KAUST's HALO group launched a CubeSat in 2023 for high-precision Earth observation in the Gulf region, combining GNSS Reflectometry and hyperspectral sensing. The satellite monitors vegetation, soil, agriculture, and ecosystem health, providing detailed estimates of irrigation water use and vegetation health. The Extreme Statistics (XSTAT) research group at KAUST focuses on the mathematical modeling and prediction of extreme weather and climate events. Why it matters: These KAUST initiatives enhance climate resilience in the region through advanced monitoring, statistical modeling, and predictive capabilities.
KAUST scientists are developing models to predict extreme weather events like the 2009 Jeddah flood, which caused significant damage. Prof. Ibrahim Hoteit's team is using data from satellites, international sources, and local entities like PME and the Jeddah Municipality to build high-resolution models. The aim is to improve predictions of extreme rain events by one or two days and issue timely warnings. Why it matters: Improving extreme weather prediction is crucial for mitigating the impact of climate change in vulnerable regions like the GCC.
KAUST's Atmospheric and Climate Modeling group, led by Georgiy Stenchikov, is using high-resolution global and regional climate models to predict climate change in the Middle East, focusing on local atmospheric and oceanic processes. The group developed coupled regional atmospheric and oceanic models for the Red Sea, accounting for the climate effect of aerosols, especially dust, which is significant in the region. They found that dust strongly affects the Red Sea, causing high optical depth and solar cooling effect, particularly in the southern part, impacting energy balance and circulation. Why it matters: Improving regional climate models with specific attention to dust and aerosols is crucial for predicting and mitigating the environmental impacts of climate change in arid regions like the Middle East.
This paper introduces a deep vision-based framework for predicting coastal floods under climate change, addressing the challenges of limited training data and high-dimensional output. The framework employs and compares various deep learning models, including a custom compact CNN architecture, against geostatistical and traditional machine learning methods. A new synthetic dataset of flood inundation maps for Abu Dhabi's coast is also provided to benchmark future models.
Researchers have developed a CNN-based deep learning model for predicting coastal flooding in cities under various sea-level rise scenarios. The model utilizes a vision-based, low-resource DL framework and is trained on datasets from Abu Dhabi and San Francisco. Results show a 20% reduction in mean absolute error compared to existing methods, demonstrating potential for scalable coastal flood management.