Description
The Current Extreme Precipitation Risk dataset captures the near-term risk of extreme precipitation across the globe using proprietary forecasts of rolling precipitation totals. This current risk layer looks at the maximum of the daily extreme precipitation conditions across the globe for the next 1-3 days.
Technical Description
This methodology classifies extreme precipitation events using a 3-day rolling sum of forecast precipitation data. The classification system employs six categories (0-5) based on cumulative precipitation thresholds, where Category 0 represents no significant event (<25mm), and Categories 1-5 represent progressively more severe precipitation events. The higher threshold values (100mm, 200mm, and 300mm for Categories 3-5) are derived from the P-Cat precipitation event classification system developed by Slinskey et al. (2019), which provides standardized categories for identifying extreme precipitation events based on their potential hydrological impacts. The 3-day rolling accumulation window is particularly valuable for hydrological applications as it reflects the typical timescale over which precipitation events develop their full impact on watersheds and drainage systems.
The model generates a daily time series where each image represents the risk category based on the 3-day period ending on that date. The current extreme precipitation risk class is then derived by taking the maximum of these rolling risk classifications across the 1-3 day forecast period.