9306909 Fisher This research will determine the effect of spatial configuration on nutrient retention in stream-riparian ecosystems. Streams are complex, heterogeneous systems consisting of a wetted channel, saturated sediments below and adjacent to the wetted channel, and a strip of terrestrial vegetation (the riparian zone) at each lateral edge. The extent and shape of each of these varies greatly in space and time. As water flows downstream, it moves to varying degrees among these components. Because of several favorable properties, streams of arid lands will be used as model systems for this study. The overall goal of this project is to determine how configuration affects the extent to which materials transported in this water are retained and recycled within a given reach. This goal will be met by addressing four questions around which the research is organized. First, how does spatial pattern influence nutrient retention? Spatial configuration and its effects will initially be considered for a single point in time. This will be accomplished by mapping the stream ecosystem a two scales to determine its physical configuration, describing hydrologic connections among these elements, and the consequences of flow path for retention of nutrients. Configuration will be manipulated experimentally to determine the effect of different patterns on the retention process. Second, how does disturbance affect this relationship? Several individual flash flood and drought events will be studied in terms of their effect on configuration and recovery of system function between events. Third, how do configuration and function interact over large spatial and temporal scales? This analysis will involve scaling up reach-scale processes to include geomorphic variation generated by valley floor width. Temporal scaling will be accomplished by considering the effect of variable flood and drought regimes; their frequency , intensity, and distribution in time. Fourth, how general are results derived from North American desert streams for this ecosystem type worldwide? This question will be answered with a comparative study of the pattern-process relationship among streams of similar deserts in Australia and Spain. %%% This research is significant in that it will contribute to basic scientific understanding of how pattern and process interact in complex ecosystems. It is an advance over black box, well-mixed reactor models that have been commonly used in ecosystem ecology. Research results will be useful to applied ecology as well. Streams are open systems which link terrestrial uplands with downstream reservoirs, groundwater supplies, or estuaries. Retention and transformation of stream load in transit affects the quality of water delivered to recipient systems. Riparian and other wetland ecosystems are increasingly used for this "nutrient filtration" capacity. Results of this research will aid in understanding this process in natural streams and optimizing it in managed and human-designed systems. ***