Ecological responses to artificial drought in two Australian rivers with differing water extraction

Abstract

Human demands for water reduce total discharge and extend the natural durations of low flows in streams and rivers worldwide. Although these 'artificial droughts' are predicted to increase, their ecological effects are poorly understood, hampering management of the timing and volume of extractions to minimise damage to stream ecosystems. This study hypothesised that artificial drought would reduce wetted area, increase electrical conductivity and nutrient concentrations, and promote algal growth in a heavily extracted river when compared to a river relatively unaffected by water extraction. These responses were predicted to affect aquatic macroinvertebrate community composition, producing declines in taxon richness and density and increases in representation by drought-tolerant groups such as odonates and semi-aquatic coleopterans and hemipterans. At five sites, samples were collected weekly to monthly from riffles before, during and after greatest water extractions until flows increased again. As discharge declined, wetted area and velocity fell while conductivity increased. Changes in physical habitat were most marked at sites on the impacted river but temporal variations in wetted area strongly correlated with site-specific riffle morphology, mediating local effects on periphyton biomass and benthic invertebrates. Periphyton biomass increased as flow declined except at the site with the most severe artificial drought. This latter unexpected response apparently reflected nutrient limitation, likely due to restricted supply as runoff and nutrient transfer fell. The richness and density of aquatic macroinvertebrates did not decline with discharge, but community composition changed as taxa tolerant of low flow became more abundant, with some sensitive taxa not found in the impacted river. These changes were more strongly associated with the number of low-flow events over the long term (365 d) than shorter terms (< 90 d), suggesting cumulative effects of artificial droughts on aquatic invertebrate composition in the impacted river.