Study of Shrub Cover and Height Using LIDAR Data in a Mediterranean Area

Abstract

Abstract In this work we studied the height and coverage of shrub vegetation using light detection and ranging (LIDAR) data. The maximum dominant heights of vegetation were measured in the field in 83 stands of a 0.5-m radius, and the data were compared with figures for heights obtained from LIDAR data in concentric areas with different radii. The minimum root mean square error (RMSE) between the field measurements and LIDAR data was found for radii between 1.5 and 2.25 m, RMSE being 0.26 m. When the slopes are low and an accurate digital terrain model is obtained, it was shown that the radius can be reduced. Shrub heights were also studied in plots of 100 m2. In this case, the 95th percentile of the LIDAR data included in each plot was the best predictor of height with R2 of 0.71 and a RMSE of 0.13 m. For detecting the presence of shrub vegetation, the highest accuracy was obtained when the canopy height model and a spectral image were combined (overall accuracy of 90%).