Raincast: A seasonal forecasting system for drought management

Raincast is a seasonal forecasting system for predicting rainfall and reservoir inflows which alerts when water levels in reservoirs are below the guarantee curve. It integrates statistical methods to characterize the hydroclimatic variability, and machine learning techniques used in artificial intelligence. By combining these tools, Raincast is able to provide a more accurate long-term prediction of rainfall and streamflows.
Technology demonstrated in relevant environment.
Representative model or prototype system, which is well beyond that of TRL 5, is tested in a relevant environment. Represents a major step up in a technology’s demonstrated readiness. Examples include testing a prototype in a high-fidelity laboratory environment or in a simulated operational environment.
The system has been tested in a relevant environment, and used to forecast precipitation and reservoir inflows for an important company located in the Basque Country (Spain). Good performance indicators were achieved during these tests, and new technological improvements were additionally adopted. Furthermore, Raincast has been experimentally validated for a short number of local cases, but it requires a stronger validation using additional testing cases in order to confirm previous results and improve the accuracy of the outputs.

How does it work?

Meteobit provides a reliable solution for the forecasting of rainfall and streamflows at seasonal and sub-seasonal scales commonly used by the public water supply and agricultural sectors, and other industrial end users. Using ensemble weather forecasting for precipitation, Raincast has been designed to provide critical forecast guidance as well as a true sense of forecast risk. Meteobit's product is a valuable tool for anticipating hydrological droughts and managing water resources.