Abstract
Precipitation is a key component of the hydrological cycle and one of the most important variables in weather and climate studies. Accurate and reliable precipitation data are crucial for determining climate trends and variability. In this study, eleven different precipitation datasets are compared, six reanalysis and five observational datasets, including the reanalysis datasets ERA5 and WFDE5 from the ECMWF family, to quantify the differences between the widely used precipitation datasets and to identify their particular strengths and shortcomings. The comparisons are focused on the common time period 1983 through 2016 and on monthly, seasonal, and inter-annual times scales in regions representing different precipitation regimes, i.e., the Tropics, the Pacific Inter Tropical Convergence Zone (ITCZ), Central Europe, and the South Asian Monsoon region. For the analysis, satellite-gauge precipitation data from the Global Precipitation Climatology Project (GPCP-SG) are used as a reference. The comparison shows that ERA5 and ERA5-Land are a clear improvement over ERA-Interim and show in most cases smaller biases than the other reanalysis datasets (e.g., around 13% high bias in the Tropics compared to 17% for MERRA-2 and 36% for JRA-55). ERA5 agrees well with observations for Central Europe and the South Asian Monsoon region but underestimates very low precipitation rates in the Tropics. In particular, the tropical ocean remains challenging for reanalyses with three out of four products overestimating precipitation rates over the Atlantic and Indian Ocean.
Highlights
In meteorology, precipitation is usually defined as rain, snow, sleet, or hail falling towards the surface from a cloud
This study focuses on the Tropics, in their full coverage and separated into land- and ocean-only regions, and two additional regions within the Tropics that are known for special precipitation regimes, including regions with known high biases
The Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Climate Data Record (PERSIANN-CDR) is a retrospective precipitation dataset based on multi satellite data, developed by the Center for Hydrometeorology and Remote Sensing (CHRS) and designed to be used for climate and hydrological studies
Summary
Precipitation is usually defined as rain, snow, sleet, or hail falling towards the surface from a cloud. A clear limitation is, that biases can be introduced by non-perfect models including, for instance, unresolved processes relevant to clouds and precipitation formation or uncertainties in used parameterizations and initial conditions This is relevant in regions with sparse observations such as over the oceans or in high-latitude regions where there is little effective constraint of the reanalysis solution and the fields are largely driven by the model physics and parameterizations (e.g., [17]). The aim of this study is to quantify the differences between the widely used precipitation datasets generated from different data sources and to identify their particular strengths and shortcomings including the recent reanalysis product ERA5 and the bias-corrected product WFDE5. E-OBS is a daily gridded dataset with a high spatial resolution that covers the European region over land and is based on station data collated by the European Climate Assessment and Dataset (ECA&D) project. More information about the ERA5 dataset can be found in [16]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
