Towards rapid, rigorous and reproducible ecohydrologic classification of data scarce regions

Background/Question/Methods

An objective account of the spatial variation in ecohydrologic processes remains urgently needed to support freshwater management in many data-scarce regions. Classification systems can meet this need, but measured flow records are insufficient to support assessments in most areas. Synthesizing public data and open-source software with approaches established in data-rich settings, we developed a method to classify national or regional extents according to differences in key environmental controls of water quantity and quality. This hierarchical approach indicates the number of classes best supported by the data, while facilitating the consideration of classification systems with more or fewer groups.

Results/Conclusions

Application to Colorado (USA), generated divisions that corresponded to flow regime classes previously defined from long-term discharge records in minimally altered catchments. As a proof of concept, we therefore applied the approach in Ecuador, where streamflow records are very limited. The resulting classification agreed with major divisions in an existing map of global freshwater ecoregions, while further discriminating areas of potentially distinct ecohydrologic function at the sub-national scale. If the results are treated as a rigorously generated hypothesis, then this protocol provides a transferrable approach to catchment classification in data-scarce regions, and could ground policy processes such as the definition of environmental flow standards.