Classifying silvicultural systems (coppices vs. high forests) in Mediterranean oak forests by Airborne Laser Scanning data

Abstract

Forest classification by silvicultural systems (coppices vs. high forests) is important for forest resource assessment as such systems relate to a wide variety of ecosystem services. In this paper the potential of Airborne Laser Scanning (ALS) data for Mediterranean oak forests classification of coppices with standards vs. high forests was investigated in three study areas in Italy. We addressed the following issues: can coppices and high forests be distinguished using a raster Canopy Height Model (CHM)? Which are the most efficient CHM-derived metrics? Does the scale of analysis influence the classification potential of CHM metrics? Our results show that CHM in grid format (1-m² pixel) provides support information to classify silvicultural systems.

Keywords:

• ALS
• Canopy Height Model
• image segmentation
• classification