AI-Enhanced Strategic Foresight Analysis for Multi-Hazard Land-Sea Interface Management in the Valencian Community

Societies face growing challenges driven by the compounded effect of climate change and interconnected hazards, alongside broader environmental issues. The Land-Sea Interface (LSI), where several ecosystems converge and interact, represents a complex environment with unique dynamics and interdependencies. These characteristics pose significant challenges for impact assessment and planning and require appropriate methodologies to navigate its complexities effectively.
The exploration and development of innovative tools for multi-hazard impact and adaptation planning have become essential for understanding and unraveling the interplay between multiple pressures while exploring future scenarios, anticipating uncertainties, and supporting informed and robust decisions.
In this setting, strategic foresight analysis has emerged as a proactive approach to address complex challenges, helping organizations to anticipate and prepare for future risks and opportunities. Its inherent interdisciplinarity and capacity to offer insights into complex dynamics make it useful to enhance building systemic resilience, and transdisciplinary collaboration, enabling the integration of diverse knowledge systems and stakeholder perspectives into adaptive planning and decision-making. Within this study, an AI-enhanced strategic foresight analysis, specifically tailored to the Valencia region's coastal wetlands, is proposed to respond to the critical need to understand the multidimensional dynamics of Land-Sea Interactions (LSI). The Valencia coastal case, highly prone to multiple hazards, is distinguished by complex interactions between anthropogenic and climate-driven pressures, which amplify their combined impacts on vital ecological.
The methodological approach integrates traditional foresight tools (i.e., horizon scanning, megatrend analysis, and scenario planning) with robust,  innovative science-based approaches. Specifically, megatrend analysis was conducted using Copernicus climate data, and scenarios based on a Cumulative Impact Assessment (CIA) supported by Generative Artificial Intelligence (Generative-AI).AI tools were employed for data analysis and to generate impact weights that accurately reflect the effects of multiple hazards and anthropogenic pressures on different ecosystems. Local stakeholders and expert involvement played a central role in these tasks, ensuring that model development and application on the Valencia case were aligned with local priorities and challenges. 
Looking at the key outcomes of the appraisal, the m
megatrend analysis revealed increasing trends in climate pressures, such as sea-level rise, storm surges, coastal erosion, and air temperature median scores over time. The most marked values were shown in the southern wetlands and on the Albufera coast, underpinning their experience to compounded pressures due to their proximity to the coast.  
Finally, scenario analysis indicated a progressive intensification of cumulative impacts for different RCPs (RCP 4.5 and 8.5) and time horizons (2050 and 2100). In particular, in all scenarios, forests and seminatural areas consistently exhibited the highest cumulative impact scores due to their sensitivity to hazards, in particular sea level rise, storm surges, and air temperature changes.
However, coastal wetlands stood out as the most critical category in future scenarios, due to their exposition to the interplay of multiple marine-driven hazards.
This study facilitated the co-design of a novel CIA approach, providing insights into multi-hazard impacts and a solid foundation for further research while enhancing decision-making processes in spatial planning and  equipping stakeholders with actionable insights  enhancing resilience and long-term preparedness.