Paper detail

A new low-cost technique improves weather forecasts across the world

Computer-generated forecasts divide the earth&#39;s surface into gridboxes, each now ~25% of the size of London, and predict one value per gridbox. If weather varies markedly within a gridbox forecasts for specific sites inevitably fail. A completely new statistical post-processing method, using ensemble forecasts as input, anticipates two gridbox-weather-dependant factors: degree of variation in each gridbox, and bias on the gridbox scale. Globally, skill improves substantially; for extreme rainfall, for example, useful forecasts extend 5 days ahead. Without post-processing this limit is < 1 day. Relative to historical forecasting advances this constitutes ground-breaking progress. The key drivers, incorporated during calibration, are meteorological understanding and abandoning classical notions that only local data be used. Instead we simply recognise that &#34;showers are showers, wherever they occur worldwide&#34; which delivers a huge increase in calibration dataset size. Numerous multi-faceted applications include improved flash flood warnings, physics-related insights into model weaknesses and global pointwise re-analyses.