Submission history
Received: 11 March 2022
Accepted: 31 March 2023
Published in print: 19 May 2023
Acknowledgments
We thank A. Temme, C. Martin, S. Nerem, and J. Rosentreter for comments on the results and K. Yang and K. Bogan for suggestions and assistance on the visualizations. We thank P. Lin for sharing the latest version of the GRADES data and H. Beck for sharing the latest version of the MSWEP data for our analyses. Assistance for processing climate and human water use datasets was provided by K. Yang. We also thank L. Patterson for help with downloading the in situ data of reservoirs and L. Pitcher and E. Knight for proofreading our manuscript. We thank C. Schwatke, C. Birkett, and S. Cooley for making the altimetry-derived water level data publicly available.
Funding: This study was supported by NOAA Cooperative Agreement with CIRES (NA17OAR4320101) to F.Y., NASA NIP grant (80NSSC18K0951) to B.L., and the âClimate Change Initiative Grant (4000125030/18/I-NB) to J.-F.C. and M.B.-N.â
Author contributions: F.Y. conceptualized the project. F.Y., J.W., J.-F.C., and M.B.-N. estimated and analyzed the water storage variability and trends. F.Y., B.R., and B.L. developed the statistical models. F.Y. and B.R. estimated the sedimentation-induced storage loss in reservoirs. F.Y. evaluated the potential impacts of drying lakes on population with inputs from B.L., B.R., and Y.W. F.Y. conducted the validation. F.Y. performed the visualization with inputs from B.L., J.W., and Y.W. F.Y. wrote the manuscript with inputs from B.L., B.R., and J.W. All authors read and commented on drafts of this paper.
Competing interests: The authors declare no competing interests.
Data and materials availability: The Landsat images, including Landsat 5 Thematic Mapper, Landsat 7 Enhanced Thematic Mapper-plus, and Landsat 8 Operational Land Imager, are available from the US Geological Survey at
http://earthexplorer.usgs.gov and the Google Earth Engine platform at
https://earthengine.google.com. ICESat and ICESat-2 data are available from the National Snow and Ice Data Center (NSIDC) at
https://nsidc.org/data. Water levels derived from ICESat-2 are available at
https://doi.org/10.5281/zenodo.4489056. Water level products from radar altimeters can be downloaded from the Hydroweb at
http://hydroweb.theia-land.fr, the Database for Hydrological Time Series of Inland Waters (DAHITI) at
https://dahiti.dgfi.tum.de/en, and the USDA Global Reservoir and Lake Monitor database at
https://ipad.fas.usda.gov/cropexplorer/global_reservoir. The CryoSat-2 data are available from the European Space Agency (ESA) at
https://earth.esa.int/eogateway/catalog/cryosat-products. The Global Reservoir Bathymetry Dataset can be downloaded from
https://dataverse.tdl.org/dataset.xhtml?persistentId=doi:10.18738/T8/TO5HJG. The Global Surface Water (GSW) dataset is available from
https://global-surface-water.appspot.com/ and the Google Earth Engine platform at
https://earthengine.google.com. Reservoir sedimentation survey data from the US Army Corps can be accessed at
https://water.usace.army.mil/ and
https://nicholasinstitute.duke.edu/reservoir-national-trends/sediment/. USGS gauge data can be downloaded from
https://waterdata.usgs.gov/nwis/, US Army Corps gauge data can be downloaded from
https://water.usace.army.mil/ and
https://nicholasinstitute.duke.edu/reservoir-data/, California Department of Water Resources gauge data can be downloaded from
https://cdec.water.ca.gov/, gauge data from Texas Water Development Board can be downloaded from
https://waterdatafortexas.org/reservoirs/statewide, gauge data from Spain can be downloaded from
https://ceh.cedex.es/anuarioaforos/afo/embalse-nombre.asp, and gauge data from Bureau of Meteorology in Australia can be downloaded from
http://www.bom.gov.au/waterdata/. The HydroLAKES database can be downloaded from
https://www.hydrosheds.org/page/hydrolakes. The Georeferenced global Dams And Reservoir dataset (GeoDAR) can be downloaded from
https://doi.org/10.5281/zenodo.6163413. The database of Roller-Compacted Concrete (RCC) dams can be accessed at
http://www.rccdams.co.uk/. The Global Lake area, Climate, and Population (GLCP) dataset can be downloaded at
https://portal.edirepository.org/nis/mapbrowse?packageid=edi.394.4. The HydroSHEDS dataset can be downloaded from
https://hydrosheds.org/page/overview. The Climatic Research Unit (CRU) data are available from
https://crudata.uea.ac.uk/cru/data/hrg/. ECMWF Reanalysis v5 (ERA5) data are available from
https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5. Modern-Era Retrospective analysis for Research and Applications version 2 (MERRA-2) are available from
https://gmao.gsfc.nasa.gov/reanalysis/MERRA-2/. Multi-Source Weighted-Ensemble Precipitation (MSWEP) are available from
https://www.gloh2o.org/mswep/. Global Historical Climatology Network data are available from
https://www.ncei.noaa.gov/products/land-based-station/global-historical-climatology-network-monthly. Global Land Evaporation Amsterdam Model (GLEAM) data are available from
https://www.gleam.eu/. Global Reach-scale A priori Discharge Estimates for SWOT (GRADES) dataset can be downloaded from
https://www.reachhydro.org/home/records/grades. The reconstructed human water use data derived from four global hydrologic models can be downloaded from
https://zenodo.org/record/1209296#.YZPcr2DMKM8. The water body masks delineated from the GSW dataset, lake volume time series derived from Landsat images and satellite altimeters, lake volume trend estimates, attribution, and all validation analyses are available on the Zenodo data repository at
https://zenodo.org/record/7946043. Water storage trends and drivers of studied large lakes are available in the interactive map at
https://cires.colorado.edu/globallakes.
Code availability: R scripts that were used to process hydroclimate and human water use data, to derive water levels from ICESat and ICESat-2, to construct lake water storage time series, to estimate the mean rate of sedimentation in reservoirs, to estimate trends in lake water storage and basin aridity, to conduct validation, to construct regression model ensemble, and to estimate affected population are available on CodeOcean at
https://codeocean.com/capsule/0322198/tree/v1. JavaScript scripts for mapping water areas from Landsat images and IDL Scripts for processing CyoSat-2 data are available in the supplementary materials.