Ecohydrological regionalisation of Australia: a tool for management and science <p>This project classified Australia&rsquo;s unregulated riverine flow regimes using two approaches. The first approach used long-term greater than 15 years data from 830 gauging stations located continent-wide. A Bayesian soft clustering technique was used to group streams of similar flow regimes, resulting in the identification of 12 flow regime classes differentiated according to seasonality, predictability and extent of intermittency. The second approach was based on spatially modelled variation in flow regime derived from available GIS sources and an updated 9 second DEM for 1.2 million stream segments. This classification recognised 30 different flow regime types within 10 metagroups; the metagroups were similar to the groups derived from gauged data.</p> <!--break--> <p>This research showed that the consistent, continent-wide characterisation of climate and landscape variation differentiates many ecologically relevant components of flow regime. It provides a rigorous foundation for future ecological investigations of the importance of the flow regime, testing of scenarios associated with projected future climate change and application to management of the nation&rsquo;s water resources. </p> <p>In river systems, the flow regime is a key driver of the ecology, with features of the riverine landscape (climate, topography, geology and vegetative cover) broadly shaping patterns in the flow regimes at large spatial and temporal scales. This project explored relationships between environmental features of the riverine landscape and riverine flow regimes to develop a continental-scale regional classification of Australia&rsquo;s rivers.</p> <p>Spatial variation in flow regimes and their linkages to landscape controls are currently not well understood in Australia. With many aquatic habitats and biota threatened by hydrologic changes due to changing land-use, water resource development and extraction as well as the future threatening impacts of projected climate change, the ability to predict and therefore plan for change is vital. Environmental management of rivers as well as planning for the impacts of global climate change need to be based on predicted changes in flow regime. The ability to do so is constrained unless we know the role of the natural flow regime, how much flow regimes vary between rivers and regions and the extent to which such variation results in natural changes to riverine ecology. </p> <p>The &ldquo;ecohydrology paradigm&rdquo;, which is based on functional relationships between hydrology and biota, provides the theoretical basis for predicting the impacts of changes to flow regimes. Ecohydrological regionalisation allows rivers to be placed into a spatial context and thereby maximise the extent to which insights gained in one river or region may be meaningfully applied to another (both within Australia and elsewhere). Regions that can be subdivided into groups of stream segments and/or river catchments that are hydrologically distinctive at landscape scales can be expected to discriminate differences in ecological character. Knowledge of patterns in hydrological character should make it possible to infer spatial patterns in aquatic ecosystem structure and function. An explicit spatial context would allow researchers to develop meaningful generalizations about the interaction between hydrology and ecology in Australia, and provide the benchmark against which the response of biological communities to hydrological alteration can be assessed.</p> <p>This project classified Australia&rsquo;s unregulated riverine flow regimes to provide a rigorous foundation for future ecological investigations of the importance of the flow regime, testing of scenarios associated with projected future climate change and application to management of the nation&rsquo;s water resources. Two alternative approaches to this task were used.</p> <p>First, we used long-term data greater than 15 years derived from 830 gauging stations located across the continent to generate a comprehensive set of hydrologic metrics describing the six key facets of the flow regime: magnitude, timing, variability, predictability, duration and rates of change. Exploratory analyses aimed at determining the minimum record length required to accurately estimate hydrologic character found that a minimum of 15 years was necessary. This enabled a greater geographic cover than available from gauging stations with greater than 25 years of record. A Bayesian clustering technique (Autoclass) was used to group streams of similar flow regime. The final solution resulted in the recognition of 12 different flow regime types differentiated according to seasonality, predictability and extent of intermittency. A strong latitudinal gradient in flow regime type driven by spatial variation in climate (tropical versus temperate) was noted. A secondary longitudinal gradient was also noted principally due to the distinctiveness of flow regimes in south-western Australia. External validation analyses revealed that geographic, climatic and some topographic factors were generally strong discriminators of flow regime classes. Comparison with existing terrestrial environmental and biophysical classification schemes revealed generally poor concordance illustrating the need to consider the aquatic environment independently of terrestrial regionalisation. Variation in the extent to which the classification accounted for variation at the within-drainage or regional scale was noted due to the continental scale of the classification.</p> <p>Second, a separate classification, based on spatially modelled variation in flow regime derived from available GIS sources and an updated 9 second DEM for 1.2 million stream segments, was performed. This classification recognised 30 different flow regime types within 10 metagroups. These metagroups were similar to the groups derived from gauged data. Additional flow regime types not recognised in the previous classification were identified in the modelled approach and were found to occur in areas, principally in the arid zone, that are poorly gauged.</p> <p>We showed that the consistent, continent-wide characterisation of climate and landscape variation differentiates many of the ecologically relevant components of the flow regime. The ecohydrological environment classification will support applications at broad regional to continental scale while supplying the context for finer scale regional or catchment studies. We showed, for instance, how it might be used to explore patterns of diversity in the hydroecological characteristics of streams at a range of spatial scales. The classifications may also be applied to the task of designing a more representative network of gauging stations. <br /> <br /> <br /> &nbsp;</p> 2009-03-11T00:06:26Z 2009-07-03T01:48:22Z Dr Brad Pusey’s project, Ecohydrological regionalisation of Australia: a tool for management and science, (Australian Rivers Institute Griffith University,) involved a review of hydrological classifications for Australia that included a review of anthropogenic factors influencing stream flow in unregulated streams. Then a transparent classification process was undertaken using a Bayesian soft clustering technique. Comparisons showed the existing classifications were a poor description of flow regime variation except at the broadest scale of tropical versus temperate. Several features developed during the course of the project have been incorporated within the River Analysis Package and will be made publicly available in subsequent releases. In addition the project prepared, cleaned and infilled stream flow data for each of 830 gauges; and new hydrologic metrics (a total of 120 metrics), GIS maps of the classification and stream network. These data will into RAP and eWater CRC catchment Modelling Toolkits to assist adoption. The outcomes of the project are ready to form the framework upon which environmental flow rules will be developed within the TRaCK program. 002321