Hydrological responses to climate and land use change in upper Langat Basin, Peninsular Malaysia

It is undeniable that climate and land use changes affect the major components of hydrological cycle such as streamflow and water availability in river basins.However, the magnitude and contributions of their effects are still uncertain. It is highly required to study the individual and combined impacts of changes in climate and land use on streamflow and consequently flood vulnerability in tropical river basins particularly those basins which experience rapid changes in land use. Langat River basin which experiences flooding as a response to both climate variability and rapid urbanization and industrialization was chosen to investigate the hydrological responses of the basin to both factors. This research aimed to evaluate past and probable future environmental changes and their effects on the hydrological response of the Hulu Langat river basin. More specifically the study provides a framework to analyze the historic trends of major hydro-climatic variables as secondary data including rainfall, temperature and streamflow (1984-2010) and land cover changes (land use maps of 1984-2006-2008) and also to quantify their individual and combined impacts on streamflow through designed scenarios of different combinations of land use and climate conditions. The trends of historical time series of temperature, rainfall and streamflow were determined with the Mann-Kendall method. The study applied the semi-distributed hydrological model, SWAT, and calibrated and validated against observed streamflow using SWAT-CUP (SUFI-2). The land use/land cover changes were examined using LCM model and future land use change scenarios were developed using the combination of Markov chain and Multi-Layer Perceptron (MLP_Markov) model. The climate change scenarios were based on the combination of one RCM (PRECIS) and stochastic weather generator (LARS-WG). To this end, the daily temperature and precipitation projection of A1B emission scenario by the PRECIS model and A2, B1 emission scenarios by LARS-WG for 2020s, 2050s and 2080s were superimposed in the calibrated SWAT model. Annual streamflow in three stations (F1, F2, and F4) out of four exhibited significant increasing trend at α = 0.01 significance level.The annual streamflow had been significantly increasing at the rate of 0.036,ii 0.023 and 0.001 ×103 m3/year in those stations,respectively, from 1980 to 2010.Seasonal streamflow showed significant increasing trend in the upper part of basin in all seasons. Significant increasing trend for monthly streamflow was detected for January and August, but insignificant trends were found in May,September and November in all hydrological stations. Streamflow was primarily correlated to rainfall (P<0.05). Land use change analysis indicated that forest cover, short-term crops and permanent crops were notably decreased from 1984 to 2006. The area of permanent crop and forest area in 2006 was 48.5 % and 3.1% lower than that in 1984, respectively, but the built-up and residential areas were increased distinctly. The simulation results of the hydrological model under land use change scenarios showed that the magnitude of streamflow was impacted by land use change in all scenarios. Streamflow was highest due to expansion of urban area and contraction of forest and croplands. Future land use change would account for increase in streamflow about 20%, 61% and 71% for the 2020s, 2050s and 2080s land use scenario maps, respectively. Under only climate change scenarios, streamflow was predicted to increase 8.7% in 2020s, 54.7 % in 2050s and 21% in 2080s under A2 scenario, 14.7%, 21.4% and 75.7% for the 2020s, 2050s and 2080s under B1 scenario, respectively. Streamflow was also projected to decrease about 3.1% and 10.4% under A1B scenario for the 2020s and 2050s, but it showed an increase of about 28% in the 2080s. The intensification of streamflow is in maximum percent (92%) in 2080s under B1 scenario. The combined scenarios of land use and climate indicated that climate variables change coupled with land use change have resulted significant increase in streamflow about 9%, 20.1%, 9% for the 2020s, 2050s and 2080s under A2 scenario, and 14.9%, 24.4% and 19.7% for the 2020s,2050s and 2080s under B1 scenario. Under A1B scenario, 19% decrease in streamflow in the 2020s was predicted, but the streamflow increases 25.6% and 65.4% for the 2050s and 2080s, respectively. This study concludes that under different designated scenarios, streamflow is expected to experience significant increase in the future decades as a consequence of land use and climate changes, but changes in land use have led to greater streamflow changes against climate change. Consequently, there is an urgent need to reverse land use change trends such as afforestation, conservation on natural habitat and control on urban expansion for future development programs and land use planning to mitigate probability of possible flooding which is a major problem in this region. Keywords: climate change, hydrological processes, hydroclimatic variables,land use conversion, trend analysis, streamflow

Preview Text FH 2016 26 IR.pdf Download (2MB) | Preview

Actions (login required)

View Item