Analysing gap dynamics in forest canopies with landscape metrics based on multi-temporal airborne laser scanning surveys – A pilot study

Abstract

For a long time gaps or openings in the forest canopy have been of considerable interest to forest ecologists and to forest managers. In the context of disturbances induced by climate change, canopy gap dynamics are of particular interest, since they can indicate imminent damage to forest resources and irreversible trends such as forest decline. Here, statistical significance is crucial for establishing whether any imminent large-scale threat to the sustainability of forest resources exists. In order to be able to assess significance, we applied the Boolean model, a null or reference model from random set statistics. The Boolean model served as a theoretical benchmark for testing the significance of the observed trends in forest canopy gap dynamics. As a pilot study we analysed airborne laser scan (ALS) data collected in the Krycklan catchment area (Northern Sweden) in 2006 and 2015. The data were analysed using eight different landscape metrics. Despite the moderate resolution of our ALS data the landscape metrics have proved to be useful tools for monitoring canopy gap dynamics of forest ecosystems. The Boolean model has been successful in ascertaining statistical significance and the model parameters indicated important trends. In the Krycklan catchment area, there was no significant trend of canopy gap dynamics indicating any harmful development between 2006 and 2015. On the contrary, we found evidence for gaps closing in and gap locations becoming more random whilst the canopy cover increased between the two survey years.