skip to main content

DOE PAGESDOE PAGES

Title: A remote sensing method for estimating regional reservoir area and evaporative loss

Evaporation from the water surface of a reservoir can significantly affect its function of ensuring the availability and temporal stability of water supply. Current estimations of reservoir evaporative loss are dependent on water area derived from a reservoir storage-area curve. Such curves are unavailable if the reservoir is located in a data-sparse region or questionable if long-term sedimentation has changed the original elevation-area relationship. In this paper, we propose a remote sensing framework to estimate reservoir evaporative loss at the regional scale. This framework uses a multispectral water index to extract reservoir area from Landsat imagery and estimate monthly evaporation volume based on pan-derived evaporative rates. The optimal index threshold is determined based on local observations and extended to unobserved locations and periods. Built on the cloud computing capacity of the Google Earth Engine, this framework can efficiently analyze satellite images at large spatiotemporal scales, where such analysis is infeasible with a single computer. Our study involves 200 major reservoirs in Texas, captured in 17,811 Landsat images over a 32-year period. The results show that these reservoirs contribute to an annual evaporative loss of 8.0 billion cubic meters, equivalent to 20% of their total active storage or 53% of totalmore » annual water use in Texas. At five coastal basins, reservoir evaporative losses exceed the minimum freshwater inflows required to sustain ecosystem health and fishery productivity of the receiving estuaries. Reservoir evaporative loss can be significant enough to counterbalance the positive effects of impounding water and to offset the contribution of water conservation and reuse practices. Our results also reveal the spatially variable performance of the multispectral water index and indicate the limitation of using scene-level cloud cover to screen satellite images. Finally, this study demonstrates the advantage of combining satellite remote sensing and cloud computing to support regional water resources assessment.« less
Authors:
ORCiD logo [1] ;  [2] ;  [3] ;  [4]
  1. Texas A & M Univ., Corpus Christi, TX (United States). School of Engineering and Computing Sciences
  2. Stanford Univ., CA (United States). Dept. of Earth System Science
  3. Texas A & M Univ., Corpus Christi, TX (United States). Center for Coastal Studies
  4. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Publication Date:
Report Number(s):
LA-UR-17-24804
Journal ID: ISSN 0022-1694; TRN: US1800800
Grant/Contract Number:
AC52-06NA25396; GEO/OAD-1342869
Type:
Accepted Manuscript
Journal Name:
Journal of Hydrology
Additional Journal Information:
Journal Volume: 555; Journal ID: ISSN 0022-1694
Publisher:
Elsevier
Research Org:
Texas A & M Univ., Corpus Christi, TX (United States); Stanford Univ., CA (United States); Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org:
USDOE; Texas Comprehensive Research Fund (TCRF) (United States); National Science Foundation (NSF)
Country of Publication:
United States
Language:
English
Subject:
58 GEOSCIENCES; evaporation; reservoir; Google Earth engine; MNDWI; freshwater
OSTI Identifier:
1415402

Zhang, Hua, Gorelick, Steven M., Zimba, Paul V., and Zhang, Xiaodong. A remote sensing method for estimating regional reservoir area and evaporative loss. United States: N. p., Web. doi:10.1016/j.jhydrol.2017.10.007.
Zhang, Hua, Gorelick, Steven M., Zimba, Paul V., & Zhang, Xiaodong. A remote sensing method for estimating regional reservoir area and evaporative loss. United States. doi:10.1016/j.jhydrol.2017.10.007.
Zhang, Hua, Gorelick, Steven M., Zimba, Paul V., and Zhang, Xiaodong. 2017. "A remote sensing method for estimating regional reservoir area and evaporative loss". United States. doi:10.1016/j.jhydrol.2017.10.007. https://www.osti.gov/servlets/purl/1415402.
@article{osti_1415402,
title = {A remote sensing method for estimating regional reservoir area and evaporative loss},
author = {Zhang, Hua and Gorelick, Steven M. and Zimba, Paul V. and Zhang, Xiaodong},
abstractNote = {Evaporation from the water surface of a reservoir can significantly affect its function of ensuring the availability and temporal stability of water supply. Current estimations of reservoir evaporative loss are dependent on water area derived from a reservoir storage-area curve. Such curves are unavailable if the reservoir is located in a data-sparse region or questionable if long-term sedimentation has changed the original elevation-area relationship. In this paper, we propose a remote sensing framework to estimate reservoir evaporative loss at the regional scale. This framework uses a multispectral water index to extract reservoir area from Landsat imagery and estimate monthly evaporation volume based on pan-derived evaporative rates. The optimal index threshold is determined based on local observations and extended to unobserved locations and periods. Built on the cloud computing capacity of the Google Earth Engine, this framework can efficiently analyze satellite images at large spatiotemporal scales, where such analysis is infeasible with a single computer. Our study involves 200 major reservoirs in Texas, captured in 17,811 Landsat images over a 32-year period. The results show that these reservoirs contribute to an annual evaporative loss of 8.0 billion cubic meters, equivalent to 20% of their total active storage or 53% of total annual water use in Texas. At five coastal basins, reservoir evaporative losses exceed the minimum freshwater inflows required to sustain ecosystem health and fishery productivity of the receiving estuaries. Reservoir evaporative loss can be significant enough to counterbalance the positive effects of impounding water and to offset the contribution of water conservation and reuse practices. Our results also reveal the spatially variable performance of the multispectral water index and indicate the limitation of using scene-level cloud cover to screen satellite images. Finally, this study demonstrates the advantage of combining satellite remote sensing and cloud computing to support regional water resources assessment.},
doi = {10.1016/j.jhydrol.2017.10.007},
journal = {Journal of Hydrology},
number = ,
volume = 555,
place = {United States},
year = {2017},
month = {10}
}