Maria Katherina Dal Barco
On the occasion of International Youth Day, we spoke to Maria Katherina Dal Barco about her research on using Machine Learning for climate risk assessment in the Veneto coastal area.
How does your two-step Machine Learning approach improve the understanding and management of multi-risk events in coastal areas like Veneto?
The two-step Machine Learning approach enhances the understanding and management of multi-risk events in coastal areas like Veneto by effectively addressing two critical aspects. Firstly, it identifies the main drivers of the impacts recorded by the Emergency archive of the Veneto Region between 2009-2019, allowing for a clear understanding of the factors that contribute most significantly to these events. Secondly, it estimates future risks under various Representative Concentration Pathway (RCP) scenarios, a comprehensive view of potential future outcomes. Additionally, a sensitivity analysis was conducted to elucidate the impact of different combinations of the selected hazards (i.e., precipitation, sea-level, and wind speed) on the overall risk. This thorough approach ensures a more informed and proactive risk management strategy for coastal regions.
What are the most significant findings from your study on the cumulative impacts of extreme climate events in the Veneto coastal area?
The study presented at the 3rd International Conference on Natural Hazards and Risks in a Changing World concerns the cumulative impacts of extreme climate events in the Veneto coastal area, which yielded several significant findings. By incorporating exposure and vulnerability patterns – such as geomorphology, land-use, soil permeability, urbanisation levels, and surface characteristics of different municipalities – it was possible to better detect the main drivers of risk. To better analyse the coastal municipalities of the Veneto region, these have been divided into four groups, each sharing similar features in terms of orientation and geomorphological characteristics (e.g., river valley, lagoon, river delta).
A key discovery is that vulnerability and exposure factors can significantly intensify the effects of extreme weather events. For instance, the southern coastal areas (i.e., the province of Rovigo), characterised by larger natural areas, have experienced fewer impacts. Conversely, more urbanised and densely populated municipalities, such as Venice, Jesolo, and Chioggia, are the most affected. This differentiation underscores the importance of localised analysis and tailored risk management strategies.
How can your integrated approach, which considers hazard, exposure, and vulnerability, improve the accuracy and effectiveness of climate impact assessments?
Our integrated approach can significantly improve the accuracy and effectiveness of climate impact assessments by analysing the vulnerability and exposure characteristics of each group, hence understanding how these factors amplify or mitigate the occurrence of impacts. This detailed analysis enables the identification of areas at higher risk and develop more precise and effective mitigation strategies.
What were the key insights or lessons you gained from attending the 3rd International Conference on Natural Hazards and Risks?
Attending the 3rd International Conference on Natural Hazards and Risks provided numerous key insights and lessons. By participating in various sessions, I had the opportunity to enrich my knowledge through the ideas and applications shared by fellow colleagues. This exchange of information was incredibly valuable, as it exposed me to new methodologies and tools that could be potentially implemented in my own case study of the coastal area of the Veneto region.
Additionally, the conference allowed me to reconnect with colleagues I had met during previous conferences, workshops, and visiting periods abroad, as well as to establish new professional relationships. These interactions fostered a collaborative environment where we could discuss and explore innovative approaches to managing and mitigating natural hazards.
Civil protection agencies can better utilise our research by fostering continuous collaboration with the scientific community. This partnership will ensure that agencies are up-to-date with cutting-edge methodologies, results, and potential applications.
How has your participation in the conference enriched your approach to using Machine Learning for climate risk analysis and adaptation planning?
My participation in the conference significantly enriched my approach to using Machine Learning for climate risk analysis and adaptation planning. Exposure to cutting-edge research and practical applications demonstrated how Machine Learning could be further optimised to detect main drivers of risk and estimate future risks under various scenarios. The discussions and presentations on new methodologies highlighted the importance of incorporating diverse data sources and advanced analytical techniques to improve the accuracy and effectiveness of climate impact assessments.
Moreover, learning about the latest tools and methods to inform local stakeholders and communities underscored the necessity of transparent and accessible communication in climate risk management. This reinforced the need for collaboration between scientists and local authorities to ensure that the insights gained from Machine Learning analyses are effectively utilised in adaptation planning.
What message would you like to convey to the CP community regarding the importance of integrating Machine Learning into multi-risk assessment and climate adaptation planning?
The integration of Machine Learning into multi-risk assessment and climate adaptation planning is paramount for enhancing our capacity to manage and mitigate the impacts of climate change. Machine Learning models are uniquely capable of processing vast amounts of heterogeneous data efficiently, reducing computational time, and improving the communication and dissemination of results. By leveraging these advanced analytical tools, civil protection agencies can develop more precise and actionable insights into the drivers and dynamics of climate-related risks.
How do you think CP agencies can better utilise your research to develop more effective disaster risk management pathways?
Civil protection agencies can better utilise our research by fostering continuous collaboration with the scientific community. This partnership will ensure that agencies are up-to-date with cutting-edge methodologies, results, and potential applications. Additionally, it is strongly recommended that local authorities and the scientific community cooperate to create a standardised impact dataset. Such a dataset would greatly enhance the quality of multi-risk studies and facilitate knowledge sharing across different regions and disciplines, leading to more robust and effective disaster risk management pathways.
What are the key takeaways from your research that should be prioritised by the civil protection community to enhance resilience against climate impacts in coastal areas?
Utilising Machine Learning models to integrate diverse data sources will provide a more comprehensive understanding of risks, enabling better prediction and management of extreme weather events. Additionally, acting as a bridge between the scientific community and local populations will increase awareness and preparedness, empowering communities to respond more effectively to extreme weather events and other climate-related hazards.