This paper presents a review of concepts related to wildfire risk assessment, including the determination of fire ignition and propagation (fire danger), the extent to which fire may spatially overlap with valued assets (exposure), and the potential losses and resilience to those losses (vulnerability). This is followed by a brief discussion of how these concepts can be integrated and connected to mitigation and adaptation efforts. We then review operational fire risk systems in place in various parts of the world. Finally, we propose an integrated fire risk system being developed under the FirEUrisk European project, as an example of how the different risk components (including danger, exposure and vulnerability) can be generated and combined into synthetic risk indices to provide a more comprehensive wildfire risk assessment, but also to consider where and on what variables reduction efforts should be stressed and to envisage policies to be better adapted to future fire regimes. Climate and socio-economic changes entail that wildfires are becoming even more a critical environmental hazard; extreme fires are observed in many areas of the world that regularly experience fire, yet fire activity is also increasing in areas where wildfires were previously rare. To mitigate the negative impacts of fire, those responsible for managing risk must leverage the information available through the risk assessment process, along with an improved understanding on how the various components of risk can be targeted to improve and optimize the many strategies for mitigation and adaptation to an increasing fire risk.

Among the numerous natural hazards, landslides are one of the greatest, as they can cause enormous loss of life and property, and affect the natural ecosystem and their services. Landslides are disasters that cause damage to anthropic activities and innumerable loss of human life, globally. The landslide risk assessed by the integration of susceptibility and vulnerability maps has recently become a manner of studying sites prone to landslide events and managing these regions well. Developing countries, where the impact of landslides is frequent, need risk assessment tools that enable them to address these disasters, starting with their prevention, with free spatial data and appropriate models. Our study shows a heuristic risk model by integrating a susceptibility map made by AutoML and a vulnerability one that is made considering ecological vulnerability and socio-economic vulnerability. The input data used in the State of Guerrero (México) approach uses spatial data, such as remote sensing, or official Mexican databases. This aspect makes this work adaptable to other parts of the world because the cost is low, and the frequency adaptation is high. Our results show a great difference between the distribution of vulnerability and susceptibility zones in the study area, and even between the socio-economic and ecological vulnerabilities. For instance, the highest ecological vulnerability is in the mountainous zone in Guerrero, and the highest socio-economic vulnerability values are found around settlements and roads. Therefore, the final risk assessment map is an integrated index that considers susceptibility and vulnerability and would be a good first attempt to challenge landslide disasters.

The risks associated with landslides are increasing the personal losses and material damages in more and more areas of the world. These natural disasters are related to geological and extreme meteorological phenomena (e.g., earthquakes, hurricanes) occurring in regions that have already suffered similar previous natural catastrophes. Therefore, to effectively mitigate the landslide risks, new methodologies must better identify and understand all these landslide hazards through proper management. Within these methodologies, those based on assessing the landslide susceptibility increase the predictability of the areas where one of these disasters is most likely to occur. In the last years, much research has used machine learning algorithms to assess susceptibility using different sources of information, such as remote sensing data, spatial databases, or geological catalogues. This study presents the first attempt to develop a methodology based on an automatic machine learning (AutoML) framework. These frameworks are intended to facilitate the development of machine learning models, with the aim to enable researchers focus on data analysis. The area to test/validate this study is the center and southern region of Guerrero (Mexico), where we compare the performance of 16 machine learning algorithms. The best result achieved is the extra trees with an area under the curve (AUC) of 0.983. This methodology yields better results than other similar methods because using an AutoML framework allows to focus on the treatment of the data, to better understand input variables and to acquire greater knowledge about the processes involved in the landslides.

