Monitoring water volume variations in the Nam Ngum hydropower dam using satellite observations combined with HydroWEB and G-REALM water level data

Tóm tắt

Continuous and frequent monitoring of surface water variations in hydropower reservoirs is essential for effective energy production, flood mitigation, and water management across agricultural, industrial, and domestic sectors. This study focuses on monitoring water volume variations in the Nam Ngum hydropower reservoir, the largest water body in the Lao People's Democratic Republic (Lao PDR), from August 2022 to August 2024. The reservoir’s surface water extent was estimated using imagery acquired from optical sensors onboard Sentinel-2, Landsat-8 and Landsat-9 satellites, as well as radar sensors onboard Sentinel-1 satellite for cloud-covered conditions. The reservoir’s water level was derived from satellite altimetry data provided by the HydroWEB and G-REALM databases. In situ measurements of water level and volume, provided free of charge by the reservoir operating company, were used for validation. Results indicated that water levels derived from satellite altimetry data ranged from 200÷212 m, exhibiting an extremely high correlation with in situ measurements (R = 99.94% and RMSE = 0.1418 m). The reservoir’s water extent varied between 350 and 485 km2, with a strong correlation to water level records (R = 98.48%). Finally, the estimated water volume variations of the reservoir closely followed the in situ observations (R = 99.63%). These findings demonstrate the effectiveness of satellite data in monitoring variations in water levels at hydropower dams. Such information is crucial for water resource management, particularly in downstream regions during storm seasons. However, this study has some key limitations: (1) the reliance on optical imagery restricts its applicability during cloudy periods, which are common in tropical regions; and (2) the HydroWEB and G-REALM only provide data for a limited number of lakes with surface areas larger than 100 km2