Evaporation (ET) is an important component of the water cycle and serves as a link between the global water cycle, energy cycle, and carbon cycle. Therefore, precise quantification of evapotranspiration is crucial for understanding various Earth system processes and subsequent social applications. The existing evapotranspiration retrieval methods either have limited spatiotemporal coverage or are largely influenced by input data selection, simplified model physics, or a combination of both. Here, we have developed a water balance based evapotranspiration dataset (ET-WB) using an independent conservation of mass method for global land and selected 168 major watersheds. We utilized multi-source datasets from satellite products, in-situ measurements, reanalysis, and hydrological simulations (23 precipitation datasets, 29 runoff datasets, and 7 reserve change datasets) to generate 4669 probability based unique combinations of evapotranspiration datasets between May 2002 and December 2021.
Three global datasets based on field observations were collected, including the Climate Research Unit Time Series (CRU TS) database, the Global Precipitation Climatology Center (GPCC) project, and the unified dataset of the National Oceanic and Atmospheric Administration's Climate Prediction Center (CPC Unified). These data typically rely on point scale rain gauges collected from around the world to interpolate gridded global products. In order to enrich our research, six remote sensing products were collected, namely Multi Satellite Integrated Retrieval (IMERG) for global precipitation measurement, Global Precipitation Climatology Project (GPCP), Estimation of Precipitation Climate Data Record from Remote Sensing Information Using Artificial Neural Networks (PERSIANN-CDR), Tropical Rainfall Measuring Mission with 3B43 algorithm, Global Precipitation Climatology Project (GPCP), Estimation of Precipitation Climate Data Record from Remote Sensing Information Using Artificial Neural Networks (PERSIANN-CDR), Tropical Rainfall Measurement Task using 3B43 algorithm (TRMM 3B43), Global Precipitation Satellite Mapping (GSMaP) and Climate Data Record using 3B43 algorithm. Disaster team infrared precipitation and station data (CHIRPS).
Generate the ET-WB dataset using the land water balance method. Through the Global Runoff Data Center (GRDC), https://www.bafg.de/GRDC/EN/Home/homepage_node.html ）Calculations were conducted on 168 major river basins worldwide, excluding Antarctica and Greenland.
We compared our results with four auxiliary global ET datasets and previous regional studies, and then rigorously discussed uncertainties, their possible sources, and potential methods to limit them. The seasonal cycle of global ET-WB shows a unimodal distribution, with the highest (median: 65.61 millimeters per month) and lowest (median: 36.11 millimeters per month) in July and January, respectively. The distribution range of different subsets is approximately ± 10 millimeters per month. The auxiliary ET products exhibit similar characteristics within the year, but with some overestimation or underestimation, which is entirely within the scope of the ET-WB set. We found that from 2003 to 2010, the global ET-WB gradually increased, then decreased during the period of 2010-2015, and then significantly decreased in other years, mainly due to changes in precipitation. Multiple statistical indicators show that in most watersheds, the accuracy of monthly ET-WB is quite good (e.g. relative deviation of ± 20%), which has improved on an annual scale. The long-term average annual ET-WB varies within 500-600 millimeters by -1 and is consistent with the observed trend estimates of four auxiliary ET products (543-569 mm yr-1). Although there are regional differences, it proves that aliens are increasing in the context of climate warming. The current dataset may contribute to several scientific assessments centered on water resource management for the benefit of the entire society.
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:11738.11.NCDC.ZENODO.DB6440.2024
10.5281/zenodo.8339655