The subalpine zone is the transition between forest and alpine vegetation communities. In Norway, as in many other nations, low productivity or non-merchantable forests, like the subalpine zone, are not routinely subject to inventory programs. Awareness of expected changes in the sub-alpine zone as a result of a warmer climate, and the interest in full carbon accounting at the national level, has dictated a need for data capture in these mountainous areas. We propose an approach for integrating strip samples of Light Detection and Ranging (LiDAR) data with Landsat imagery to delineate the subalpine zone. In the current study the subalpine zone was defined according to international definitions based on tree heights and canopy cover. The three-dimensional measurements of forest structure obtained from LiDAR enable a delineation of the subalpine zone. The approach was implemented using 53 LiDAR sample strips in Hedmark County, Norway, and validated with field measurements at 26 locations. The subalpine zone boundaries were found to be within one Landsat pixel, on average, when validated using an image gradient technique. Furthermore, binomial logistic regression was used to upscale the LiDAR classes to the entire county (27 400 km2) using satellite images and information derived from a digital terrain model. The result from the binomial logistic regression was a probability map suitable for monitoring changes in the extent and location of the subalpine zone. The probability surface was separated into hard classes by calibrated alpha-cuts derived using density estimation to support the information needs of inventory stratification and area estimation.