The correlation matrix reveals that human-controlled factors, specifically Seed Variety ($0.68$) and Irrigation Schedule ($0.56$), have the strongest positive relationship with crop yield. Interestingly, rainfall shows a slight negative correlation ($-0.25$), suggesting that in this dataset, more water isn't always better and that managed irrigation is a much more reliable predictor of success.

This KDE plot demonstrates a significant overlap in yield efficiency between "Sustainable" and "High-Risk" areas. It proves that high productivity isn't exclusive to resource-rich land; through intensive management and fertilizer use, high-risk areas can achieve yields comparable to sustainable ones, albeit at a higher environmental or resource cost.

The Logistic Regression model was designed to prioritize catching as many sustainable zones as possible. While the matrix shows some false positives—classifying high-risk areas as sustainable—it successfully identified 69% of truly sustainable locations. This "safety-first" approach is vital for agricultural planning where missing a sustainable opportunity is costlier than investigating a false lead.

Comparing the drivers of yield with the drivers of sustainability highlights a critical gap: high production does not automatically translate into environmental health. While seed variety and irrigation are the "engines" of high yield, they don't necessarily align with the natural soil and water availability that define a sustainable zone. This visual emphasizes the importance of choosing management strategies that favor both output and resource longevity.