Author(s): Leonore Boelee, Razak Bahrom, Haslina Amir, Nor Zamri Sondor, Emma Brown, Faizah Ahmad, Geoff Wilson
Linked Author(s): Leonore Boelee
Keywords: Water resource management, operational systems, NWP, forecasting
Abstract: The Selangor River lies on the west coast of Peninsular Malaysia, 70 km north of Kuala Lumpur. The Selangor catchment area is approximately 2000 km2 and has two major dams located in the upstream part of the catchment: Sungai Selangor Dam and Sungai Tinggi Dam. Lembaga Urus Air Selangor (LUAS) is the Malaysian government agency responsible for water resources management of the Selangor catchment, with their main challenge being to balance an increase in water demand, whilst trying to manage the catchment's environmental state. The key to this is the amount of water that is released from the two reservoirs. The current manual decision-making process (based on staff experience) does not provide LUAS with the efficiency and accuracy that is required in a catchment where sustainability is key and water resources are not always abundant. A solution was required to improve the management of the current resources, thus reducing the need for the costly and environmentally contentious development of new infrastructure. The operational Decision Support System (DSS) for sustainable water resources management of the Sungai Selangor catchment is a non-structural tool developed to support LUAS in optimising the reservoir releases and water abstractions in the catchment. The DSS is a fully automated system that is driven by a combination of live, telemetered gauged data from various sources including the InfoBanjir telemetry database and Numerical Weather Prediction (NWP) rainfall forecasts from the Malaysian Meteorological Department (MMD). Simulations are carried out automatically on an hourly basis, to calculate and forecast whether water levels at the main abstraction location (SSP1) are being maintained at the required levels. If the required water levels at SSP1 are not met, being either too high or too low, the model is re-run using an adjusted reservoir release time series. The DSS should also be able to help LUAS in the future to increase the sustainability of management of water resources in the Selangor catchment. This paper describes the approach developed to support LUAS to meet this objective, in a manner that can be used in real-time, and can be transferred to other catchments in the future. The advantages and disadvantages of potentially expanding this system to use ensemble forecasts, data assimilation and optimisation algorithms are discussed, along with suggestions for further research
Year: 2017