Evaporative water loss of 1.42 million global lakes
Gang Zhao, Yao Li, Liming Zhou and Huilin Gao
Summarized by Michael Hallinan
What data was used? A series of geospatial data containing information on global lakes that are over 0.1 square kilometers (approximately 328 square feet) was sourced from HydroLAKES, a database centered around mapping the global freshwater. The data included a total of 1,427,688 water bodies, of which 6715 are reservoirs. In addition to this, three sets of meteorological data from TerraClimate, ERA5, and GLDAS were used to cancel bias as each dataset was developed independently through different institutes with a wide range of input sources. Lastly, a series of lake ice coverage and lake evaporation data were obtained from the Natural Snow and Ice Data Center and previous studies, respectively.
Methods: Using the geospatial data on lakes, a series of calculations was performed to estimate potential water loss due to evaporation of lakes and reservoirs. This was performed through calculating the change in heat stored by the body of water using the density, specific heat of water, water depth, and change in water temperature. Then an estimation of lake evaporation rate was performed using vapor pressure, net radiation, change in heat, surface area, as well as wind and other environmental data. In addition to this, further data processing occurred to account for ice coverage as well as to remove biases in satellite-sourced data caused by cloud coverage.
Results: This study created a dataset of evaporative water loss from 1958 to 2018 containing estimates of monthly evaporative loss of over 1.42 million lakes world wide. The most notable observations of this dataset are that the long-term average global lake evaporation has increased by 3.12 cubic kilometers per year in volume (roughly 0.75 cubic miles) while the average currently is 1500±150 cubic kilometers (roughly 932 cubic miles). This trend is likely a result of three main factors: Around 58% of this increase is a consequence of increased evaporation rate, 23% is caused by decreasing lake ice coverage, and 19% stems from an increase in lake surface area. In addition to this, these three factors have an identifiable pattern in their global distribution. High-latitude and high-altitude regions such as Tibetan Plateau and northern Eurasia show amplified effects of climate change on ice duration and as a result evaporation, likely having accelerated evaporation in the future.

Why is this study important? This dataset is essential to understanding global evaporative loss and the response of bodies of water to global warming. This dataset is the first of its kind to provide long-term monthly evaporation data on a global scale. This information can be used in the context of water availability estimations as well as in climate models. Although previous studies about water evaporation have been performed, many of them focused on only a few environmental parameters such as lake surface temperature, lake and river ice, or other attributes. This knowledge will be imperative in improving our overall understanding of the effects of lake evaporation
The big picture: A dataset of evaporation data comprising 1.42 million lakes from 1958 to 2018 was formed through a mixture of geospatial, meteorological, and lake ice coverage data. This dataset is the first of its kind and can be used to better understand water availability as well as water bodies’ reaction to climate change. Lastly, through this dataset it was discovered that there is an increase in water evaporation of about 3.12 cubic kilometers per year in volume (roughly 0.75 cubic miles).
Citation: Zhao, G., Li, Y., Zhou, L. et al. Evaporative water loss of 1.42 million global lakes. Nat Commun 13, 3686 (2022). https://doi.org/10.1038/s41467-022-31125-6