Resource information
1. The restoration of native, forested riparian habitats is a widely accepted method for improving degraded streams. Little is known, however, about how the width, extent and continuity of forested vegetation along stream networks affect stream ecosystems. 2. To increase the likelihood of achieving restoration goals, restoration practitioners require quantitative tools to guide the development of restoration strategies in different catchment settings. We present an empirically based model that establishes a relationship between a 'stress' imposed at different locations along a stream by the spatial pattern of land cover within catchments, and the response of biologically determined ecosystem characteristics to this stress. The model provides a spatially explicit, quantitative framework for predicting the effects of changes in catchment land cover composition and spatial configuration on specific characteristics of stream ecosystems. 3. We used geospatial datasets and biological data for attached algae and benthic macroinvertebrates in streams to estimate model parameters for 40 sites in 33 distinct catchments within the mid-Atlantic Piedmont region of the eastern U.S. Model parameters were estimated using a genetic optimisation algorithm. R² values for the resulting relationships between catchment land cover and biological characteristics of streams were substantially improved over R² values for spatially aggregated regression models based on whole-catchment land cover. 4. Using model parameters estimated for the mid-Atlantic Piedmont, we show how the model can be used to guide restoration planning in a case study of a small catchment. The model predicts the quantitative change in biological characteristics of the stream, such as indices of species diversity and species composition, that would occur with the implementation of a hypothetical restoration project.
