New data sources in process-based hydrological modeling

New types of hydrological data from different parts of the water cycle, from many locations and at many times, create conditions to fundamentally improve our hydrological modeling tools. Some examples that can help make models more robust, in a time of changes in regards to both society and climate, include: high-resolution satellite measurements of water levels in lakes and watercourses, innovative applications of drones for snow and watercourse mapping, and precipitation analyses based on signals from cell towers.

The ability to generate hydrological data using satellites, drones, cell towers, and other measurement systems has never been greater than now. Machine learning and data-driven models are other methods for new data sources that together create entirely new conditions for developing hydrological models with higher resolution in time and space and better descriptions of different hydrological processes. The new models, in turn, create conditions to better utilize various types of data for calibration, assimilation, and uncertainty estimations.

Additionally, SMHI has a strong international position and actively contributes to the Agenda 2030 work for climate transition with service exports to data-poor countries. Here, for example, combinations of satellite and drone data can help improve model results from World-Wide HYPE so they can be fully used in warning services and local community planning.

Research lead: David Gustafsson

A drone above water.