Detecting human interferences to low flows through base flow recession analysis

Human activities directly and indirectly interfere with naturally occurring hydrologic processes. Indirect human interferences, such as those resulting from land use and land cover changes, have been studied and modeled in many watersheds throughout the world. These kinds of human interferences usually have considerable impacts on the rising and falling limbs and the peak of a hydrograph. Direct human interferences, such as water withdrawals and return flows, can affect the base flow recession process, which is not greatly affected by the land surface condition. Direct human interferences are usually assumed to be known but are difficult to observe and model. The withdrawals and return flows are frequently added to or subtracted from estimates of naturally occurring streamflow, underlying an assumed linear relationship between streamflow and water withdrawal and return flow. This paper presents an empirical analysis to incorporate groundwater pumping and return flow as variables in the recession process. This technique is subsequently applied to the Salt Creek Watershed, a highly urban watershed located in the Chicago area. On the basis of the historical streamflow observation during 1946-2006, the analysis provides results that explain the impact of the aforementioned direct human activities on the base flow recession process. Moreover, this empirical method allows estimates of human water uses, such as groundwater pumping and return flow, to be made as long as a sufficiently long term streamflow data record is available. This analysis is verified with the data from the Salt Creek Watershed and can also be applied to other watersheds subject to intensive human interferences.