Geospatial assessment for crop physiological and management improvements with examples using the simple simulation model. Issue 2 (27th March 2020)

• Record Type: Journal Article
• Authors: Sinclair, Thomas R.; Soltani, Afshin; Marrou, Helene; Ghanem, Michel; Vadez, Vincent
• Is Part Of: Crop science. Volume 60:Issue 2(2020)
• Page Start: 700
• Page End: 708
• Publication Date: 2020-03-27
• Subjects: Crop science -- Periodicals; Cultures -- Périodiques; Cultures de plein champ -- Périodiques; Crop science; Nutzpflanzen; Zeitschrift; Pflanzenbau; Periodicals; 633
• DOI: 10.1002/csc2.20106
• Languages: English
• ISSNs: 0011-183X

Abstract

Abstract: Physiological and management modifications to increase crop yields require an appreciation of the impact of these changes on a geospatial basis. It is quite possible that the yield response to any modification may vary from positive in one location to negative in another location. Therefore, tools to undertake geospatial analysis are required to assess the overall impact. It is argued that mechanistic models based on the physics and physiology of plant development, growth, and yield formation are required for such tasks. Several criteria in selecting model tools are discussed. First, models that need to be "calibrated" are not suited for geospatial assessments because the calibration processes causes the model to be an empirical representation of the calibration data and limited to the calibration environment. Extreme caution is needed to extrapolate model use beyond the domain of calibration, since geospatial analysis requires the model to be run for a range of geographical locations over a number of growing seasons. Second, to readily understand the output from simulations across space and time, models needs to be compact and transparent so output that seems inconsistent or not intuitively obvious can be tracked to the critical features in the model. Models with a smaller number of parameters are likely to be more transparent. Finally, it is necessary that the robustness of the model has been tested against a range of environmental conditions. In this paper, we discuss the example of the Simple Simulation Model (SSM) as an option that meets these criteria.