Evaluation of Maximum Entropy (Maxent) Machine Learning Model to Assess Relationships between Climate and Corn Suitability

Given the impact that climate change is projected to have on agriculture, it essential understand mechanisms and conditions drive agricultural land suitability. However, existing literature does not provide sufficient guidance best modeling methodology study crop suitability, there even less research how evaluate accuracy of such models. Further, studies yet demonstrate use Maximum Entropy (Maxent) model in predicting presence yield large-scale field crops United States. In this study, we investigate application Maxent predict suitability present novel methods evaluating its predictive ability. a correlative machine learning often used cropland for corn production contiguous States under current bioclimatic conditions. We developed Maxent’s ability through three comparisons: (i) classification suitable units comparison results with another similar species distribution (Random Forest Classification), (ii) output response curves thresholds, (iii) correlation predicted observed extent yield. determined was superior Random Forest, especially areas which likely but currently associated presence. also predictions correlated strongly statistics were consistent regarding range variable values corn. concluded an effective method could be applied broader issues agriculture–climate relationships.