Keiko Ioki

Abstract Introduction Enrichment planting is a widely used method of forest restoration that involves planting seedlings from native species in degraded forests. Variability of the micro‐environmental conditions within the target area largely affects the growth of the seedlings; therefore, area‐wide and high‐resolution data that reflect habitat quality are vital to the success of such projects. Light Detection and Ranging (LiDAR) data provide opportunities to characterize the three‐dimensional spatial heterogeneity of forests over broad spatial extents. Objectives We assessed the use of airborne LiDAR data in the evaluation of site suitability for the enrichment planting of dipterocarp trees in Malaysian Borneo. Methods Predictor variables derived from the LiDAR data were used to represent the topographical and forest‐structure information and to predict the growth of 10 planted dipterocarp species. We fitted field‐measured diameter growth data collected from the trees 15–20 years after planting and LiDAR predictors from Random Forest regression models. Results Prediction models were developed for eight species with r² values of 0.06–0.43, while models for the other two species were unable to provide reasonable fits. Selected significant predictors in the models agreed well with the known traits and habitat preferences. For the four species with effective prediction models, we generated growth prediction maps to illustrate their growth potential. Conclusions This study highlights the utility of micro‐habitat information derived from airborne LiDAR data in forest restoration. The prediction maps generated here could contribute to enrichment planting guidelines in broad‐scale restoration schemes.