Improving the description of human activities potentially affecting rural stream ecosystems

Abstract

Stressor (or human activity) gradients that quantify variation in the magnitude and type of human activity among sites are an important and widely applicable tool for aquatic monitoring and assessment. These gradients are typically determined from regional land cover data. We predicted that their performance could be improved by incorporating less generalized depictions of human activities. Using data from 479 rural, headwater basins we calculated four human activity gradients (HAGs) that differed in the level of detail (coarse, fine) and spatial explicitness (aspatial, spatial) used to describe human activity. Results demonstrated that the addition of fine detailed information was valuable as it resulted in a HAG that captured subtle differences in the extent of human activity among study units. In comparison, the addition of spatially explicit data added little novel information to the HAG. Analysis of fish and benthic macroinvertebrate samples from 160 of the 479 basins indicated that the addition of fine detailed and spatially explicit information significantly increased the ability of the HAG to predict variation in aquatic assemblages. We concluded that HAGs can better meet the requirements of monitoring and assessment programs if detailed and spatially explicit descriptions of human activity are used along with more typically available land cover data.