Modeling Soil-crop Interactions for Site Specific Decision Support in Nitrogen Management Concerning Aspects of Yield Production and Water Protection.
K. Christian Kersebaum, Inst for Landscape Systems Analysis, ZALF, Eberswalder Str. 84, 15374 Muencheberg, Germany, Karsten Lorenz, Univ of Applied Science Eberswalde, Friedrich-Ebert-Str. 28, 16225 Eberswalde, Germany, Hannes Isaak Reuter, European Commission, Soil and Waste Unit, TP-280, Inst of Environment and Sustainability, Joint Research Centre, Ispra, Italy, and Ole Wendroth, Univ of Kentucky, Dept. of Plant and Soil Sciences, N-122M Agr. Sci. North, Lexington, KY 40546.
Spatial variability within fields influences soil water, nutrient and crop growth dynamics. Within the framework of precision farming, technical solutions are provided to consider these spatial patterns for site specific management, to improve productivity and nutrient efficiency and to reduce pollution to the environment. Depending on climatic conditions, spatial yield patterns are often not stable in time because weather conditions favor different processes from year to year. For the farmer, it is important to apply an appropriate amount of fertilizer corresponding to the site-specific yield expectations to optimize nutrient use efficiency. His aims are often in conflict with the interest of water suppliers to keep nutrient loads of the seepage water low. The estimation of site specific management limitations which are required to maintain the groundwater quality as well as their consequences for crop production are of high interest for both sides. Simulations of water and nitrogen dynamics and crop growth of winter wheat were performed for two fields in different landscapes of Germany with the model HERMES considering within-field soil variability. Spatial yield data and soil mineral nitrogen contents of three years were used to validate the model. Real time model based site specific fertilizer recommendations were compared to soil sample/ N-sensor based nitrogen fertilization recommendation for 2 years concerning crop yields and residual mineral nitrogen after harvest. Results show that model recommendations were 40 kg N ha-1 less than the measurement based recommendation without yield losses. Results of zero plots indicated a decrease of the annual mineralization after 3 years of no nitrogen application. A virtual fertilization trial was performed using long term weather data to estimate fertilization effects on the average nitrogen concentration of seepage over more than 20 years. The results were used to determine zones of stable high and low yields and areas where fertilizer application should be limited to meet the drinking water standard with respect to nitrate concentration. The estimated fertilization limits can be used as an overlay for the site specific fertilizer application. The investigated field showed only small areas where fertilizer application should be restricted due to unfavorable conditions for crop growth.