Abstract
Abstract Flooding is among the most destructive natural disasters, causing widespread property damage, loss of life, and public health challenges, particularly affecting vulnerable populations who lack access to early warning systems. To address this, nations are adopting advanced hazard prediction and risk assessment tools. Google Earth Engine (GEE), combined with machine learning (ML), has proven instrumental in improving flood prediction accuracy and timeliness. This study focuses on assessing coastal flood risks in Libya, a country with a 1770-km Mediterranean coastline, identified as highly vulnerable by the World Bank due to rising sea levels. The research applies ML and GEE to create detailed flood risk maps for Libya, aiming to inform policymakers and urban planners about high-risk areas. Findings reveal that a 1-m sea level rise could threaten approximately 8480 km2 of coastal terrain. Particularly vulnerable regions include salt marshes (sabkhas) (e.g., Misrata, Tawergha, Abu Kammash, and Ghazil), the Bomba Gulf, and coastal zones in Tripoli, such as Tajoura and Souq al-Juma. These areas are at heightened risk, underscoring the urgency to rethink urban development strategies in flood-prone regions. The study advocates for resilience-oriented urban design, incorporating climate-resilient infrastructure and smart city management to protect communities. The integration of GEE and ML offers Libyan decision-makers a powerful toolkit for proactive flood risk management. By adopting these technologies, Libya can mitigate the adverse impacts of coastal flooding, safeguarding lives, infrastructure, and ecosystems along its vulnerable Mediterranean coastline.