Monitoring fire-affected areas using temporal profiles of spectral signal from time series MODIS satellite images

<p>Vegetation phenology is an important element of vegetation characteristics that can be useful in vegetation monitoring especially when satellite remote sensing observations are used. In that sense temporal profiles extracted from spectral signal of time series MODIS satellite images can be used to characterize vegetation phenology and thus to be helpful for monitoring vegetation recovery in fire-affected areas. The aim of this study is to explore the phenology patterns and the vegetation recovery pattern of various wildfires occurred in Greece during the period 2000-2020. Satellite remote sensing data from MODIS satellites in the period from 2000 to 2020 were acquired and processed to extract the temporal profiles of the spectral signal for selected areas inside and outside the fire-affected areas. This dataset and time period analyzed together with the time that these fires occurred gave the opportunity to create temporal profiles for some years before and some years after the fire. Different metrics linked to key phenological events have been created and used to assess vegetation recovery in the fire-affected areas. Apart of the use of the original spectral data we estimated and used vegetation indices commonly found in vegetation studies as well as in burned area mapping studies. In this study we explore the strength and the use of these time series satellite data to characterize vegetation phenology as an aid to assess the fire-affected areas and to monitor their vegetation recovery.</p>