1211 | 554 | time series with vegetation growth index for mediterranean forest classification

Forest masses are constantly changing due to man’s action and climate change. The study of the biodiversity, the monitoring of the species’state and the distribution of the vegetation in the territory becomes an increasingly necessary field of research to study the impacts and the effects of these changes._x000D_
Currently, the increase in satellite missions focused on the study of the Earth is vertiginous. It is worth to point out the Sentinel missions within the European Space Observation Programme of Copernicus (ESA) and for this topic, the Sentinel-2 mission with a constellation of two satellites taking images of our territory every 5 days with the visible and near infrared spectrum. _x000D_
In this way, a new index based on percentage variations has been proposed among the four spectral bands that include the visible portion of green and red and two rededge bands. We have called this new index as Vegetation Growth Index (VGI). With this index, we have carried out a three-year temporary study, between 2018 and 2020, where temporary VGI patterns have been determined for the predominant tree species of the studied forested masses with dendrometric monitoring for validation in Montseny Natural Park (Catalonia, SPAIN). These spectral time patterns follow a particular phenological behaviour during the year. When the sequence of the vegetative cycle is not met with the VGI, it most likely implies a change indicator derived from forest management, biotic or abiotic diseases, cover land replacement, etc._x000D_
Therefore, this model of temporary approximation to spectral information provides us with a new tool for classifying and detecting changes in forest masses. Opens new working lines to exploit current and future satellite data to better understand our Mediterranean forests and with sufficient field information, it can be applied to all forest masses._x000D_

Anna Tardà
Institut Cartogràfic i Geològic de Catalunya

ID Abstract: 554

You might also like