|JEGO, GUILLAUME - Agriculture And Agri-Food Canada|
|Rotz, Clarence - Al|
|GELANGER, GILLES - Agriculture And Agri-Food Canada|
|TREMBLAY, GAETAN - Agriculture And Agri-Food Canada|
|CHARBONNEAU, EDITH - Laval University|
|PELLERIN, DORIS - Laval University|
Submitted to: Canadian Journal of Plant Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/23/2015
Publication Date: 9/18/2015
Citation: Jego, G., Rotz, C.A., Belanger, G., Tremblay, G., Charbonneau, E., Pellerin, D. 2015. Simulating forage crop production in a northern climate with the Integrated Farm System Model. Canadian Journal of Plant Science. 95:745-757.
Interpretive Summary: In the current century, agriculture will face major challenges regarding food production and environmental protection. The dairy industry is a major agricultural sector around the world, and global milk production has increased in the last three decades by more than 50%. Perennial and annual forage production for animal feed uses a significant portion of Canada’s total agricultural land, especially in the provinces of Quebec and Ontario, where dairy production is dominant. Previous studies have shown that dairy production’s contribution to environmental pollution is a source of concern. Many studies have evaluated the effects of management strategies or climate variability on agricultural production and the environment at the field scale, and some indicate that field-based activities, such as manure application, may offset efforts made through animal feeding and manure handling to improve the environmental and economic sustainability of dairy farms. Therefore, a comprehensive and integrated approach is required to design eco-efficient, farm-scale management strategies. This can be achieved using models such as the Integrated Farm System Model (IFSM) that simulate the economic and environmental implications of practices at the whole-farm scale. The ability of IFSM to simulate crop production in northern regions, such as Eastern Canada, characterized by short growing seasons and long periods of soil freezing and snow cover, needed to be verified. When crop growth parameters, soil properties, crop management, and weather data were precisely set, IFSM predicted the yield and nutritive value of perennial forage crops as accurately as field-scale crop models. The set of crop growth parameters for timothy and alfalfa presented in this study extends the potential area of application of IFSM to northern regions such as Canada and Northern Europe. The results of the present study reinforce previous results showing that even though improvements can be made, IFSM accurately predicted the forage yield and fiber concentration of each cut within the growing season of a timothy–alfalfa mixture. The regional yield evaluation of the model for corn, soybeans, and barley confirmed the robustness of the model, which can be easily calibrated for simulating annual yields of these crops under northern climate conditions and provides accuracy comparable with field-scale crop models. The verified farm model can now be used to evaluate strategies for mitigating environmental impacts and adapting dairy farms to climate variability.
Technical Abstract: Whole-farm simulation models are useful tools for evaluating the effect of management practices and climate variability on the agro-environmental and economic performance of farms. A few process-based farm-scale models have been developed, but none have been evaluated in a northern region with a short growing season and a long period with snow cover. The study objectives were 1) to calibrate and evaluate the grass sub-model of the Integrated Farm System Model (IFSM) for simulating the yield and nutritive value of timothy and alfalfa, grown alone or in a mixture, using experimental field data from across Canada, and 2) to assess IFSM for simulating the yield of major annual crops grown on dairy farms in Eastern Canada using regional yield data from two contrasting regions. Several timothy and alfalfa datasets combining sites, years, harvests, and N fertilization rates were used to calibrate and evaluate the model. For timothy, the model’s performance was satisfactory for simulating dry matter yield and P uptake, with a normalized root mean square error (NRMSE) below 30% and a model efficiency (EF) higher than 0.60. For N uptake, the scatter was large (NRMSE = 49%), but the EF (0.44) was still acceptable. The neutral detergent fiber concentration was well simulated at all but one site, where it was underestimated (EF < 0). For alfalfa, the model’s performance was fairly good for predicting dry matter yield (NRMSE = 30%, EF = 0.51), and the NRMSE was good (<20%), but the EF was negative for N uptake and neutral detergent fiber concentration because of a small range in the measured data. For P uptake, the range of measured data was also small, leading to poor NRMSE and EF values, but the average P uptake was well simulated (mean error close to 0). The model’s performance for simulating the yield of annual crops was generally good, with a bias less than or equal to 10% and an NRMSE less than 30%. However, the EF was negative in some cases, indicating that the model was not always able to capture the inter-annual variability. Adding timothy and alfalfa to the grass sub-model of IFSM and verifying the model’s performance for annual crops confirmed that IFSM can be used in northern areas of North America. In addition, the model was able to simulate the yield and nutritive value of a timothy–alfalfa mixture, which is the most common perennial mixture used in Canada.