Table of Contents
|
June to August 2024 Edition
Tracking Hydropower Dam Operations Using Satellites A Case of the 2018 Kerala Floods By Sarath Suresh – PhD scholar at the University of Washington In parts of the world characterized by high precipitation and steep topography, hydroelectric dams often play the dual role of power generation and flood control. Improper and uncoordinated management of such dams during extreme and unexpected precipitation events can have disastrous consequences. Numerous cases of flooding events that have been exacerbated due to insufficient storage conditions in hydropower dams have been reported worldwide. Issues also arise when critical data regarding reservoir operations are not shared between trans-boundary agencies thereby making flood preparedness extremely challenging. As such, a need for an open and transparent reservoir monitoring framework that can provide publicly-accessible reservoir operations data has become apparent. Recent studies by SASWE Labs at the University of Washington have explored the use of a purely satellite data and hydrological model-based reservoir monitoring framework to this effect. The Reservoir Assessment Tool (RAT 3.0), which utilizes high frequency remote sensing-based surface area and reservoir storage estimation alongside hydrological modelled inflow was developed and tested for various basins across the world. Here, we look at its implementation and efficacy in tracking dam operations during a disastrous flooding event that took place in 2018 in the southern Indian state of Kerala. Kerala is a coastal and mountainous state in the southwestern region of India, bounded by the Arabian sea to its west and the Western Ghats Mountain ranges to the east. It has a tropical climate with high temperatures, humidity, and substantial rainfall (~2900mm annual average). During the second and third week of August 2018, hydropower reservoir operators in the state were caught unawares by unexpected and extremely high precipitation triggered by a northward propagating cyclonic disturbance in the Arabian Sea. 35 of the 61 dams in the state were forced to be opened in a never-before-seen scenario, causing widespread flooding, leading to the loss of over 400 lives and an estimated $5 billion worth of property damage.
Figure 1. (i) and (ii): Widespread inundation due to 2018 Kerala Floods, (iii) Opening of all 5 shutters of the Idukki Reservoir Although the root cause for the flood was attributed to the extreme rainfall, it was determined by studies that the various reservoirs across the state were already at or near full reservoir level and thus had lost any flood moderation capabilities. Based on these studies, a revision of the rule curves of the dams were recommended by the Central Water Commission of India. Read more about the 2018 Kerala floods here: 2018 Kerala floods - Wikipedia Application of RAT 3.0 was carried out across 19 reservoirs in the state. Data from various satellite sources such as the Landsat series and SRTM mission from NASA, and the Sentinel Constellation from the European Space Agency is integrated in RAT to provide high frequency estimates of reservoir area and storage levels. This is combined with hydrological model-based inflow to generate reservoir outflow estimates. The effectiveness of RAT 3.0 was gauged by assessing how well the fully satellite sensor-based framework captured the rapidly evolving dynamics of the flood and reservoir state. Application of RAT 3.0 in monitoring the state of the reservoirs in Kerala during the flood event showed promising results. In general, RAT 3.0 was found to capture the temporal trend of reservoir storage and pinpoint the sudden shift in filling or release decisions made by the dam operator with good accuracy. This translated to reliable updating of downstream flood risk in near real-time for improving flood preparedness, even though the absolute magnitudes were sometimes found to need bias correction. Figure 2 depicts surface area time series for a prominent hydropower dam in the state as obtained using RAT. Flood cushioning, defined as the intentional lowering of storage to accommodate for expected future precipitation can be clearly observed here. In the case of the reservoirs in Kerala, the timing of the flood cushioning is based on rigid rule curves developed more than 50 years ago.
Figure 2. Surface Area time series for Idamalayaar reservoir obtained from RAT3.0 RAT further showed that by August 8th, about ten days before the peak precipitation event, 16 out of the 19 reservoirs had reached 90% of the maximum surface area. This indicates that flood cushioning was provided too early or was inadequate and that the monsoon precipitation had filled up the flood pool, effectively nullifying any flood moderating capabilities of the reservoirs. Finally, RAT was also able to pick up period of high inflow and outflow as shown below. Maximum inflow and outflow was found to occur around August 17th, which matches reports from state and national agencies such as the Central Water Commission, India.
Figure 3. Inflow and Outflow heat plots obtained from RAT. The study concluded that such a framework that uses only satellite-based data and hydrological modelling can be a powerful tool for water managers to monitor and track reservoir operations. Although they are not a direct substitute for in-situ measurements, the possibilities and reach of satellite-based observations in reservoir management are exciting. The RAT tool was subsequently adopted by the Kerala Center for Water Resources Development and Management (CWRDM) and runs operationally on their local servers. Data generated from RAT is made accessible to the public via a web-based interface. The tool is expected to aid the agency in providing early warning during potential future flood events and in policy making regarding reservoir operation. Read more about this study here: https://doi.org/10.1016/j.rse.2024.114149 The CWRDM front end tool can be accessed here: https://cwrdm.kerala.gov.in/reservoir-assessment-tool-kerala-rat-30 Bio: Sarath Suresh is a second year PhD student at the University of Washington, working with Professor Faisal Hossain at SASWE Labs in the department of Civil and Environmental Engineering. Sarath is from the southern state of Kerala, India and holds a Master’s degree in Ocean Engineering from the Indian Institute of Technology Bombay. His research focuses on Satellite remote sensing and Hydrology, with a special emphasis on hydropower dam operations and flood risk analysis. Email: saraths96@gmail.com, LinkedIn: sarathsuresh96 |
