Machine learning techniques have been increasingly used in flood management worldwide to enhance the effectiveness of traditional methods for flood susceptibility mapping. Although these models have achieved higher accuracy than traditional ones, their application has not yet reached full maturity. We focus on applying machine learning models to create flood susceptibility maps (FSMs) for Thessaly, Greece, a flood-prone region with extreme flood events recorded in recent years. This study utilizes 13 explanatory variables derived from topographical, hydrological, hydraulic, environmental and infrastructure data to train the models, using Storm Daniel—one of the most severe recent events in the region—as the primary reference for model training. The most significant of these variables were obtained from satellite data of the affected areas. Four machine learning algorithms were employed in the analysis, i.e., Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF) and eXtreme Gradient Boosting (XGBoost). Accuracy evaluation revealed that tree-based models (RF, XGBoost) outperformed other classifiers. Specifically, the RF model achieved Area Under the Curve (AUC) values of 96.9%, followed by XGBoost, SVM and LR, with 96.8%, 94.0% and 90.7%, respectively. A flood susceptibility map corresponding to a 1000-year return period rainfall scenario at 24 h scale was developed, aiming to support long-term flood risk assessment and planning. The analysis revealed that approximately 20% of the basin is highly prone to flooding. The results demonstrate the potential of machine learning in providing accurate and practical flood risk information to enhance flood management and support decision making for disaster preparedness in the region.

As cities have expanded into floodplains, the need for their protection has become crucial, prompting the evolution of flood studies. Here, we describe the operational tools, methods and processes used in flood risk engineering studies in the 1970s, and we evaluate the technological progress up to the present day. To this aim, we reference relevant regulations and legislation and the recorded experiences of engineers who performed hydrological, surveying and hydraulic studies in the 1970s. These are compared with the operational framework of a contemporary flood risk assessment study conducted in the Pikrodafni basin in the Attica region. We conclude that, without the technologically advanced tools available today, achieving the level of detail and accuracy in flood mapping that is now possible would have been unfeasible, even with significant human resources. However, ongoing urban development and growth continue to encroach upon flood plains that have existed for centuries, contributing to increased flood risk.

Nowadays, resilience has become an indispensable term in several aspects and areas of research and life. Reaching consensus on what actually constitutes “resilience,” “community,” and “community resilience” is still a task that guarantees a vivid exchange of opinions, sometimes escalating into debates, both in the scientific community and among practitioners. Figuring out how to practically apply resilience principles goes even a step further. This study attempts to circumvent the need for a universal agreement on the definition of “community resilience,” which may still be immature, if not impossible, at this time. We accomplish this by proposing a practical methodological approach with concrete methods on how to agree and implement commonly accepted community resilience principles in the context of technology development and pilot testing for disaster management. The proposed approach was developed, tested, and validated in the context of the Horizon 2020 EU-funded project Search and Rescue. Major aspects of the approach, along with considerations for further improvement and adaptation in different contexts, are addressed in the article.

Machine learning techniques have been increasingly used in flood management worldwide to enhance the effectiveness of traditional methods for flood susceptibility mapping. Although these models have achieved higher accuracy than traditional ones, their application has not yet reached full maturity. We focus on applying machine learning models to create flood susceptibility maps (FSMs) for Thessaly, Greece, a flood-prone region with extreme flood events recorded in recent years. The study utilizes 13 explanatory variables derived from topographical, hydrological, hydraulic, environmental and infrastructure data to train the models, using the Daniel storm—one of the most severe recent events in the region—as the primary reference for model training. The most significant of these variables were obtained from satellite data of the affected areas. Four machine learning algorithms were employed in the analysis, i.e., Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost). Accuracy evaluation revealed that tree-based models (RF, XGBoost) performed better, achieving Area Under the Curve (AUC) values of 97%. A flood susceptibility map corresponding to a 1000-year return period rainfall scenario at 24 h scale was developed, aiming to support long-term flood risk assessment and planning. The results demonstrate the potential of machine learning in providing accurate and practical flood risk information to enhance flood management and support decision-making for disaster preparedness in the region.