Severidad y regeneración en grandes incendios forestales: análisis a partir de series temporales de imágenes Landsat

[EN] The main objective of this study is to take a close look at post-fire recovery patterns in forestry areas under different burn severity conditions. We also investigate the time that forestry ecosystems take to recover their pre-fire condition. In this context, this study analyses both the level...

ver descrição completa

Detalhes bibliográficos
Autores: Martínez, S., Chuvieco, E., Aguado, I., Salas, Jorge
Tipo de documento: artigo
Data de publicação:2017
País:España
Recursos:Universitat Politècnica de València (UPV)
Repositório:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:espanhol
OAI Identifier:oai:riunet.upv.es:10251/92734
Acesso em linha:https://riunet.upv.es/handle/10251/92734
Access Level:Acceso aberto
Palavra-chave:Incendios forestales
GeoCBI
Regeneración
Severidad
LandTrendr
Landsat
Wildland fires
Recovery
Burn severity
Descrição
Resumo:[EN] The main objective of this study is to take a close look at post-fire recovery patterns in forestry areas under different burn severity conditions. We also investigate the time that forestry ecosystems take to recover their pre-fire condition. In this context, this study analyses both the level of severity in Uncastillo forest wildfire (7.664ha), one of the greatest occurred in Spain in 1994, and the pattern of natural recovery in the following decades (until 2014) using annual Landsat time series (sensors TM&ETM+). Burn severity has been estimated by means of PROSPECT and GeoSAIL radiative transfer models following methodologies described in De Santis and Chuvieco (2009). On the other hand, recovery processes have been assessed from spectral profiles using the LandTrendr model (Landsat-based Detection of Trends in Disturbance and Recovery) (Kennedy et al., 2010). Results contribute to a further understanding of the post-fire evolution in forestry areas and to develop effective strategies for sustainable forest management.