In this study we combine information from aerial LIDAR and hemispherical
images taken in the field with ForeStereo—a forest inventory device—to assess the
vulnerability and to design conservation strategies for endangered Mediterranean fir
forests based on the mapping of fire risk and canopy structure spatial variability.
We focused on the largest continuous remnant population of the endangered tree
species Abies pinsapo Boiss. spanning 252 ha in Sierra de las Nieves National Park
(South Andalusia, Spain). We established 49 sampling plots over the study area.
Stand structure variables were derived from ForeStereo device, a proximal sensing
technology for tree diameter, height and crown dimensions and stand crown cover
and basal area retrieval from stereoscopic hemispherical images photogrammetry.
With this information, we developed regression models with airborne LIDAR data
(spatial resolution of 0.5 points∙m−2). Thereafter, six fuel models were fitted to the
plots according to the UCO40 classification criteria, and then the entire area was
classified using the Nearest Neighbor algorithm on Sentinel imagery (overall
accuracy of 0.56 and a KIA-Kappa Coefficient of 0.46). FlamMap software was used
for fire simulation scenarios based on fuel models, stand structure, and terrain data.
Besides the fire simulation, we analyzed canopy structure to assess the status and
vulnerability of this fir population. The assessment shows a secondary growth forest
that has an increasing presence of fuel models with the potential for high fire spread
rate fire and burn probability. Our methodological approach has the potential to be
integrated as a support tool for the adaptive management and conservation of
A. pinsapo across its whole distribution area (<4,000 ha), as well as for other
endangered circum-Mediterranean fir forests, as A. numidica de Lannoy and
A. pinsapo marocana Trab. in North Africa.
