The Diurnal Cycle of Precipitation according to Multiple Decades of Global Satellite Observations, Three CMIP6 Models, and the ECMWF Reanalysis

Summary

This validation study assesses how three state-of-the-art CMIP6 climate models and ERA5 reanalysis represent the sub-daily (diurnal) cycle of precipitation using high-resolution satellite observations from NASA's IMERG product across the tropical and subtropical domain (60°N–60°S). The analysis reveals persistent model limitations: systematic underestimation of oceanic precipitation amplitudes, overestimation over land in ERA5, and consistent phase shifts with simulated precipitation peaks occurring 1–4 hours too early, particularly over land. These findings underscore that improving representation of sub-daily precipitation processes—critical for weather and climate prediction—remains an unresolved challenge in contemporary climate model development.

UK applicability

Whilst this study focuses on tropical and subtropical precipitation patterns, the methodological approach and findings on model biases in sub-daily precipitation timing are relevant to UK climate science and weather prediction, where accurate representation of diurnal convection is important for operational forecasting. The identified phase-shift biases may have implications for UK-focused downscaling and impact assessments that rely on CMIP6 models.

Key measures

Diurnal precipitation amplitude (percentage of mean); diurnal phase (time of maximum precipitation); model-observation bias; interannual variability in timing of precipitation maxima

Outcomes reported

The study evaluated the representation of sub-daily precipitation cycles in three CMIP6 climate models and ERA5 reanalysis using NASA's IMERG satellite product across the tropics and subtropics. It quantified systematic biases in precipitation amplitude and phase across oceanic and land regions.

Topic tags