Culminating Project Title
Demonstrating Spatial Patterns of Crop Productivity in a Minnesota Corn Field Using Hierarchical Multiple Regression Models and Ordination
Date of Award
Culminating Project Type
Geography - Geographic Information Science: M.S.
Geography and Planning
School of Public Affairs
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
Crop Productivity, Remote Sensing, Soil Zonation, Hierarchical Multiple Regression, Ordination, Vegetation Indices
Accurate and timely assessment of within-field crop vigor heterogeneity is essential for detecting field-wide crop productivity and yield, contributing to improvements in the management of corn fields. Yet, studies designed to explore the spatial heterogeneity of crop vigor in corn over different productivity zones, where soil nutrient characteristics are known to limit crop productivity during the growing season, as yet been reported. We assessed whether changes in temporal weather conditions within a growing season, contribute to crop vigor variability. Furthermore, we evaluated whether within-season changes in precipitation and temperature contribute to variable nutrient concentrations within different productivity zones. More so, we utilized random forest regression to calculate the relative importance of predictor variables to crop vigor variability. We then employed hierarchical multiple regression (HMR) to build several regression models to determine whether the collinearity of variables (soil characteristics) showed a significant improvement in the R2 i.e., the proportion of explained variance in crop vigor response. The principal component analysis (PCA) was employed to find components that express as much of the inherent variability of the complete data set as possible as well as, to plot how variables map relative to field productivity or management zones. We inferred that, changes in precipitation and temperature during the growing season influence soil nutrient concentrations within productivity zones especially, potassium, calcium, nitrogen, phosphorus, and magnesium. We hypothesize that, significant and yet subtle crop vigor differences can be observed within field productivity zones attributed to the heterogeneity of soil macro nutrient concentrations within corn fields, using the combined utility of remote sensing and hybrid statistical approaches. Thus, aiding farmers to ascertain, early season, whether they will obtain a poor harvest or not, improve on soil nutrient use efficiency, and field management practices to ensure a bumper harvest.
Mpofu, Ndumezulu, "Demonstrating Spatial Patterns of Crop Productivity in a Minnesota Corn Field Using Hierarchical Multiple Regression Models and Ordination" (2021). Culminating Projects in Geography and Planning. 12.