A remote sensing-based approach was implemented to detect the effect of a late spring frost on beech forests in the Mediterranean mountain region. The analysis of spatio-temporal variability of frost effects on normalised difference vegetation index (NDVI) highlighted the distribution of the canopy damage across the forest according to geomorphic factors such as slope, aspect, and altitude.
Context Increased intensity and frequency of extreme temperatures such as late spring frosts and heat waves represent the main drivers affecting forest ecosystem structure and composition in the Mediterranean region.
Aims The main objective of this study was to evaluate the effects of a late spring frost disturbance, which occurred during spring 2016 in southern Italy, through the assessment of the spatial pattern of the damage to the beech forest canopy associated with the peak decrease in normalised difference vegetation index (NDVI), and the analysis of the NDVI temporal recovery after this frost disturbance.
Methods The forest areas affected by frost were detected through the NDVI differencing technique based on Landsat 8 (OLI/TIRS) imagery time series. The influence of local geomorphic factors (i.e., aspect, elevation, and slope) on forest NDVI patterns was assessed by means of a generalized additive model (GAM).
Results A rather counterintuitive NDVI patterns emerged according to the forest exposition, with NDVI significantly higher on the north facing areas than the southerly ones. The main canopy damage occurred at about 1250 m and reached up to 1500 m asl, representing the altitudinal range affected by the frost disturbance. Finally, the full canopy recovery occurred within 3 months of the frost event.
Conclusion The analysis of seasonal Landsat 8 image time series related to local geomorphic factors, such as aspect, slope, and altitude, and plant phenology on a frost event date, contributed to highlight the NDVI spatio-temporal variation and canopy recovery of a Mediterranean mountain beech forest.
Late spring frost, Beech forest, Landsat 8, Forest disturbance, NDVI differencing
Nolè, A., Rita, A., Ferrara, A.M.S. et al. Annals of Forest Science (2018) 75: 83.
For the read-only version of the full text: https://rdcu.be/4FzD
The dataset generated and analyzed during the current study is available in the Figshare repository, https://doi.org/30410.6084/m9.figshare.6409235.v1. (Nolè et al. 2018).