Local topographic variation directly influences crop yields in unirrigated agricultural fields. Topography-driven surface processes affect soil moisture and nutrient distributions throughout a field, thereby introducing spatial heterogeneity in soil fertility where crops are grown. This research used traditional and secondary terrain attributes derived from fine-resolution topographic surface data to model estimated crop yield for a row crop field in southern Ontario. A moving-window Pearson’s correlation technique was used to assess individual influence between topographic attributes and crop yield. Influential variables were then selected as input variables for a geographic weighted regression (GWR) model where crop yields were estimated and compared to observed values. Slope and relative topographic position were consistently selected as effective explanatory variables for the predictive GWR models regardless of crop type. Yields were sufficiently predicted for each crop type, with calculated coefficient of determination values equaling R2 = 0.80 for corn, R2 = 0.73 for wheat and R2 = 0.71 for soybeans. The model performed better in areas of the field with greater variation, suggesting that this method works best in variable terrain. These results indicate that local topographic variation plays a significant role in crop yield and various topographic attributes should be included in crop yield estimation models.