Analysis of radar and thermal satellite data time-series for understanding the long-term impact of land surface temperature changes on forests

Abstract

<p>Forests are globally an important environmental and ecological resource since they retrain water through their routes and therefore limit flooding events and soil erosion from moderate rainfall. They also act as carbon sinks, provide food, clean water and natural habitat for humans and other species, including threatened ones. Recent reports stressed the vulnerability of EU forest ecosystem to climate change impacts (EEA, 2012) (IPPC, et al., 2014). Climate change is a significant factor in the increasing forest fires and tree species being unable to adapt to the severity and frequency of drought during the summer period. Consequently, the possibility of increased insect pests and tree diseases is high as trees have been weakened by the extreme weather conditions. In Cyprus, there are two types of pine trees that exists on Troodos mountains, Pinus Nigra and Pinus Brutia, that may have been influenced by the reduced snowfall and extended summer droughts during the last decades.</p><p> </p><p>The overarching aim of this project is to research the impact of Land Surface Temperature on Cypriot forests on Troodos mountains by analysing time-series of radar and thermal satellite data. Impacts may include forest decline that does not relate to fire events, decreased forest density and alternations to timing of forest blooming initiation, duration and termination. Radar systems emitted pulses that can penetrate forest canopy due to the size of its wavelength and, therefore, collect information between tree branches without being affected by clouds. This presentation will focus on radar analysis conducted; testing of various methods, and how the processing pipeline has been automated.</p><p> </p><p>The project ‘ASTARTE’ (EXCELLENCE/0918/0341) is co-financed by the European Regional Development Fund and the Republic of Cyprus through the Research Innovation Foundation.</p>

Geoinformatics and Machine Learning for Comprehensive Fire Risk Assessment and Management in Peri-Urban Environments: A Building-Block-Level Approach

“…), rendering fire risk assessment a critical priority [2]. In particular, climate change has increased the vulnerability of forest ecosystems, as it is a major contributor to the rise in forest fires and tree species' inability to adapt to the intensity and frequency of summer droughts [3]. In light of this, peri-urban zones are more prone to wildfires, leaving people's lives, properties, and the natural environment/ecosystem exposed to increased disaster risk.…”

High-Resolution Fire Risk Assessment and Management Planning in Peri-urban Areas via Coupling Geoinformatics, Machine-learning and Field Observations. Pilot Application in Attica Region, Greece.

“…Climate change has led to increases in the vulnerability of forest ecosystems, as it is a major contributor to the rise of forest fires and tree species' inability to adapt to the intensity and frequency of summer droughts (Prodromou et al, 2020). Peri-urban zones, mainly due to uncontrolled urban-sprawl and lack of proper planning, are more vulnerable to wildfires leaving people's lives and properties, as well as the surrounding natural environment and ecosystem, exposed to increased disaster risk.…”