Fire is a natural phenomenon that has played a critical role in transforming the environment and maintaining biodiversity at a global scale. However, the plants in some habitats have not developed strategies for recovery from fire or have not adapted to the changes taking place in their fire regimes. Maps showing ecological vulnerability to fires could contribute to environmental management policies in the face of global change scenarios. The main objective of this study is to assess and map ecological vulnerability to fires on a global scale. To this end, we created ecological value and post-fire regeneration delay indices on the basis of existing global databases. Two ecological value indices were identified: biological distinction and conservation status. For the post-fire regeneration delay index, various factors were taken into account, including the type of fire regime, the increase in the frequency and intensity of forest fires, and the potential soil erosion they can cause. These indices were combined by means of a qualitative cross-tabulation to create a new index evaluating ecological vulnerability to fire. The results showed that global ecological value could be reduced by as much as 50 % due to fire perturbation of poorly adapted ecosystems. The terrestrial biomes most affected are the tropical and subtropical moist broadleaf forest, tundra, mangroves, tropical and subtropical coniferous forests, and tropical and subtropical dry broadleaf forests.

Background: Wildfires play a significant and complex role in ecosystems, influencing various aspects of their functioning and structure. These natural disturbances can positively and negatively impact ecosystems, shaping landscapes, nutrient cycles, biodiversity, and ecological processes. This study focuses on assessing and integrating the different factors that affect the ecological vulnerability to wildfires at the European scale. Our methodology follows three steps. Firstly, ecological values based on biological distinctiveness and conservation status were estimated to understand pre-fire conditions better. Secondly, we obtain vegetation’s coping capacity (or resistance) to the impacts of fire, considering the functional traits of plants and fire characteristics through a fire extreme scenario. Finally, post-fire recovery time was calculated by considering the species-specific recovery time, recovery starting time, growth recovery rate, and the environmental constraints affecting the optimal vegetation response. These three variables were combined using a dynamic model that assumed the change of value due to wildfires integrated throughout the recovery time.Results: Our results indicate that the tundra biome emerges as the most ecologically vulnerable to fire, primarily due to its high ecological values and long recovery time, which outweigh its moderate coping capacity. Following closely, the temperate conifer forests also exhibit high vulnerability driven by their high recovery time, despite moderate ecological and coping capacity values. The boreal forests rank next, with moderate vulnerability due to their long recovery time and moderate coping capacity. The Mediterranean region, although having moderate ecological values and recovery time, shows a notable vulnerability influenced by lower coping capacity. The temperate broadleaf and mixed forests demonstrate relatively lower vulnerability owing to their balanced ecological values, moderate recovery time, and substantial coping capacity. Lastly, the temperate grasslands, savannas, and shrublands are the least vulnerable, benefiting from lower ecological values and the fastest recovery time, alongside moderate coping capacity, which collectively reduce their overall fire vulnerability.Furthermore, we found that coping capacity is the factor that most influenced ecological vulnerability to wildfires.Conclusions: The study identifies key zones for European or national policies on fire prevention and post-wildfire regeneration. It offers insights into effective forest management and conservation policies, applicable to current conditions. Additionally, the methods can predict future ecological vulnerability to wildfires based on climatic and socio-economic trends.

Landslides are recognized as high-impact natural hazards in different regions around the world; therefore, they are extensively researched by experts. Landslide inventories are essential to identify areas that are likely to be affected in the future, thereby enabling interventions to prevent loss of life. Today, through combined approaches, such as remote sensing and machine learning techniques, it is possible to apply algorithms that use data derived from satellite images to produce landslide inventories. This work presents the performance of five machine learning methods—k-nearest neighbor (KNN), stochastic gradient descendent (SGD), support vector machine radial basis function (SVM RBF Kernel), support vector machine (SVM linear kernel), and AdaBoost—in landslide detection in a zone of the state of Guerrero in southern Mexico, using continuous change maps and primary landslide factors, such as slope angle, terrain orientation (aspect), and lithology, as inputs. The models were trained with 2/3 of ground truth samples of 671 slidden/non-slidden polygons. The obtained inventory maps were evaluated with the remaining 1/3 of ground truth samples by generating a confusion matrix and applying the Kappa concordance coefficient, accuracy, precision, recall, and F1 score as evaluation metrics, as well as omission and commission errors. According to the results, the AdaBoost classifier reached greater spatial and statistical coherence than the other implemented methods. The best input layer combination for detection was the continuous change maps obtained by the linear regression and image differencing detection methods, together with the slope angle, aspect, and lithology conditioning factors.