An IMERG-Based Optimal Extended Probabilistic Climatology (EPC) as a Benchmark Ensemble Forecast for Precipitation in the Tropics and Subtropics

Significance Statement Precise precipitation forecasts in the tropics and subtropics are relevant for a large and growing population. To gauge the success of improvements, an adequate baseline is needed. Here we use satellite-based rainfall estimates from 2001 to 2019 to define a climatological reference forecast that we call extended probabilistic climatology (EPC), as it combines rainfall observations from a window of +/- 15 days around the date of interest. We show that this simple approach outperforms current weather forecast models in some areas and forecast aspects but is inferior in others. To foster the use of this new benchmark in the scientific and forecasting communities, we provide the EPC data, scripts, and an interactive web tool to display EPC forecasts for selected locations. Current numerical weather prediction models show limited skill in predicting low-latitude precipitation. To aid future improvements, be it with better dynamical or statistical models, we propose a well-defined benchmark forecast. We use the arguably best available high-resolution, gauge-calibrated, gridded precipitation product, the Integrated Multisatellite Retrievals for GPM (IMERG) final run in a +/- 15-day window around the date of interest to build an empirical climatological ensemble forecast. This window size is an optimal compromise between statistical robustness and flexibility to represent seasonal changes. We refer to this benchmark as extended probabilistic climatology (EPC) and compute it on a 0.1 degrees x 0.1 degrees grid for 40 degrees S-40 degrees N and the period 2001-19. To reduce and standardize information, a mixed Bernoulli-Gamma distribution is fitted to the empirical EPC, which hardly affects predictive performance. The EPC is then compared to 1-day ensemble predictions from the European Centre for Medium-Range Weather Forecasts (ECMWF) using standard verification scores. With respect to rainfall amount, ECMWF performs only slightly better than EPS over most of the low latitudes and worse over high-mountain and dry oceanic areas as well as over tropical Africa, where the lack of skill is also evident in independent station data. For rainfall occurrence, EPC is superior over most oceanic, coastal, and mountain regions, although the better potential predictive ability of ECMWF indicates that this is mostly due to calibration problems. To encourage the use of the new benchmark, we provide the data, scripts, and an interactive web tool to the scientific community.

Automated summary

This study defines a climatological reference forecast called extended probabilistic climatology (EPC) using satellite-based rainfall estimates from 2001 to 2019 to gauge the success of improvements in precipitation forecasts. The EPC outperforms current weather forecast models in some areas and forecast aspects, providing a new benchmark for the scientific and forecasting communities.