A5055741456- Meteorological Phenomena and Simulations
- Model Reduction and Neural Networks
- Climate variability and models
- Oceanographic and Atmospheric Processes
- Oil Spill Detection and Mitigation
- Marine and coastal ecosystems
- Advanced Mathematical Modeling in Engineering
- Fluid Dynamics and Turbulent Flows
- Wind and Air Flow Studies
- Lattice Boltzmann Simulation Studies
- Ocean Waves and Remote Sensing
- Water Quality Monitoring and Analysis
- Coastal and Marine Management
- Atmospheric and Environmental Gas Dynamics
- Hydrological Forecasting Using AI
- Fluid Dynamics and Vibration Analysis
- Diffusion and Search Dynamics
- Maritime Navigation and Safety
- Water Quality Monitoring Technologies
- International Maritime Law Issues
- Air Quality and Health Impacts
- Tropical and Extratropical Cyclones Research
King Abdullah University of Science and Technology
2020-2024
The King's College
2023
University of Science and Technology
2021
Abstract The Red Sea, home to the second-longest coral reef system in world, is a vital resource for Kingdom of Saudi Arabia. Sea provides 90% Kingdom’s potable water by desalinization, supporting tourism, shipping, aquaculture, and fishing industries, which together contribute about 10%–20% country’s GDP. All these activities, those elsewhere region, critically depend on oceanic atmospheric conditions. At time mega-development projects along coast, global warming, authorities are working...
Abstract A risk analysis is conducted considering an array of release sources located around the NEOM shoreline. The are selected close to coast and in neighboring regions high marine traffic. evolution oil spills released by these simulated using MOHID model, driven validated, high-resolution met-ocean fields Red Sea. For each source, simulations over a 4-week period, starting from first, tenth twentieth days month, covering five consecutive years. total 180 thus for source location,...
Abstract Generating high‐resolution flow fields is of paramount importance for various applications in engineering and climate sciences. This typically achieved by solving the governing dynamical equations on meshes, suitably nudged toward available coarse‐scale data. To alleviate computational cost such downscaling process, we develop a physics‐informed deep neural network (PI‐DNN) that mimics mapping information into their fine‐scale counterparts continuous data assimilation (CDA)....
Abstract Backward-in-time predictions are needed to better understand the underlying dynamics of physical fluid flows and improve future forecasts. However, integrating backward in time is challenging because numerical instabilities caused by diffusive nature systems nonlinearities governing equations. Although this problem has been long addressed using a nonpositive diffusion coefficient when backward, it notoriously inaccurate. In study, physics-informed deep neural network (PI-DNN)...
Abstract Data assimilation (DA) plays a pivotal role in diverse applications, ranging from weather forecasting to trajectory planning for autonomous vehicles. A prime example is the widely used ensemble Kalman filter (EnKF), which relies on filter's linear update equation correct each of forecast member's state with incoming observations. Recent advancements have witnessed emergence deep learning approaches this domain, primarily within supervised framework. However, adaptability such models...
To support accidental spill rapid response efforts, oil simulations may generally need to account for uncertainties concerning the nature and properties of spill, which compound those inherent in model parameterizations. A full detailed these sources uncertainty would however require prohibitive resources needed sample a large dimensional space. In this work, variance-based sensitivity analysis is conducted explore possibility restricting priori set uncertain parameters, at least context...
Data assimilation (DA) plays a pivotal role in diverse applications, ranging from climate predictions and weather forecasts to trajectory planning for autonomous vehicles. A prime example is the widely used ensemble Kalman filter (EnKF), which relies on linear updates minimize variance among of forecast states. Recent advancements have seen emergence deep learning approaches this domain, primarily within supervised framework. However, adaptability such models untrained scenarios remains...
Abstract Obtaining accurate high-resolution representations of model outputs is essential to describe the system dynamics. In general, however, only spatially- and temporally-coarse observations states are available. These can also be corrupted by noise. Downscaling a process/scheme in which one uses coarse scale reconstruct solution states. Continuous Data Assimilation (CDA) recently introduced downscaling algorithm that constructs an increasingly representation continuously nudging large...
Data assimilation (DA) plays a pivotal role in diverse applications, ranging from climate predictions and weather forecasts to trajectory planning for autonomous vehicles. A prime example is the widely used ensemble Kalman filter (EnKF), which relies on linear updates minimize variance among of forecast states. Recent advancements have seen emergence deep learning approaches this domain, primarily within supervised framework. However, adaptability such models untrained scenarios remains...
This study explores the use of Bayesian Neural Networks (BNNs) for estimating chlorophyll-a concentration ([CHL-a]) from remotely sensed data. The BNN model enables uncertainty quantification, offering additional layers information compared to traditional ocean colour models. An extensive in situ bio-optical dataset is utilized, generated by merging 27 data sources across world’s oceans. demonstrates remarkable capability capturing mesoscale features and circulation patterns, providing...
Detecting mesoscale ocean eddies provides a better understanding of the oceanic processes that govern transport salt, heat, and carbon. Established eddy detection techniques rely on physical or geometric criteria, they notoriously fail to predict are neither circular nor elliptical in shape. Recently, deep learning have been applied for semantic segmentation eddies, relying outputs traditional algorithms supervise training neural network. However, this approach limits network’s predictions...
A risk analysis is conducted considering several release sources located around the NEOM shoreline. The are selected close to coast and in neighboring regions of high marine traffic. evolution oil spills released by these simulated using MOHID model, driven validated, high-resolution met-ocean fields Red Sea. For each source, simulations over a 4-week period, starting from first, tenth twentieth days month, covering five consecutive years. total 48 thus for source location, adequately...
Efficient downscaling of large ensembles coarse-scale information is crucial in several applications, such as oceanic and atmospheric modeling. The determining form map a theoretical lifting function from the low-resolution solution trajectories dissipative dynamical system to their corresponding fine-scale counterparts. Recently, physics-informed deep neural network ("CDAnet") was introduced, providing surrogate for efficient downscaling. CDAnet demonstrated efficiently downscale noise-free...
Obtaining accurate high-resolution representations of model outputs is essential to describe the system dynamics. In general, however, only spatially- and temporally-coarse observations states are available. These can also be corrupted by noise. Downscaling a process/scheme in which one uses coarse scale reconstruct solution states. Continuous Data Assimilation (CDA) recently introduced downscaling algorithm that constructs an increasingly representation continuously nudging large scales...