Using crop simulation to optimize variable rate experimentation

corn, economic optimum N rate, forecast, modeling, APSIM, in-season nitrogen management, nutrientrecommendation

Mandrini, German ; Mieno, Taro, ; ,Paulson, Nicholas D., ; ,Martin, Nicolas F. ; Bullock ; David S.

关键词: corn, economic optimum N rate, forecast, modeling, APSIM, in-season nitrogen management, nutrientrecommendation;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/101594/MANDRINI-THESIS-2018.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Researchers working on a USDA-sponsored research project are exploring a new concept of on-farm experimentation (OFE). These trials are implemented by farmers at their fields, in a similar way to how they would plant a regular production crop. This concept generates large amounts of data at low cost that, after processing, will generate local models about the yield response function within a field.At the time of this work, the research group is running more than 100 trials in different states and countries. There are questions related to how to optimize OFE. To address those questions, the APSIM crop growth model was used to simulate the concept of running on-field trials, use that information to calculate the Economic Optimum Nitrogen Rate (EONR), and finally use that EONR in a regular crop production. Spatially variable layers of data that characterized a field were transformed into APSIM parameters. Daily weather events were obtained from historical weather data for the field’s county. Economic analysis of different strategies was performed, which involved testing if the increase in revenues due to including more variables or running more trials outperforms the cost, and how weather affects the results. The results will help to optimize the actual protocol that is guiding the implementation of the trials. Key results obtained by this research were: (1) The value of conducting trials and using that information for N-management advice was 9.8 $/ha. (2) The added value of gathering soil sampling data at the same time was 7.4 $/ha. (3) The optimal time to stop running trials and start using the information for N-management advice was one or two years, depending on the weather. (4) Conducting trials and using that information for N-management advice decreased N-leaching by 10.4 kg/ha. Performing soil sampling tests together with running trials made N-management advice increase the efficiency and reduced N-leaching by 5.9 kg/ha more. (5) A tentative rule for deciding if a one trial year is sufficient or if one more year is needed was obtained by determining the likelihood of the weather of the trial year compared with the historic weather. These results provide insights that will be helpful to optimize the protocol that guide OFE and help farmers increase profits in the fastest way and decrease the environmental impact of nitrogen fertilization.