ABSTRACT

Leroy, Jeroen D. M.S., Purdue University, May, 2004. Modeling Lake Level Variations Using L-THIA in the Lake Maxinkuckee Watershed. Major Professor: Bernie Engel.

Land-use changes driven by human activity have profound environmental implications, therefore a long-term understanding of changes in land and water use is central to the debate of sustainability. The purpose of this study was to examine temporal variations in the rainfall-runoff relationship of the Maxinkuckee Watershed, in northern Indiana, with varying land uses over approximately a 50 year period. In order to do so, the daily time step Long-Term Hydrologic Impact Analysis (L-THIA) model and GIS based L-THIA model were used. These models estimated daily and average annual runoff from the Maxinkuckee Watershed, using land use maps created for 1951, 1972 and 1997, daily precipitation data and hydrologic soil group information. L-THIA estimated daily runoff values were integrated into an EXCEL spreadsheet program to model lake level incorporating all direct inputs, such as precipitation and runoff, and also all outputs, such as evaporation, ground water interaction and flow out of the lake through a weir and culverts. This was then used as a tool to quantify the impact of land use change on the total runoff from the surrounding land. Calibration of the model was done using the first 10 years of the study period using the Nash Sutcliffe Efficiency (NSE) statistic to determine performance. Due to an obvious underestimation in runoff created by the daily L-THIA model, scenarios with increased curve number (CN) values in increments of 10% were run until the best correspondence between estimated and observed lake levels was obtained. Validation was then done using the remaining study period, and a statistical analysis was performed in order to establish the accuracy of the rainfall-runoff model. A 40% increase in CN values resulted in the highest NSE value during both the calibration and validation phase with NSE values of 0.56 and 0.39, respectively. In order examine future impacts of land use change, two scenarios using increased urbanization, by two and four times, were then applied. Finally, the average annual runoff values were used to determine differences in estimated NPS pollutants associated with the varying land use.