxan and relatives: Software for spatial conservation prioritisation. Spatial Conservation Prioritisation: Quantitative Methods and Computational Tools (eds A. Moilanen, K.A. Wilson & H.P. Possingham), pp 185195. Oxford University Press, Oxford.
  • Bates, B.C., Kundzewicz, Z.W., Wu, S. & Palutikof, J.P. (2008) Climate Change and Water. Technical Paper of the Intergovernmental Panel on Climate Change. IPCC Secretariat, Geneva.
  • Bond, N.R., Lake, P.S. & Arthington, A.H. (2008) The impacts of drought on freshwater ecosystems: an Australian perspective. Hydrobiologia, 600, 316.
  • Bunn, S.E. & Arthington, A.H. (2002) Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management, 30, 492507.
  • Environmental Protection Agency (2005) Wetland Mapping and Classification Methodology – Overall Framework – A Method to Provide Baseline Mapping and Classification for Wetlands in Queensland. Queensland Government, Brisbane, Qld.
  • ESRI (2002) ArcGIS. Environmental Systems Research Institute, Redlands, CA.
  • Fausch, K.D., Torgersen, C.E., Baxter, C.V. & Li, H.W. (2002) Landscapes to riverscapes: bridging the gap between research and conservation of stream fishes. BioScience, 52, 483498.
  • Fullerton, A.H., Burnett, K.M., Steel, E.A., Flitcroft, R.L., Pess, G.R., Feist, B.E., Torgersen, C.E., Miller, D.J. & Sanderson, B.L. (2010) Hydrological connectivity for riverine fish: measurement challenges and research opportunities. Freshwater Biology, 55, 22152237.
  • Gasith, A. & Resh, V.H. (1999) Streams in Mediterranean climate regions: abiotic influences and biotic responses to predictable seasonal events. Annual Review of Ecology and Systematics, 30, 5181.
  • Haig, S.M., Melhman, D.W. & Oring, L.W. (1998) Avian movements and wetland connectivity in landscape conservation. Conservation Biology, 12, 749758.
  • Halls, A.S., Lieng, S., Ngor, P. & Tun, P. (2008) New research reveals ecological insights into Dai fishery. Catchments and Culture, 14, 812. Mekong River Commission, Vientiane.
  • Hamilton, S.K., Sippel, S.J. & Melack, J.M. (2002) Comparison of inundation patterns among major South American floodplains. Journal of Geophysical Research-Atmospheres, 107, 80388052.
  • Hermoso, V., Kennard, M.J. & Linke, S. (2012) Integrating multi-directional connectivity requirements in systematic conservation planning to prioritise fish and waterbird habitat in freshwater systems. Diversity and Distributions, 18, 448458.
  • Hermoso, V., Linke, S., Prenda, J. & Possingham, H.P. (2011) Addressing longitudinal connectivity in the systematic conservation planning of fresh waters. Freshwater Biology, 56, 5770.
  • Hughes, J.M., Schmidt, D.J. & Finn, D.S. (2009) Genes in streams: using DNA to understand the movement of freshwater fauna and their riverine habitat. BioScience, 59, 573585.
  • Kennard, M.J. (2010) Identifying high conservation value aquatic ecosystems in northern Australia. Interim Report for the Department of Environment, Water, Heritage and the Arts and the National Water Commission. Tropical Rivers and Coastal Knowledge (TRaCK) Commonwealth Environmental Research Facility, Charles Darwin University, Darwin. ISBN: 978-1-921576-23-2.
  • Kennard, M.J., Pusey, B.J., Olden, J.D., Mackay, S.J., Stein, J.L. & Marsh, N. (2010) Classification of natural flow regimes in Australia to support environmental flow management. Freshwater Biology, 55, 171193.
  • Leathwick, J.R., Rowe, D., Richardson, J., Elith, J. & Hastie, T. (2005) Using multivariate adaptive regression splines to predict the distribution of New Zealand′s freshwater diadromous fish. Freshwater Biology, 50, 20342052.
  • Linke, S., Turak, E. & Nel, J. (2011) Freshwater conservation planning: the case for systematic approaches. Freshwater Biology, 56, 620.
