Using LiDAR and forest inventory data to identify late development stages in broad-leaved forest

Palop-Navarro, E., Bañuelos, M. J., Quevedo, M. 2016. Using LiDAR and forest inventory data to identify late development stages in broad-leaved forest. Ecosistemas 25(3): 35-42. Doi.: 10.7818/ECOS.2016.25-3.04 Old-growth forests are generally scarce because of historic human exploitation of forest ecosystems. Their structure – characterized by trees of varying sizes and ages, presence of forest gaps and abundant deadwood – is particularly hard to recover after homogenization derived from forest exploitation. Moreover, the conservation of forest specialists, like saproxylic insects or woodpeckers, depends on this type of structure. Characterizing and identifying this structure is therefore important for conserving and restoring this type of forests and their communities. Forest structure and composition has been historically recorded in forest inventories. These inventories provide valuable information, albeit of discrete nature, and may be less efficient in areas of difficult access. From this regard, LiDAR (Light Detection and Ranging) technology provides tridimensional and continuous information of forest structure, making it a useful tool for complementing field work of forest inventories. In this study we evaluated the possibility of quantifying late successional stages in broad-leaved forests in the NW corner of the Iberian Peninsula from public LiDAR data (density 0.5 points m-2). First, we compared LiDAR-derived variables with field-derived variables from the 4th National Forest Inventory (IFN4), which showed the existence of correlations, especially at maximum and averaged heights, though the adjustment varies considerably with the diameter of the inventory considered. Later we built simple predictive models for the identification of old-growth forests in IFN4 plots (binary variable Bosque Viejo). The final model was built by variables namely Altura desvest and Altura media2 (indicators of heterogeneous structures and size varying trees in late successional stages of development), together with dominance of different tree species in plots derived from the IFN4. Results show that publically available Spanish LiDAR data can be used to identify forest structures compatible with old-growth broadleaved forests.