Climate change and its impacts on hydrological regime of the Langat River Basin

• Main Author: Amirabadizadeh, Mahdi
• Format: Thesis
• Language: English
• Published: 2015
• Online Access: http://psasir.upm.edu.my/id/eprint/58115/1/FK%202015%2084IR.pdf; http://psasir.upm.edu.my/id/eprint/58115/

The water resources in the Langat River Basin are the main sources of water supply for different usage in the Klang Valley area that includes the city of Kuala Lumpur. In this study the rainfall data and the maximum, minimum and mean temperatures data were investigated for the presence of annual and seasonal trends. The Mann-Kendall test and the Theil-Sen's Slope method were used to detect the existence and magnitude of changes in the significant trends. The analytical results indicated that there were significant increasing trends in the annual and seasonal precipitation as well as the maximum and minimum temperatures at the 95% confidence level. This study also investigated the ability of the multiple linear (Statistical Downscaling Model) and nonlinear regression (Artificial Neural Network) methods with different complexity in downscaling and projection of climate variables in the Langat River Basin. These statistical downscaling models have been calibrated and validated using the NCEP/NCAR predictors in single station approach. The statistical validation of the generated precipitation, maximum and minimum temperatures on a daily scale,illustrated that the SDSM performs with better accuracy than the ANN model. The SDSM showed much ability to catch the wet spell and dry spell length than the ANN model. The calibrated models show more accuracy in simulating the temperature when compared with the capture of the variability of the precipitation. The better performing SDSM model was applied in projecting regional variables for two future periods (2030s and 2080s) by using predictors of the Coupled Global Climate Model version 3.1 under the A2 emissions scenario. The SDSM predicts an increase in mean monthly precipitationfor two future periods. This downscaling model predicts a similar pattern for maximum and minimum temperatures during future periods. The GEV distribution was fitted to the observed and generated daily rainfall, maximum and minimum temperatures in two future periods (2030s and 2080s) as well as baseline period using the Maximum Likelihood Method (MLE) at different stations. The comparison between the return values for precipitation and maximum and minimum temperatures indicated that the precipitation increases more than the temperature at all stations under future scenarios. Results of sensitivity analysis during the calibration process indicated that the mean monthly streamflow was sensitive to changes in seven parameters (v_ALPHA_BNK, v_CH_K2, r_SOL_K(…), r_CN2, v_EPCO, v_GW_REVAP, r_REVAPMN) out of 19 parameters. Four evaluation index values namely, NSE, PBIAS, RSR, and R2 of 0.62, 5.7, 0.61, and 0.63, respectively indicated that the calibration was reasonable. These indexes during the validation period were 0.55, 3.5, 0.67, and 0.56 respectively. The SWAT modelwas applied to predict the values of the mean monthly discharges in the Hulu Langat basin for the three periods which are the baseline, 2030s, and 2080s and these values are 14.15, 24.20, and 29.42 m3/s, respectively.The majorcontribution of this study was to identify the SDSM model as the more reliable downscaling model for the study area, which can be developed further by using of more General Circulation Model (GCM) outputs.