Speaker
Description
Grasslands are an important component of land use in Europe, and the differ considerably in species composition, yield potential and management. Typical use cases include grazing with animals and mowing for haymaking, but mixed use and other forms of use also exist. Wet grasslands have recently come into focus, as many of them have been drained in the past to prolong the production season and improve access with heavy machinery. Under drainage, the previously accumulated organic matter in the soils got in contact with oxygen, and began to mineralise in high rates. The massive amounts of CO2 that are consequently emitted contribute considerably to European nations’ CO2 balance, which is why rewetting strategies are now being developed to reverse this process.
Mechanistic agroecosystem models are often used to assess the performance of crop and grassland systems, looking at productivity and related ecosystem services and disservices. For grassland systems, the MACSUR project has earlier demonstrated that simulation models are much less developed and supported as compared to cropland systems. One of the reasons may be that grasslands systems seem much more complex to simulate, because of the multiple species being involved at the same time.
We are presenting here the recent advances of the MONICA model (Nendel et al. 2011) for grassland systems. The most important features include (i) the determination of grassland systems in Germany and their mowing frequency using remote sensing, (ii) the determination of lowland grasslands with frequent contact to ascending groundwater and the groundwater distance dynamics in these areas, (iii) the simultaneous simulation of carbon, nitrogen and biomass dynamics, and (iv) the response of the species communities to changes in groundwater levels.
Remote sensing provides a wall-to-wall 10m resolution map of permanent grassland systems in Germany (Schwieder et al. 2022). The mowing frequency can be used to provide indications for the use intensity. Based on this, the input of additional nitrogen can be assumed to inform MONICA. From groundwater head data and remote sensing, a groundwater distance map (first plant-accessible aquifer) has been created (Raza et al., under review). This map includes the average minimum and maximum distance within a year. MONICA assumes a sine wave between min and max value, and uses this groundwater baseline for daily updates from evapotranspiration and precipitation terms. Ascending groundwater is computed using empirical capillary rise rates, in dependency of soil texture and groundwater distance. This allows to simulate additional water supply to the grassland community, related productivity benefits and the effect of other biophysical processes (Khaledi et al. 2024b). The biggest challenge for mono-species simulation models is the fact that grassland communities change their composition in response to changes in water supply (Khaledi et al. 2024a). This requires that MONICA will need to communicate with a competition model. In this case, the GRASSMIND model (Taubert et al. 2020) will deliver species community changes in response to water supply, but also other environmental and management factors, so that MONICA can simulate the correct biomass growth on the basis of this information.
References
Khaledi, V., G. Lischeid, B. Kamali, O. Dietrich, M.F. Davies and C. Nendel (2024a): Challenges of including wet grasslands with variable groundwater tables in large-area crop production simulations. Agriculture, under revision.
Khaledi, V., R. Baatz, D. Antonijević, M. Hofmann, O. Dietrich, G. Lischeid, M.F. Davies, C. Merz and C. Nendel (2024b): Modelling water, carbon and nitrogen fluxes in a wet grassland at contrasting water tables. Sci. Total Environ., under revision.
Nendel, C., M. Berg, K.C. Kersebaum, W. Mirschel, X. Specka, M. Wegehenkel, K.O. Wenkel and R. Wieland (2011): The MONICA model: Testing predictability for crop growth, soil moisture and nitrogen dynamics. Ecol. Model. 222 (9), 1614–1625.
Raza, A., G. Lischeid and C. Nendel (under review): Predicting groundwater levels at high spatial resolution for Brandenburg using machine-learning models. J. Hydrol.
Schwieder, M., M. Wesemeyer, D. Frantz, K. Pfoch, S. Erasmi, J. Pickert, C. Nendel and P. Hostert (2022): Mapping grassland mowing events across Germany based on time series of Sentinel-2 and Landsat 8 data. Remote Sens. Environ. 269, 112795.
Taubert, F., J. Hetzer, J.S. Schmid and A. Huth (2020), The role of species traits for grassland productivity. Ecosphere 11(7), e03205.
| Keywords | Simulation; Modelling; Remote sensing; Grassland, Ecosystem services |
|---|
