The application of predefined weather patterns over India within probabilistic medium-range forecasting tools for high-impact weather

A method is presented for deriving probabilistic medium-range (1-to-2-week) weather pattern forecasts for India. This method uses an existing set of 30 objectively derived daily weather patterns, which provide climatological representations for unique states in the large-scale circulation over India. Weather pattern forecast probabilities are based on the number of ensemble members objectively assigned to each weather pattern. Two summer monsoon case studies illustrate the best use of the forecasting tool within medium-range guidance, such as highlighting the most likely weather pattern transitions and relating these to the likelihood of weather impacts. Forecast skill is evident out to at least 10-15 days. Winter dry period weather patterns have the highest forecast skill, closely followed by retreating monsoon weather patterns. In contrast, monsoon onset and break monsoon weather patterns have the lowest forecast skill. Finally, a prototype weather pattern forecast climatology application is presented for use in highlighting when extreme rainfall is more likely than normal. This application is based on weather pattern empirical probabilities of threshold exceedances using a high-resolution regional reanalysis. The transitional pre- and post-monsoon seasons have the greatest variability in rainfall across all possible weather patterns, with a slight dip in variability during the main summer monsoon season. In contrast, very little variability across weather patterns is evident during the relatively dry winter months. This highlights the times of year when a climatology-based weather pattern forecasting approach may have its greatest benefits over that of a basic daily climatology.

Automated summary

A method for probabilistic medium-range weather pattern forecasts in India is presented, using a set of 30 daily weather patterns. The forecasts show skill up to 10-15 days and can be used to highlight weather pattern transitions and likelihood of weather impacts. The study also presents an application for highlighting periods of extreme rainfall based on weather pattern probabilities. The research demonstrates the importance and benefits of a weather pattern forecasting approach over basic daily climatology.