Informal Seminar: Soil-moisture atmosphere coupling hotspots and their importance for sub-seasonal and seasonal prediction

Due to its inherent long-memory and profound role in modulating surface exchanges of heat and moisture, soil moisture is a major source of potential forecast skill at sub-seasonal and seasonal timescales. In some regions, known as “hotspots”, soil moisture can influence near-surface temperature and precipitation locally, but can also influence remote areas by exciting quasi-stationary Rossby waves in the atmosphere.

In this talk I will give an overview of these important predictability mechanisms and discuss a recent study which highlighted the quality of land-surface initial conditions and the current representation of these hotspots as limiting factors in the skill of forecasting systems contributing to the C3S multi-model outlook. Regions where forecasting systems overestimate the strength of soil-moisture atmosphere coupling tend to have particularly low skill.

I will also discuss ongoing efforts at ECMWF to address these limitations through improved land-surface data assimilation and increased model complexity, as part of the CERISE project. One notable development is the incorporation of time-varying vegetation, which is held static in current operational systems but is thought to provide a source of predictability at seasonal timescales.