Every summer, typhoons threatening the Korean Peninsula draw their energy from the warm waters of the Northwest Pacific Ocean. In recent years, the frequency and intensity of extreme weather events, such as heat waves, droughts, and heavy rains, have been increasingly linked to rising sea surface temperatures (SST). Accurate predictions of SST are crucial for enhancing climate and weather forecasts. However, satellite observations, which are essential for broad and continuous monitoring, often suffer from data gaps due to clouds, precipitation, and other observational challenges, hindering long-term, high-resolution climate analysis. To address this issue, a research team at UNIST, led by Professor Jungho Im from the Department of Civil, Urban, Earth, and Environmental Engineering, has developed a groundbreaking artificial intelligence (AI) model called PARAN (Physics-Assisted Reconstruction Adversarial Network). This model restores missing satellite data and generates continuous, high-resolution SST datasets with unprecedented accuracy. PARAN, published in Remote Sensing of Environment, uses a Generative Adversarial Network (GAN), an AI architecture initially designed for image synthesis. The researchers trained the GAN using high-frequency satellite data and thermodynamic insights from numerical weather prediction (NWP) models. This integrated approach ensures that the AI-produced SST data aligns closely with real-world physical conditions, even when observations are incomplete. "Traditional methods like linear interpolation or statistical models often struggle to preserve the fine details of SST, especially during rapid temperature changes," explained Sihun Jung, the study's first author. "Our AI model not only surpasses these methods in accuracy but also maintains high fidelity in challenging conditions, making it a powerful tool for climate monitoring." Professor Im highlighted the broader implications of this technology, noting its particular significance for the Northwest Pacific, a region highly susceptible to frequent typhoons and climate variability. "By providing high-resolution SST data, we can significantly improve weather forecasts and climate models," he said. "In the long run, this technology could also play a critical role in early warning systems for marine disasters, such as marine heat waves, thereby helping to protect communities and ecosystems." The ocean stores about 90% of Earth's surface energy, and SST is a crucial boundary where heat exchange between the ocean and atmosphere takes place. Higher SSTs can transfer heat upwards, fueling typhoons, intensifying heat waves, and increasing the risk of heavy rainfall events. Despite its importance, continuous, high-resolution monitoring of SST remains difficult due to data gaps in satellite imagery. PARAN addresses these challenges by generating SST data at a spatial resolution of 2 kilometers and on an hourly basis. This level of detail allows scientists to better understand the oceanic conditions that directly influence regional weather and climate patterns. The model's ability to integrate physical oceanic principles sets it apart from conventional methods, making it particularly effective in maintaining accuracy during periods of rapid temperature change. The potential applications of this technology extend beyond improving weather forecasts. It can aid in predicting and mitigating the impacts of marine heat waves and other environmental hazards, contributing to more resilient coastal communities and ecosystems. By enhancing the overall quality and reliability of SST data, PARAN promises to advance both short-term and long-term climate science, ultimately leading to better-informed decision-making in regions prone to extreme weather events. Industry experts and climatologists have welcomed this development, praising the innovative use of AI to overcome longstanding data limitations in climate science. The technology demonstrates the growing role of AI in refining and enhancing environmental data, a trend that is expected to drive further advancements in climate modeling and forecasting. Companies like UNIST, which focus on interdisciplinary research combining AI and environmental science, are poised to make significant contributions to addressing global climate challenges.