A new study led by Professor Jonghun Kam’s team at POSTECH (Pohang University of Science and Technology) has issued a stark warning about Pakistan’s future, forecasting that the country will face extreme "super floods" and "extreme droughts" roughly every 15 years due to climate change. The research, published in Environmental Research Letters, uses a cutting-edge artificial intelligence model to project these increasingly severe weather events in the upper Indus Basin (UIB), a region critical to Pakistan’s water supply and agricultural stability. The study highlights how global warming is intensifying climate extremes, especially in high-altitude regions where glaciers are rapidly melting. Pakistan’s major rivers, including the Indus, depend heavily on glacial meltwater, making them highly sensitive to temperature shifts. As a developing nation in the Global South, Pakistan faces heightened vulnerability to climate change due to limited resources and infrastructure for climate monitoring and adaptation. Traditional climate models have struggled to accurately predict extreme weather in complex mountainous terrain like Pakistan’s, often underestimating flood risks or overestimating rainfall. To address this, the POSTECH team developed AI models trained on historical river flow data and real-world observations. By learning from past patterns, the AI significantly improved the accuracy of predictions for extreme events, offering a more reliable alternative to conventional methods. The results are alarming: the upper Indus River is projected to experience major floods and severe droughts every 15 years, while other rivers in the region could face such extreme events as frequently as every 11 years. These findings underscore the urgent need for region-specific water management strategies tailored to individual river basins rather than relying on broad, generic policies. Professor Kam emphasized that this AI-driven approach is not only vital for Pakistan but could also serve as a model for other climate-vulnerable, data-scarce regions worldwide. The technology enables more precise, localized climate forecasting, empowering governments to prepare for and respond to climate risks effectively. The research was conducted by Professor Kam’s team from POSTECH’s Division of Environmental Science and Engineering, in collaboration with doctoral student Hassan Raza and Professor Dagang Wang’s team from Sun Yat-sen University in China.