Integration of LULC change/prediction and hydrological modeler for assessment of the effect of LULC Change on peak discharge in Sari Watershed, Sumbawa Island, Indonesia

Abstract

Flood is the main hazard that frequently occurs in Sari Watershed, Sumbawa Island, West Nusa Tenggara Province, Indonesia. The Land Use Land Cover (LULC) change, especially from the forest into non-forest coverage, may have contributed to the increased surface runoff and flooding frequency downstream. Because of the contradicting effects that these processes have on the risk of flooding, it is difficult to predict what the long-term impact of LULC changes will be on flooding in Sari Watershed. This study evaluated the effect of LULC change and prediction on peak discharge in the Sari watershed from 1990 to 2030. LULC maps of 1990, 1995, 2000, 2005, 2010, 2015, and 2020 were obtained by Landsat images classification, whereas 2025 and 2030 were from prediction. LULC change and prediction analysis used the ANN (Artificial Neural network)-CA (Cellular Automata) Markov model. Peak discharge for each year was calculated using the HEC-HMS model. The relationship between LULC change and peak discharge was analyzed with correlation analysis. The result showed that the trend of peak discharge dynamic was consistent with the tendencies of LULC change, particularly from forest coverage to non-forest coverage. Based on correlation analysis, the peak discharge dynamic correlated with the shift in LULC during the analysis period. The study result may provide a reference for flood risk reduction and LULC planning in Sari Watershed.