Land Suitability Assessments for Maize Production in Ontario, Canada, using a Weighted Overlay Method and Random Forest Algorithm

Abstract

This thesis aims to conduct land suitability assessments (LSAs) for maize production in Ontario, Canada for 2018 and 2080 under the Representative Concentrative Pathway (RCP) 4.5 scenario following the Food and Agriculture Organization of the United Nations (FAO) guidelines while considering climate, topography, and soil related factors. Two different approaches were used; namely a GIS-based weighted overlay technique, which is a conventional approach, and a novel statistical learning approach that employs the random forest (RF) algorithm with a new continuous measure of suitability introduced in this thesis. Both the conventional and statistical approach indicate that global warming will create more opportunities by 2080 for cultivating maize in Ontario with approximately 55% (546 000 km2) and 19% (183 000 km2) of Ontario’s land being suitable (highly and moderately suitable) for maize cultivation, respectively due to increasing growing season length, temperature, and precipitation. Regardless of the selected approach, there will be great economic significance involved with cultivating maize in Ontario by 2080 due to climate change while reducing food insecurity within Ontario and Canada.