Modeling and assessing the effects of land use changes on runoff generation with the CLUE-s and WetSpa models

Abstract

Land use change is an important determinant of hydrological processes and is known to affect hydrological parameters such as runoff volume, flood frequency, base flow, and the partitioning into surface flow and subsurface flow. The main objective of this research was to assess the magnitude of the effect of land use changes on runoff parameters, using the Baghsalian watershed in Iran as a case study site. At first, land use maps of years 1986 and 2012 were prepared using synthetic method, and then simulation was done based on land use changes in the 1986 to 2012 period. Land use map of year 2030 was simulated using CLUE-s model. Spatially distributed hydrological WetSpa model was used to simulate runoff at daily scale with land use maps of 1986, 2012, and 2030. Total volume of runoff, peak flow, and surface flow were compared. The accuracy of the WetSpa model simulation was assessed with the Nash-Sutcliffe efficiency, which had values of 0.61 and 0.56% for the calibration and validation dataset, respectively. The aggregation measure criterion was also calculated and had values of 64 and 62% for the calibration and validation periods, respectively. The main land use changes in Baghsalian watershed between 1986, 2012, and 2030 were the conversion of forest and rangeland to agriculture and residential land use types. Because of these conversions, simulated total runoff volume increased; and the rate of increase in surface runoff was larger than the rate of increase in subsurface runoff. In addition, surface and subsurface runoff increased in 2012 and 2030 compared to 1986 land use map, but the rate of increase of subsurface runoff was less than surface runoff.