Behavior and biology of many forest pests are tied to major forest disturbances and succession. Fire is the principal disturbance in the forests of the western United States. Fire regimes as well as distribution and behavior of forest pests and beneficial microbes are all strongly associated with plant communities. Thus, mapping of plant communities can facilitate our ability to understand, predict, and manage interaction among these ecological processes. A new procedure to map potential vegetation was assessed on the Priest River Experimental Forest in northern Idaho. Digital terrain modeling and satellite imagery supplied 14 predictor variables to the Most Similar Neighbor (MSN)) imputation procedure. MSN uses a multivariate difference function and canonical correlation analysis to impute class designations to unclassified elements, based on the similarity of training samples across a set of global predictor variables. MSN was 'trained' with plant community data obtained from 245 delineated stands classified to 8 subseries (groups of habitat types). These subseries were imputed to 27,775 raster cells to create a map of subseries for the Priest River Experimental Forest. MSN statistics indicated a successful classification and predicted subseries map showed a good overall match with a habitat type map made from ground observation. Accurate mapping of subseries can facilitate analysis of landscape interactions among fire and pests. Such maps would support ecosystem restoration, management, and research. Needed landscape perspective could be added to the design, execution, and analysis of studies directed at understanding ecosystem processes such as gene flow and adaptation in hosts and pests.