Green et al. (2026) Global Intensity-Duration-Frequency curves based on observed sub-daily rainfall (GSDR-IDF)

Identification

• Journal: Scientific Data
• Year: 2026
• Date: 2026-02-14
• Authors: Amy C. Green, Selma B. Guerreiro, Hayley J. Fowler
• DOI: 10.1038/s41597-026-06858-4

Research Groups

• School of Engineering and Tyndall Centre for Climate Change Research, Newcastle University, Newcastle upon Tyne, UK

Short Summary

This study presents GSDR-IDF, a global dataset of Intensity-Duration-Frequency (IDF) curves derived from over 24,000 quality-controlled sub-daily rain gauge records, providing a crucial resource for hydrological modeling, engineering design, and flood-risk assessment. It addresses the lack of comparable global IDF estimates by applying robust extreme value analysis methods to generate return levels for various durations and return periods.

Objective

• To develop and provide a global, openly accessible, traceable, and reproducible dataset of Intensity-Duration-Frequency (IDF) curves based on quality-controlled sub-daily rain gauge observations, serving as a new benchmark for global rainfall extremes.

Study Configuration

• Spatial Scale: Global, covering all major climate regions, derived from point-based rain gauge observations.
• Temporal Scale: Sub-daily rainfall extremes (1-hour, 3-hour, 6-hour, and 24-hour durations) for 10-year, 30-year, and 100-year return periods. The underlying rain gauge data record lengths vary from greater than 1 year to 104 years, with annual maxima (AMAX) time-series extending up to 2019.

Methodology and Data

• Models used:
  • Extreme Value Analysis (EVA)
  • Generalized Extreme Value (GEV) distribution, fitted using L-moments.
  • Single Gauge Analysis (SGA)
  • Regional Frequency Analysis (RFA), implemented using the lmomRFA package.
  • Power-law model (D = αt^β for depth-duration, I = αt^(β-1) for intensity-duration) for curve estimation.
• Data sources:
  • Global Sub-Daily Rainfall dataset (GSDR), comprising over 24,000 quality-controlled hourly rain gauge records.
  • Annual Maxima (AMAX) time-series derived from GSDR.
  • PXR2 dataset (Parametrized eXtreme Rain – 2 parameters) for comparison, based on ERA5 reanalysis data.

Main Results

• The GSDR-IDF dataset provides 23,985 fitted Intensity-Duration-Frequency (IDF) curves for 10-year, 30-year, and 100-year return periods, derived from over 24,000 quality-controlled sub-daily rain gauge records globally.
• The dataset includes return level estimates from both Single Gauge Analysis (SGA) and Regional Frequency Analysis (RFA), with RFA effectively doubling the number of gauges globally where extreme value analysis could be applied (4,065 to 4,181 gauges for RFA vs. 2,018 to 2,053 for SGA).
• The Generalized Extreme Value (GEV) distribution, fitted using L-moments, was found to adequately capture hourly and multi-hour rainfall extremes.
• Spatial consistency analysis revealed that differences in return level estimates between neighboring gauges increase significantly up to a separation distance of 200 kilometers, stabilizing thereafter. A global distance threshold of 100 kilometers is recommended for reliable application of IDF curves.
• SGA consistently exhibited greater variability in return level differences compared to RFA.

Contributions

• This study delivers the first global dataset of Intensity-Duration-Frequency (IDF) curves derived directly from quality-controlled sub-daily rain-gauge observations, establishing a new benchmark for global rainfall extremes.
• It provides an openly accessible, traceable, and reproducible resource that significantly enhances the accessibility and precision of global IDF estimation.
• The dataset supports a wide range of cross-disciplinary applications, including robust flood-risk assessment, hydrological modeling, engineering design, and climate-resilience planning.
• By incorporating both Single Gauge Analysis (SGA) and Regional Frequency Analysis (RFA) estimates, the dataset improves robustness and allows for direct comparison between methodologies.

Funding

• INTENSE project (European Research Council, grant ERC-2013-CoG-617329)
• IMPETUS4CHANGE (I4C) project (grant agreement ID: 101081555, HORIZON-CL5-2022-D1-02-04)
• Climate+ Co-Centre (managed by Research Ireland, Northern Ireland’s Department of Agriculture, Environment and Rural Affairs (DAERA), and UK Research and Innovation (UKRI), funded via the UK’s International Science Partnerships Fund (ISPF) and the Irish Government’s Shared Island initiative)

Citation

@article{Green2026Global,
  author = {Green, Amy C. and Guerreiro, Selma B. and Fowler, Hayley J.},
  title = {Global Intensity-Duration-Frequency curves based on observed sub-daily rainfall (GSDR-IDF)},
  journal = {Scientific Data},
  year = {2026},
  doi = {10.1038/s41597-026-06858-4},
  url = {https://doi.org/10.1038/s41597-026-06858-4}
}