Funk et al. (2026) The Climate Hazards Center Infrared Precipitation with Stations, Version 3

Identification

• Journal: Scientific Data
• Year: 2026
• Date: 2026-04-11
• Authors: Chris D Funk, Pete Peterson, Laura Harrison, Robert Saldivar, M. F. Landsfeld, Diego Pedreros, Shraddhanand Shukla, Andreas H. Fink, Frank Davenport, S. Peterson, William Turner, Austin Sonnier, Michael Budde, Karyn Tabor, J. P. Verdin, Disha Hauzaree, Mohamed Naim, Daniella Alaso, G. J. Husak
• DOI: 10.1038/s41597-026-07096-4

Research Groups

• University of California, Santa Barbara Climate Hazards Center, Santa Barbara, USA
• US Geological Survey, Earth Resources and Observation Science Center, Sioux Falls, SD, USA
• Karlsruhe Institute of Technology (KIT), Institute of Meteorology and Climate Research, Karlsruhe, Germany
• Science Systems and Applications, Inc and NASA Goddard Space Flight Center, Greenbelt, USA
• University School for Advanced Studies IUSS Pavia, Pavia, Italy
• Department of Engineering, University of Messina, Messina, Italy

Short Summary

This paper introduces and evaluates CHIRPS Version 3 (CHIRPS3), an enhanced quasi-global, high-resolution rainfall dataset that integrates satellite thermal infrared observations with a significantly expanded network of station data, demonstrating improved accuracy in representing observed precipitation mean and variance compared to its predecessor, CHIRPS2.

Objective

• To introduce and evaluate the performance of the Climate Hazards Center Infrared Precipitation with Stations, Version 3 (CHIRPS3), highlighting its improvements over previous versions and other precipitation products in terms of accuracy and data representation.

Study Configuration

• Spatial Scale: Quasi-global, spanning 60°N to 60°S and covering all longitudes, with a spatial resolution of 0.05°. Additional sub-domain products are available for Africa and Latin America.
• Temporal Scale: A 40+ year dataset from 1981 to near-present, with data available in daily, pentad, dekad, monthly, and annual timesteps.

Methodology and Data

• Models used:
  • Improved variance-preserving thermal infrared (TIR)-to-precipitation estimation method.
  • Gauge-undercatch correction applied to station data.
  • For daily products: ECMWF ERA Reanalysis v5 (ERA5) for the 'rnl' product and NASA IMERG Late V07 for the 'sat' product.
  • Performance evaluated against CHIRPS2, IMERG, PERSIANN-CCS, and GPI.
• Data sources:
  • High-resolution climatology.
  • Thermal infrared (TIR) geostationary satellite observations.
  • In-situ station observations (significantly more sources than CHIRPS2).
  • High-quality interpolated data used for validation in twelve regions with dense station coverage.

Main Results

• CHIRPS3 extends its quasi-global coverage to 60°S/N.
• The new version incorporates an improved variance-preserving TIR-to-precipitation estimation method and implements gauge-undercatch correction.
• CHIRPS3 utilizes a substantially larger number of station observations and sources compared to the original CHIRPS2 product.
• Evaluation demonstrated that CHIRPS3 represents both the observed mean and variance of precipitation more accurately than CHIRPS2.
• A case study in Morocco showed that CHIRPS3 better captures observed rainfall variability when compared to CHIRPS2.
• The study emphasizes the importance of applying gauge-undercatch correction to station data when validating CHIRPS3.

Contributions

• Development and release of CHIRPS Version 3, a significantly improved quasi-global, high-resolution rainfall dataset.
• Expansion of the dataset's spatial coverage to 60°S/N.
• Integration of an advanced variance-preserving TIR-to-precipitation estimation method and gauge-undercatch correction, enhancing data quality.
• Incorporation of a substantially larger and more diverse set of station observations, improving the blending process.
• Demonstrated superior accuracy of CHIRPS3 in capturing observed precipitation mean and variance compared to CHIRPS2 and other widely used satellite products.
• Provides practical guidance on the validation of CHIRPS3, specifically highlighting the necessity of gauge-undercatch correction for station data.

Funding

• Famine Early Warning Systems Network
• U.S. Geological Survey (USGS cooperative agreement G24AC00087, "New Tools for Forecasting and Monitoring Agroclimatic Conditions in the Developing World")
• MEASURES grant led by the University of Arizona ("Next Generation GPCP Global Precipitation with Improved Consistency, Timeliness and Resolution for Climate Change, Early Warning and Scientific Analyses")
• University Center for Atmospheric Research’s Enhancing Meteorological Networks Partnership (EMNP)

Citation

@article{Funk2026Climate,
  author = {Funk, Chris D and Peterson, Pete and Harrison, Laura and Saldivar, Robert and Landsfeld, M. F. and Pedreros, Diego and Shukla, Shraddhanand and Fink, Andreas H. and Davenport, Frank and Peterson, S. and Turner, William and Sonnier, Austin and Budde, Michael and Tabor, Karyn and Verdin, J. P. and Hauzaree, Disha and Naim, Mohamed and Alaso, Daniella and Husak, G. J.},
  title = {The Climate Hazards Center Infrared Precipitation with Stations, Version 3},
  journal = {Scientific Data},
  year = {2026},
  doi = {10.1038/s41597-026-07096-4},
  url = {https://doi.org/10.1038/s41597-026-07096-4}
}