Modeling the effect of land use and land cover changes on long-term rainfall/run-off and non-point source pollution in the upper Kelantan River Basin, Malaysia

Kelantan River basin, Malaysia is a tropical catchment receiving heavy monsoon rainfall coupled with intense land use and land cover (LULC) changes making the basin consistently flood prone. A study was conducted to model the effect of LULC changes on long-term rainfall-runoff and non-point (NPS) pollution in the upper portion of the basin. First, LULC maps corresponding to 1984, 2002 and 2013 were analyzed. The basin was delineated into four catchments (Galas, Pergau, Nenggiri and Lebir) due to it’s and size for improved results accuracy. Flood hydrographs corresponding to 1984, 2002 and 2013 LULC condition were simulated using HECHMS. Relative changes in the peak flow of the three subsequent conditions were determined for different return periods (2, 5, 10, 20, 50 and 100 years). By using flood response approach, flood source areas were identified and ranked based on the values given by gross flood index (F), per unit area index (f) and flood area index (fa) where different results were obtained for each index. Long-term runoff dynamics due to LULC changes was determined using NRCS-CN model and its modifications. NPS pollution estimation was carried out using numeric integration in a GIS environment. Soil loss was estimated using RUSLE model. Result of land use analyses showed that deforestation for logging activities, agricultural purposes and urbanization were the major land use changes observed in the basin from 1984-2013. Lebir (48557.3 m3/s) was the catchment with greatest contribution of peak discharge at the outlet under 2013 LULC condition. This is followed by Galas (43357.7 m3/s), Pergau (33126.4 m3/s) and Nenggiri (16729.1 m3/s) in that order. The use of fa index gives better ranking and is therefore, recommended in ranking sub-basins with respect to their contribution to the outlet. Results of runoff dynamic reveal that proposed modified NRCS-CN model V (MNM V) was found to give the best runoff estimation based on model goodness of fit evaluation criteria. Thus, the MNM V was selected for runoff estimation from 1984- 2014. It was observed from the results that runoff estimation increased with changes in LULC from 1984-2014 in all the selected runoff events and in all catchments. Results of spatio-temporal variation of pollutant loads in all the catchments increased with changes in LULC condition as one moves from 1984-2014, with 2013 LULC condition found as the dominant in almost all cases. NPS pollutant loads among different LULC changes also increased with changes in LULC condition from 1984- 2013; while urbanization was found to be the dominant LULC change with the highest pollutant load in all the catchments. This reveals the clear effect LULC changes on NPS pollution. Soil erosion results from RUSLE showed that 67.54% of soil loss is located under low erosion potential or 0-1ton ha-1 yr-1 otherwise known as reversible soil loss in Galas, 59.17% in Pergau, 53.32% in Lebir and 56.76% in Nenggiri all under the 2013 LULC condition. These results are higher than that of 1984 and 2002 LULC conditions.The novel methodologies developed in this study can be incorporated for regional hydrological studies and early warning systems for flood control.

Preview Text FPAS 2018 20 - IR.pdf Download (3MB) | Preview

Actions (login required)

View Item