Wednesday, November 15, 2006 - 9:35 AM
300-5

Model-Based Nitrogen Fertilization Considering Agrometeorological Data.

Kurt Christian Kersebaum, Centre of Agricultural Landscape Research, Eberswalder Str. 84, Muencheberg, 15374, Germany

The spatial and temporal coincidence of nutrient supply and the demand of the crops is especially important for nitrogen to enhance nitrogen use efficiency and to prevent groundwater contamination. Soil and crop measurements are useful tools to obtain soil or crop nutrient status. However, they only give a snapshot of a present situation with a limited temporal validity. Temporal dynamics of soil-crop nitrogen dynamics can be simulated with process oriented agricultural system models  using temporal stable soil attributes being estimated once for a field and current management and agro-meteorological data. The concept of the model based fertilizer recommendation with the model HERMES will be shown. The model includes processes of soil water and nitrogen dynamics (mineralization, denitrification, transport), crop growth and development and N uptake. Simulation case studies will show the potential benefits considering available 3 to 5 day weather forecasts for different climatic conditions for fertilizer applications. To cover the longer period between two fertilizer applications scenario calculations can be used to estimate the required nitrogen fertilization using weather scenarios derived from long term weather data of the specific location. Fertilizer recommendations can be derived by subsequent model runs at different development stages combining actualized real weather data,  weather forecast and the weather scenarios of the specific site for a predictive calculation of the nitrogen deficiency until the next fertilization date. Examples will be shown of the operational application of the model HERMES to derive nitrogen fertilizer recommendations for entire fields or site specific recommendations using spatial variable input from a GIS. Results are compared to other methods based on soil and crop measurements. Additionally, retrospective long term simulations can be used to derive fertilizer recommendations or limitations concerning water protection requirements or yield potential.