A Spatially Explicit Framework for Connecting Land Cover Changes to Stream Ecosystems

Process hierarchies within watersheds exert differential control on stream ecosystems relative to the composition, configuration, and relative connectivity of contributing land covers. Still, the scales at which watershed processes govern discrete biophysical responses across continua of space and time are largely unknown.  This is partly due to differences in the precision of digital spatial datasets when compared with response variables that are typically measured in watershed studies. Many digital land cover datasets are derived through various automated classification approaches that produce relatively large patches of land cover types. Datasets of this resolution are suitable for understanding broad-scale, long term phenomena like hydrology and stream geomorphology, but they may be inappropriate for finer-scale investigations into drivers of biological response or water chemistry.  Small stream studies, by contrast, are often undertaken with highly precise instrumentation and field measurement protocols.  These differences in precision among predictor and response variables can produce highly variable results. One way to reduce spatial sources of error is by using high-resolution data that are finer than the levels of organization under investigation. To understand the scales of select watershed processes, and the ways that land cover changes affect them, we hand digitized high-resolution aerial photographs, and analyzed bi-annual (2005 – 2010) land cover changes relative to their hydrologic distance from a set of study sites in nine watersheds in the Puget Sound lowland. Land cover changes were analyzed as discrete transitions from one type to another and in two broad categories: recovering or degrading. Additionally, we investigated the distributions of these transitions at the entire watershed, and buffershed extents to test if distributions of land covers varied by extent. These transition values were then compared against measures of habitat complexity, water quality, macroinvertebrate communities, and discharge characteristics to test their relative degree of relatedness with land cover changes.