  • Linke, S., Pressey, R.L., Bailey, R.C. & Norris, R.H. (2007) Management options for river conservation planning: condition and conservation re-visited. Freshwater Biology, 52, 918938.
  • Magalhães, M.F., Beja, P., Schlosser, I.J. & Collares-Pereira, M.J. (2007) Effects of multi-year droughts on fish assemblages of seasonally drying Mediterranean streams. Freshwater Biology, 52, 14941510.
  • Magoulick, D.D. & Kobza, R.M. (2003) The role of refugia for fishes during drought: a review and synthesis. Freshwater Biology, 48, 11861198.
  • Meir, E., Andelman, S. & Possingham, H.P. (2004) Does conservation planning matter in a dynamic and uncertain world? Ecology Letters, 7, 615622.
  • Moilanen, A., Leathwick, J. & Elith, J. (2008) A method for spatial freshwater conservation prioritization. Freshwater Biology, 53, 577592.
  • Morrongiello, J.R., Beatty, S.J., Bennett, J.C., Crook, D.A., Ikedife, D.N.E.N., Kennard, M.J., Kerezsy, A., Lintermans, M., McNeil, D.G., Pusey, B.J. & Rayner, T. (2011) Climate change and its implications for Australia's freshwater fish. Marine and Freshwater Research., 62, 10821098.
  • Nel, J.L., Roux, D.J., Abell, R., Ashton, P.J., Cowing, R.M., Higgins, J.V., Thieme, M. & Viers, J.H. (2009) Progress and challenges in freshwater conservation planning. Aquatic Conservation: Marine and Freshwater Ecosystems, 19, 474485.
  • Nel, J.L., Reyers, B., Roux, D.J., Impson, N.D. & Cowling, R.M. (2011) Designing a conservation area network that supports the representation and persistence of freshwater biodiversity. Freshwater Biology, 56, 106124.
  • Possingham, H.P., Franklin, J., Wilson, K.A. & Regan, T.J. (2005) The roles of spatial heterogeneity and ecological processes in conservation planning. Ecosystem Function in Heterogeneous Landscapes (eds. G.M. Lovett, C.G. Jones, M.G. Turner & K.C. Weathers), pp. 389406. Springer, New York.
  • Pressey, R.L., Johnson, I.R. & Wilson, P.D. (1994) Shades of irreplaceability towards a measure of the contribution of sites to a reservation goal. Biodiversity and Conservation, 3, 242262.
  • Pressey, R.L., Cabeza, M., Watts, M.E., Cowling, R.M. & Wilson, K.A. (2007) Conservation planning in a changing world. Trends in Ecology and Evolution, 22, 853592.
  • Pringle, C.M. (2001) Hydrologic connectivity and the management of biological reserves: a global perspective. Ecological Applications, 11, 981998.
  • Rea, J.H., Brinton, A.C. & Georges, A. (2009) Temporal and spatial variation in landscape connectivity for a freshwater turtle in a temporally dynamic wetland system. Ecological Applications, 19, 12881299.
  • Taylor, P.D., Fahrig, L. & With, K.A. (2006) Landscape connectivity: a return to the basics. Connectivity Conservation (eds K.R. Crooks & M. Sanjayan), pp. 2943. Cambridge University Press, Cambridge.
  • Thieme, M., Lehner, B., Abell, R., Hamilton, S.K., Kellndorfer, J., Powell, G. & Riveros, J. (2007) Freshwater conservation planning in data-poor areas: an example from a remote Amazonian basin (Madre de Dios River, Peru and Bolivia). Biological Conservation, 135, 484501.
  • Ward, J.V. (1989) The four-dimensional nature of lotic ecosystems. Journal of the North American Benthological Society, 8, 28.
  • Ward, D., Pusey, B., Brooks, A., Olley, J., Shellberg, J., Spencer, J. & Tews, K. (2011) River landscapes and aquatic systems diversity. Aquatic Biodiversity in Northern Australia: Patterns, Threats and Future (ed. B.J. Pusey), pp 522. Charles Darwin University Press, Darwin.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
jpe2191-sup-0001-FigureS1.docxWord document161KFig. S1. Conceptual diagram of analyses.
jpe2191-sup-0002-TableS1.docWord document33KTable S1. Summary of comparison between scenarios.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.