Impact Assessment of Land Use Change and Rainfall Variation on Peak Flow in Upper Nan Basin

Abstract

The Upper Nan river basin has relatively high rainfall in mountainous areas, and land-use change has been observed, resulting in frequent flooding. This study investigates the impacts of land-use change, and rainfall variation on peak flows by using the Rainfall-Runoff-Inundation (RRI) model. The inputs are daily rainfall data from four stations from the Thai Meteorological Department (TMD) during 2001-2017, land-use data from Land Development Department (LDD) during 2003, 2009, and 2012 and land-use data from Moderate Resolution Imaging Spectroradiometer (MODIS, NASA) during 2001-2017, and digital elevation model during 2001. From the analysis of annual rainfall, the average annual rainfall in the Upper Nan river basin is relatively high at 1710 mm/year compared to the average annual rainfall of the Nan river basin at 1287 mm/year. From the moving average, the annual rainfall of the four stations demonstrates a cyclic pattern. The Mann-Kendall Test of the annual and seasonal rainfall of stations shows that the null hypothesis of rainfall data has no trend cannot be rejected at the confidence level of 95%. Land-use data from LDD are not continuous. In this study, land-use data from MODIS are used and calibrated with LDD data before inputting into the RRI model. The results show that land-use change in the Upper Nan river basin is relatively small, and MODIS cannot detect urban areas effectively. Furthermore, the woody savanna type in MODIS land-use can be interpreted as agriculture or forest, so further study needs to be at the sub-basin scale. Results from the impact study show that when rainfall and land use increased by 1%, the peak flow would be increased by rainfall variation more than land-use change. The variation from rainfall has more impact on peak flow than a land-use change under the scope and limitations of this study that there is a relatively small change in land use.