We investigate residential property-price effects of the spread of the Hemlock wooly adelgid infestation northward through central portions of Connecticut and Massachusetts, USA. We find that hemlock trees and the accompanying adelgid infestation within 0.1 km buffers of properties affect sale prices, but the results do not extend to buffers of 0.5 and 1.0 km's. Further, within the 0.1 km buffer, only the healthiest hemlock trees contribute positively to property values. We investigated the robustness of the results to three data interpolation methods, Kriging, Inverse Distance Weighting and Spline, and while there was some minor difference in outcomes the results are robust to these interpolation methods. Two property-price models were estimated, a traditional hedonic model with spatial fixed effects and a repeat sale model. The models provide substantially different property-price impacts and care needs to be taken when interpreting these estimates. Both approaches are limited but in different ways; the hedonic by potentially omitted variables and the repeat-sales by a limited number of observations. Our results provide some support for the repeat-sale model as the hedonic model with spatial fixed effects underperformed when both models were estimated using the same data.