Identifying spatial clustering in change points of streamflow across the contiguous U.S. between 1945 and 2009

摘要: Much of the work investigating sudden changes in streamflow U.S. has used only a small subset all available gage data and identified single change point each gage's period record. In this paper, we apply detection clustering algorithm that uses Geological Survey flow gages with near-continuous records, detects multiple points annual streamflow, groups into geographic clusters which are not predefined by any political or hydrologic boundaries. We identify 17 spatially distinct clusters, 13 related to concurrent precipitation. Several regions display indicating time. The presence abrupt suggests natural variability climate signal may be dominating observed variations last 60 years many locations contiguous