The Effect Of Land Cover Data On Hydrological Simulation

With the deep research on global climate change, the effects of hydrological model simulation by Land Cover Change become a new study direction.The Qinhuai Watershed, located in the lower reaches of Yangtze River, containing the majority of Nanjing City and part of Zhenjiang City, is one of the fastest economic development regions, which absolutely accompany with Land Cover change significantly. Therefore, environment crisis become more and more frequent, like the shortage of water resource, flood and deterioration of water quality. At the same time, many researchers focus on the study of Global Climate Model to discover the variation tendency of our planet. As one of the momentous data in the simulate process, Land Cover data demand more accurate standard, such as high-resolution and multi-classification land cover data, to generate more accurate simulate results.In this research, by input differ-period, differ-resolution and differ-classification land cover data, SWAT model (a distributed hydrological model) was applied to study the impact of land cover data on hydrological simulation process in Qinhuai Watershed. The main conclusion as follows:1. Established the database of Qinhuai Watershed by collecting the soil, hydrological, meteorological data of this area. Based on the calibration and validation, the simulate runoff match well with the measured flow.2. Comparative analysis with land cover data of1988and2007, with the rapid economic development in Nanjing, the area of urban has expanded obviously; meanwhile the dry land and paddy field decreased greatly. Then, two typical sub-basins were chosen to identify the impact of land cover change. The simulation results show that this change leads to the increase of surface runoff.3. By inputting30meter resolution and300meter resolution land cover data, there are little difference between the simulation results. The disparity of simulation results in flood period is larger than the others and the accuracy of300meters shows little less than the30meters.4. The simulate result of12categories of land cover data is larger than the5categories', with the range of1.3%to4.11%.And the simulation flow of12categories of land cover data is smaller than the5categories'in flood period.