A5074627752- Fire effects on ecosystems
- Landslides and related hazards
- Species Distribution and Climate Change
- Fire Detection and Safety Systems
- Remote Sensing in Agriculture
- Time Series Analysis and Forecasting
- Evacuation and Crowd Dynamics
- Data Management and Algorithms
- Cultural Heritage Management and Preservation
- Remote Sensing and LiDAR Applications
- Forest ecology and management
- Computer Graphics and Visualization Techniques
CIMA Research Foundation
2022-2024
Wildfires constitute an extremely serious social and environmental issue in the Mediterranean region, with impacts on human lives, infrastructures ecosystems. It is therefore important to produce susceptibility maps for wildfire management. The defined as a static probability of experiencing certain area, depending intrinsic characteristics territory. In this work, machine learning model based Random Forest Classifier algorithm employed obtain national scale Italy at 500 m spatial...
Background Wildfires are a growing threat to many ecosystems, bringing devastation human safety and health, infrastructure, the environment wildlife. Aims A thorough understanding of characteristics determining susceptibility an area wildfires is crucial prevention management activities. The work focused on case study 13 countries in eastern Mediterranean southern Black Sea basins. Methods data-driven approach was implemented where decade past linked geoclimatic anthropic descriptors via...
PROPAGATOR is a fire spread simulator designed as stochastic cellular automaton model for rapid risk assessment. The uses high-resolution data about topography and vegetation cover, accounting different types. Key inputs include wind, fuel moisture, the ignition point. Additionally, can incorporate firefighting strategies, such modifying moisture content or implementing firebreaks. probability of influenced by type, slope, content, while fire-propagation speed calculated using Rate Spread...
Wildfires are a critical component of natural ecosystems, contributing to biodiversity by shaping habitat structures and promoting species adaptation, but also posing significant risks human life, infrastructure, air quality. can be characterized both their impact the drivers occurrence. Historical data exploration is essential for researchers build data-driven models wildfire risk assessment capture characteristics extreme events (EWE). Such may include fire perimeter records, weather...
Changes in wildfire regimes observed globally due to land use transformation, human activity and climate change are compelling the development of Forest Fire Danger Rating systems capable accurately identifying spatio-temporal patterns increased fire danger for effective risk management, with a focus on distinguishing extreme dangerous conditions proper resources deployment.  Many existing models primarily rely weather conditions, often overlooking critical factors such as fuel...
Climate change has markedly increased the intensity and frequency of wildfires, emphasizing need for predictive tools to inform adaptive management mitigation strategies. This study presents a dynamic framework assessing wildfire susceptibility, focusing on Southeastern Europe, region particularly vulnerable due diverse topographical climatic conditions. By integrating machine learning (ML) with historical records climate projections, provides high-resolution susceptibility fuel maps...
Preserving cultural heritage from natural hazards is of paramount importance due to the role that plays in supporting community resilience and economic activities. Therefore, being able map risk faced by heritage, especially a multi-risk perspective, useful provide policy-making tool which highlights hotspot areas.In this work, we present preliminary version at national scale considering flood, earthquake, landslide wildfire hazards, Italy. The exposure dataset provided Italian Ministry...
Recent decades have seen an increase in wildfires activity, posing risks to human settlements, and forcing exploration of new technologies for wildfire risk management. Utilizing Machine Learning Time Series classification, this study produces decision support maps Civil Protection system Italy, which is responsible coordinating national firefighting air fleet. Trained on past events data, the model gives daily indication occurrence aerial requests each administrative unit utilizing time...
The socio-economic changes over recent decades, marked by rural abandonment and fuel accumulation, coupled with the impact of climate change altering spatio-temporal weather patterns, have created conditions conducive to potential extreme wildfire events. Numerous management systems thus faced significant challenges, leading an additional push develop or improve decision-support tools. Forest Fire Danger Rating models been widely used in aiding daily operations planning production fire...
In the southern European countries, combination of climate change, substantial shifts in land use/land cover, and socio-economic factors acting over last decades are anticipated to increase frequency, scale, intensity wildfires unless enhanced prevention control strategies implemented. Statistical data-driven approaches widely used by researchers evaluate main variables controlling occurrences spreading. Lately machine learning proved be highly performant due its flexible non-linear nature,...
The authors present a framework designed to model wildfire risk and the future projection of patterns, also in view climate change scenarios. adopted modeling is inherently multi scale, giving results at national after data gathering process developed regional / supranational scale. assessment comprises computation susceptibility, hazard, exposures, damage layers. Machine learning techniques are used assess susceptibility hazard level, analogously [1, 2]. To this end, two-models approach has...
Comparison of Different Algorithms and Vegetation Classes’ Importance Ranking in Wildfire Susceptibility Maps. Wildfire Maps (WSM) the analysis explanatory variables affecting model’s predictions are innovative tools to support forest protection management plans. Namely, WSM identify areas subject wildfire, terms relative spatial likelihood, on base observed past events, stored spatio-temporal inventories, local environmental anthropogenic properties an area....
A change in wildfire regimes several regions around the Earth has been acknowledged recent decades, with an increase frequency of particularly severe events. Consequently, many wildfires management systems have challenged, renewing interest Forest Fire Danger Rating (FFDR) models to support preparedness and response phases. The Liguria Region (Italy) Italian Civil Protection supported independent research programs that led 2003 development FFDR model RISICO. Nowadays is used as a...
<p>Wildfires are a serious social and environmental issue in the Mediterranean basin, menacing human lives, infrastructures ecosystems. Italy, due to land cover, orography  climate, expresses complex wildfire regime that is worth investigating. Static maps, such as susceptibility, hazard risk valid allies for management use planning. In particular, susceptibility defined spatially distributed probability of experiencing at certain point, depending only...
first_page settings Order Article Reprints Font Type: Arial Georgia Verdana Size: Aa Line Spacing: Column Width: Background: Open AccessAbstract Using Crossborder Multisource Burned Area Datasets for Assessing Wildfire Susceptibility Machine Learning Techniques † by Giorgio Meschi *, Andrea Trucchia, Guido Biondi and Paolo Fiorucci CIMA Research Foundation, 17100 Savona, Italy * Author to whom correspondence should be addressed. Presented at the Third International Conference on Fire...