Forest structure modeling with combined airborne hyperspectral and LiDAR data

► Here LiDAR and hyperspectral data are combined to build models of forest attributes. ► Genetic algorithm was used to screen the predictor feature space. ► Restricted neighbor search improved the performance and returns unbiased models. ► The hyperspectral data slightly contributed to improve the LiDAR-based models. ► GA-screened models featuring 9-12 predictors are suggested.