Water supply reservoir operation in the framework of climate variability and change

Abstract

The optimal planning and operation of a reservoir system is getting more crucial particularly in view of the recent awareness of potential climate change. In particular, the incorporation of hydrologic uncertainties due to climate change into reservoir operation system requires comprehensive and long-term hydrological database which rarely available in most of the conventional reservoir design. The prime objective of the study is to formulate a multiple approach on the long-term reservoir operation optimization under the scarcity of observed hydrological data and with the influence of climate change. A combined research method using IHACRES for hydrological simulation, HadCM3 for emission scenario and Statistical Downscaling Model were developed along with a Mixed Integer Linear Programming (MILP) for reservoir operation optimization. These approaches were applied to a single purpose Sg Layang Resevoir, that is one of the most prominent water supply reservoir located in Johor State, Malaysia. The climatic variables obtained from general circulation model (GCM) were downscaled corresponding to HadCM3 emission scenario and used in climate change impact analysis. The SDSM was used to produce 100 synthetic climate time-series for 90 years of the participating station, representing the climate change projection and baseline period. With respect to the baseline data, an apparent increase in temperature (1.2 degree Celsius between time periods) and rainfall was observed. The deterministic optimization exercise is performed repetitively for a number of case scenarios based on weekly reservoir’s inflows derived from the projected climate change in a way to determine the optimal operation rule and policy which are based on total pumping volume and pumping cost. Corresponded to the future inflows, the pumping volume has shown an increase trend particularly during southwest monsoon, transition between seasons and autumn. Judged from the decreasing rate of the streamflows, a 34 to 40% increase in the projected monthly pumping volume is anticipated. An opposite scenario is observed during northeast monsoon season which shows a decreasing trend of 28% to 46%. At various degree of statistical reliability, the optimal operational pumping curves of the reservoir were established. These curves provide some basic information on the monthly pumping requirement from various sources of inflow to sustain the reservoir storage and demand. These operation curves are of very useful guidelines for reservoir operators in making decision to follow an optimal pumping operations schedule onsite. Such research findings were expected to generate a general awareness to the public water authorities on the potential long term effect of climate change to the reliability of reservoir operating system.