Elena Palenova defends her thesis "We Will Burn: What's Next?: Wildfire Risk, Recovery, and Geospatial Decision-Making in Greece".

Elena Palenova

Doctoral thesis: We Will Burn: What's Next?: Wildfire Risk, Recovery, and Geospatial Decision-Making in Greece Länk till annan webbplats, öppnas i nytt fönster.

Subject: Environmental Science
Research area: Environmental Studies
Graduate school: Natural Sciences, Technology and Environmental Studies
Faculty Examiner: Dr. Gavriil Xanthopoulos, Forest Fires, Hellenic Agricultural Organization "Demeter", Institute of Mediterranean Forest Ecosystems, Athens, Greece
Language: English

Abstract

Wildfires are intensifying globally, demanding interdisciplinary approaches that integrate both the ecological and social dimensions of fire research. This thesis addressed this need through a holistic framework that examines wildfire impacts and post-fire recovery in Greece, a region highly prone to recurrent wildfires. The aim was to advance understanding of post-fire response, vegetation recovery and restoration, and wildfire risk through an Integrated Fire Management lens, using socio-ecological and geospatial perspectives.

Stakeholder views on post-fire management were explored through Q-methodology that combines qualitative and quantitative techniques to examine subjectivity. Semi-structured, in-depth interviews with 22 participants involved in wildfire management spanning 10 affiliations were conducted. From these interviews, 376 statements were extracted, of which 44 were selected for a Q study. Six distinct discourses were identified: Statist, Abstainer, Beneficiary, Pragmatist, Revolutionary, and Contextualist, revealing contrasting but complementary perspectives and highlighting communication gaps between institutions and actors. Conducted interviews were used to develop a framework for prioritizing areas for post-fire restoration combining stakeholder input with geospatial analysis. Areas covering 77 km² were selected for restoration depending on the fire history, slope, and designation as part of the protected areas. Three locations with a total area of 31 km² were highlighted as potentially needing special attention.

Post-fire vegetation recovery was examined using spatial analysis, statistical methods, and machine learning. Satellite data from MODIS and Landsat over 2013–2025 were analysed through a chrono-sequence approach employing NDVI, EVI, and dNBR vegetation indices. Random Forest algorithm was applied for downscaling climate parameters. By combining spectral indices, topographic variables, and climate data, predictive Generalized Additive Mixed Models were developed to evaluate regeneration patterns in fire-prone areas. Results indicated that recovery correlates positively with precipitation and evapotranspiration and negatively with temperature and elevation. A related research direction emerged around the issue of power line ignitions as a wildfire source. Analysis of official data from the Hellenic Fire Service and records from WWF Greece, media sources and volunteers, along with application of Sentinel-2 satellite imagery for validating fire scars revealed systemic data limitations and proved that a total of 237 km² of burned area of selected case studies from 2018 to 2025 were directly linked to electrical causes.

The findings showed the need for developing comprehensive mixed-method approaches and the necessity for improved transparency and infrastructure maintenance to mitigate wildfire risks. Interdisciplinary design and the outcomes of this thesis contribute to the evolving field of Integrated Fire Management by comprehensive understanding of wildfire dynamics and offering evidence-based insights to guide more adaptive and inclusive wildfire governance in the Mediterranean.