A New High Resolution Experiment with the ParFlow Hydrologic Model to Forecast Soil Moisture in Germany

Abstract:

In the context of the spring and summer droughts that have affected Europe over the last three years, monitoring and forecasting soil moisture becomes increasingly important for the agricultural sector to mitigate the impacts of water stress on crops.

Here, we present a very high resolution model experiment to forecast the evolution of soil moisture. The setup has been developed within the ADAPTER project (www.adapter-projekt.de) of the Helmholtz Association of German Research Centres, which seeks to develop products to improve the resilience of agriculture to extreme weather conditions and climate change in Germany.

For our experiment, we use the hydrologic model ParFlow with its CLM (Common Land model) module at approximately 500m resolution over Germany and the neighboring regions. The model features 15 soil layers reaching from the surface to 60m depth. The soil hydraulic properties are defined on the basis of SoilGrids and of the International Hydrogeological Map of Europe, and the land cover is taken from the CORINE Land Cover database. The atmospheric forcing is provided by forecast products from the ECMWF (European Centre for Medium-Range Weather Forecasts). This setup allows for a complete three dimensional representation of the soil water budget as well as the water and energy flux exchanges with the atmosphere.

With this experimental setup, soil moisture forecasts can be performed on daily and subseasonal time scales.

Forecasts of soil moisture with ParFlow/CLM over the upcoming days provide applicable information, on e.g. the water stress of plants (via diagnostics such as plant available water) and thus the needs for irrigation, the trafficability and workability of the fields, or the presence and amount of seepage water, which can induce a leakage of nutrients and pollutants from the upper layers.

On the subseasonal scale, ParFlow/CLM is driven by ECMWF ensemble forecasts to predict the evolution of the soil water state. This is not only interesting for estimating the groundwater recharge during the autumn and winter seasons, but it gives also a probability range of upcoming agricultural droughts, which is of particular interest during the growing season.