Prediction of Hydrologic Characteristics for Ungauged Catchments to Support Hydroecological Modeling. Issue 11 (8th November 2017)

• Record Type: Journal Article
• Authors: Bond, Nick R.; Kennard, Mark J.
• Is Part Of: Water resources research. Volume 53:Issue 11(2017)
• Page Start: 8781
• Page End: 8794
• Publication Date: 2017-11-08
• Subjects: ecology -- flow -- hydrology -- metrics -- prediction -- statistical model; Hydrology -- Periodicals; 333.91
• DOI: 10.1002/2017WR021119
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Hydrologic variability is a fundamental driver of ecological processes and species distribution patterns within river systems, yet the paucity of gauges in many catchments means that streamflow data are often unavailable for ecological survey sites. Filling this data gap is an important challenge in hydroecological research. To address this gap, we first test the ability to spatially extrapolate hydrologic metrics calculated from gauged streamflow data to ungauged sites as a function of stream distance and catchment area. Second, we examine the ability of statistical models to predict flow regime metrics based on climate and catchment physiographic variables. Our assessment focused on Australia's largest catchment, the Murray‐Darling Basin (MDB). We found that hydrologic metrics were predictable only between sites within ∼25 km of one another. Beyond this, correlations between sites declined quickly. We found less than 40% of fish survey sites from a recent basin‐wide monitoring program (n = 777 sites) to fall within this 25 km range, thereby greatly limiting the ability to utilize gauge data for direct spatial transposition of hydrologic metrics to biological survey sites. In contrast, statistical model‐based transposition proved effective in predicting ecologically relevant aspects of the flow regime (including metrics describing central tendency, high‐ and low‐flows intermittency, seasonality, and variability) across the entire gauge network (median R 2 ∼ 0.54, range 0.39–0.94). Modeled hydrologic metrics thus offer a useful alternative to empirical data when examining biological survey data from ungauged sites. More widespread use of these statistical tools and modeled metrics could expand our understanding of flow‐ecology relationships. Key Points: Hydrologic information is essential to understand flow‐ecology relationships in rivers but is often unavailable for ecological survey sites Gauged flow metrics can only be reliably transferred to ungauged sites close by (up to ∼25 km) and with similar catchment areas (∼±10%) Statistical models can reliably predict a range of flow metrics for ungauged sites using climate and upstream catchment characteristics