Quantifying spatial and temporal vegetation recovery dynamics following a wildfire event in a Mediterranean landscape using EO data and GIS

Analysis of Earth Observation (EO) data, often combined with Geographical Information Systems (GIS), allows monitoring of changing land cover dynamics which may occur after a natural hazard such as a wildfire. In the present study, the vegetation recovery dynamics of one such area are evaluated by exploiting freely distributed EO data and GIS techniques. The relationships of re-growth dynamics to the topographical characteristics of the burn scar are also explored. As a case study, a typical Mediterranean ecosystem in which a wildfire occurred during 2007 is used. Vegetation recovery dynamics of the whole area under the burn scar were investigated based on chronosequence analysis of the normalized difference vegetation index (NDVI) derived from anniversary Landsat TM images. The spatio-temporal patterns of post-fire NDVI on each image date were statistically compared to the pre-fire pattern to determine the extent to which the pre-fire spatial pattern was re-established and the recovery rate. The relationships between NDVI as an expression of recovery rates and aspect were also statistically investigated and quantified using a series of statistical metrics. Results suggested a generally low to moderate vegetation recovery of the local ecosystem five years after the fire event, with the post-fire NDVI spatial pattern generally showing a gradual but systematic return to pre-fire conditions. Re-growth rates appeared to be somewhat higher in north-facing slopes in comparison to south facing ones, in common with other similar studies in Mediterranean type ecosystems. All in all, this study provides an important contribution to the understanding of Mediterranean landscape dynamics, and corroborates the usefulness particularly of NDVI in post-fire regeneration assessment via a well-established methodology which can also be transferable to other regions. It also provides further evidence that use of EO technology which combined with GIS techniques can offer an effective practical tool for mapping wildfire vegetation dynamics and ecosystem recovery after wildfire. (C) 2014 Elsevier Ltd. All rights reserved.