loading page

Monitoring the storage volume of water reservoirs using Google Earth Engine
  • Joaquim Vasques Condeça,
  • João Palma Nascimento,
  • Nuno Barreiras
Joaquim Vasques Condeça
Universidade de Lisboa

Corresponding Author:vasques.condeca@gmail.com

Author Profile
João Palma Nascimento
Universidade de Lisboa
Author Profile
Nuno Barreiras
Instituto Superior Técnico, Universidade de Lisboa
Author Profile


Recently, the satellite images have been used in remote sensing allowing observations with high temporal and spatial distribution. The use of water indices has proved to be an effective methodology in the monitoring of surface water resources. However, precise or automatic methodologies using satellite imagery to determine reservoir volumes are lacking. To fulfil that gap, this methodology proposes 3 stages: use Google Earth Engine (GEE) to select images; automatically calculate flooded surface areas applying water indices; determine the volume stored in reservoirs over those years based on the relation between the flooded area and the stored volume. The method was applied in four reservoirs and contemplate Landsat 4 and 5 ETM and Landsat 8 OLI. For the calculation of the flooded area the NDWI Indexes (McFeeters, 1996; Gao, 1996), and the MNDWI index (Xu, 2006) were applied and tested. The estimation of stored volume of water was made based on the area indices and a cross-check between real stored volume and calculated volume was made. Finally, an analysis on the selection of the best fit water indices was made. The results of every case studies herein displayed showed a quantifiable proficiency and reliability for quite a varied natural conditions. As a conclusion, this methodology could be seen as a tool for water resources management in developing countries, and not only, to measure automatically trends of stored volumes and its relation with the precipitation, and could eventually be extended to other types of surface water bodies, as lakes and coastal lagoons.
Mar 2022Published in Water Resources Research volume 58 issue 3. 10.1029/2021WR030026