Caravan MultiMet (Part 2, Forecasts): Extending Caravan with Multiple Weather Nowcasts and Forecasts

Caravan MultiMet is a novel extension to Caravan, focusing on enriching the meteorological forcing data. Our extension adds three precipitation nowcast products (CPC, IMERG v07 Early, and CHIRPS) and three weather forecast products (ECMWF IFS HRES, GraphCast, and CHIRPS-GEFS). Since all data is kept in it's original time zone (UTC+0) and the ERA5-Land data in the original Caravan data set is shifted to local time of each gauge, we also include ERA5-Land reanalysis data in this extension, matching the UTC-0 timezone for all gauges of the other forcings. This part of the extension includes the forecast products.

The inclusion of diverse data sources, particularly weather forecasts, enables more robust evaluation and benchmarking of hydrological models, especially for real-time forecasting scenarios. To the best of our knowledge, this extension makes Caravan the first open large-sample hydrology dataset to incorporate weather forecast data.

The data is also publicly available on Google Cloud Platform (GCP), and we provide below a colab with an example of how to access it which does not require downloading the entire dataset.

Additional resources:

• The Caravan GitHub repository includes further information and links to other extensions.
• The original Caravan paper