Using multi-point simulation modeling to estimate the spatial variability of wheat grain yield and protein within farmers’ fields

Description of the topic

In crop development, there is spatial and temporal variability within a site, mainly because of soil nutrient availability and the soil’s physical and chemical properties. These variables, together with climate parameters, agronomic management choices and crop genotypes, compose the whole scenario where the crop will grow and will also cause variability in yield and quality. In terms of investigations and simulations of the main factors related to yield and quality, the use of crop models and the identification of important soil attributes during each crop cycle may help determine which nutrient-supply and variety-management practices should be used in the various cropping environments. The application of simulation models in agriculture is widespread across various industries for different uses. The global grain industry applies simulation models to answer numerous questions in areas such as biochemistry, agronomy, physiology and management.

Multi-point simulations allow point-based models to be applied multiple times within a single simulation, with data communication among all discrete points. Hence, each point in space requires its own input in the crop modules, such as a Geographic Information System (GIS) dataset, where it is possible to have several types of information for each sample point within a site. Therefore, each simulation point will have a reporting output with the studied factors. Based on this, the hypothesis of this work is that through multi-point simulations within a crop model framework, it is possible to detect the spatial variability of wheat grain yield and protein within a plot, as well as to compare measured versus simulated data.

Work expectations

The suitable candidate will be responsible for the entire parameterization of the crop model. This includes estimating the spatial variability of wheat grain yield and protein within a plot using a multi-point simulation approach and evaluating simulated spatial variability versus measurement data. Data assimilation using remote sensing data is also a goal; the candidate will have access to remote/proximal sensing data from the sites and will be expected to explore the use of such data within the crop model.

Required skills

A background in crop modeling (e.g., APSIM, DSSAT) and program skills (e.g., R, Python, C++) are essential. Knowledge of remote sensing as well as GIS and statistics is desired. Knowledge of maize and wheat physiology is a plus.