Skillful precipitation nowcasting using deep generative models of radar

Precipitation nowcasting, the high-resolution forecasting of precipitation up to two hours ahead, supports the real-world socioeconomic needs of many sectors reliant on weather-dependent decision-making.

Nowcasting:

the World Meteorological Organization (WMO) defined nowcasting as a detailed analysis and description of the current weather and then forecasting ahead for a period from 0 to 6 h (WMO, 2019).

Forecasting:

the practice of predicting what will happen in the future by taking into consideration events in the past and present. Basically, it is a decision-making tool that helps businesses cope with the impact of the future’s uncertainty by examining historical data.

Why is precipitation nowcasting difficult:

precipitation prediction, poses complex tasks because they depend on various parameters to predict the dependent variables like temperature, humidity, wind speed and direction, which are changing from time to time and weather calculation varies with the geographical location along with its atmospheric variables.

Recently introduced deep learning methods use radar to directly predict future rain rates but their operational utility is limited because their lack of constraints produces blurry nowcasts at longer lead times, yielding poor performance on rarer medium-to-heavy rain events.

we present a deep generative model for the probabilistic nowcasting of precipitation from radar that addresses these challenges. Using statistical, economic, and cognitive measures, we show that our method provides improved forecast quality, forecast consistency, and forecast value. we show that our generative model ranked first for its accuracy and usefulness in 89% of cases against two competitive methods. We show that generative nowcasting can provide probabilistic predictions that improve forecast value and support operational utility, and at resolutions and lead times where alternative methods struggle.

Approaches based on deep learning have been developed that move beyond reliance on the advection equation5,6,14–19. By training these models on large corpora of radar observations rather than relying on in-built physical assumptions, deep learning methods aim to better model traditionally difficult non-linear precipitation phenomena, such as convective initiation and heavy precipitation.