Speaking at AGU Fall Meeting 2025 — Developments in ML across Earth-System Modeling
Presenting recent work on data-driven flux parameterization in the atmospheric boundary layer.
Abed Hammoud
Postdoctoral Research Associate
About
I’m a postdoctoral research associate at Princeton University in the Department of Civil and Environmental Engineering, working with Prof. Elie Bou-Zeid on data-driven parameterizations of turbulent fluxes in the atmospheric boundary layer, with Mitchell Bushuk (NOAA GFDL).
My research sits at the intersection of scientific machine learning, uncertainty quantification, and geophysical fluid dynamics. I build physics-informed and probabilistic learning frameworks for problems where models are imperfect, observations are sparse and noisy, and the underlying dynamics are chaotic — from boundary-layer turbulence and Rayleigh–Bénard convection to ocean color retrieval and oil-spill source identification.
I completed my PhD in Mechanical Engineering at KAUST under Prof. Omar Knio and Prof. Edriss S. Titi, where I developed AI-based frameworks for data assimilation and downscaling in uncertain chaotic systems. I am a 2026 Gordon and Betty Moore Foundation Postdoctoral Fellow and a Princeton–NOAA GFDL CIMES Fellow.
Interests
- Scientific Machine Learning
- Uncertainty Quantification
- Data Assimilation
- Geophysical Fluid Dynamics
- Boundary Layer Meteorology
- Remote Sensing & Ocean Color
Education
PhD, Mechanical Engineering
King Abdullah University of Science and Technology (KAUST)
MSc, Mechanical Engineering
King Abdullah University of Science and Technology (KAUST)
BEng, Mechanical Engineering
American University of Beirut
Featured Publications
Recent Publications
(2025).
From Standard to Bayesian: Revisiting Ocean Color Model Evaluation.
in AGU Earth and Space Science.
(2025).
The new Climate Change Center of Saudi Arabia — A big step towards understanding and predicting the distinct climate of the Arabian Peninsula.
in AGU Earth’s Future.
(2025).
On the potential of Bayesian neural networks for estimating chlorophyll-a concentration from satellite data.
in MDPI-RS.
(2025).
Identification of Moving Sources in Stochastic Fields: A Bayesian Inferential Approach with Application to Marine Traffic in the Mediterranean Sea.
in Comp. Geo..
(2025).
Downscaling Using CDAnet Under Observational and Model Noises: The Rayleigh-Bénard Convection Paradigm.
in Comp. Geo..
(2024).
Data Assimilation in Chaotic Systems Using Deep Reinforcement Learning.
in JAMES.
(2024).
Two-step AI-aided Bayesian source identification of urban-scale pollution.
in Atmospheric Environment.
(2024).
Oil spill risk analysis for the NEOM shoreline.
in Scientific Reports.
(2023).
Global Sensitivity Analysis of an Oil Spill Model: a Regularized Regression Approach..
in Frontiers in Marine Science.
(2023).
Semantic Segmentation of Mesoscale Eddies in the Arabian Sea: A Deep Learning Approach.
in MDPI RS.
(2023).
CDAnet: A Physics-Informed Deep Neural Network for Downscaling Fluid Flows.
in JAMES.
(2023).
A Physics-Informed Deep Neural Network for Backward in Time Prediction: Application to Rayleigh-Bénard Convection.
in AMS AIES.
(2022).
Continuous and discrete data assimilation with noisy observations for the Rayleigh-Bénard convection: a computational study.
in Comp. Geo..
(2021).
Moving source identification in an uncertain marine flow: Mediterranean Sea application.
in Ocean Engineering.
Under Review
(2026).
Reinforcement Learning, Optimal Control, and Bayesian Filtering in Data Assimilation.
Under review.
(2025).
Train Long, Think Short: Curriculum Learning for Efficient Reasoning.
Under review.
News & Talks
Awarded the Gordon and Betty Moore Foundation Postdoctoral Fellowship!
Honored to receive a Gordon and Betty Moore Foundation Postdoctoral Fellowship in support of my work at Princeton (2026–2027).
Convening a session at AGU 2025
Co-convening a session at the AGU Fall Meeting 2025.
Abstract accepted to AMS BLT '25!
Your news content here…
NCAR Exploratory Computing Allocation awarded (UPRI0030)
Awarded 2,000 GPU-hours from the NCAR Computational and Information Systems Lab as PI to support data-driven boundary-layer research (2025–2027).
Invited talk at the RWTH-Aachen Uncertainty Quantification Seminar
Gave an invited seminar on AI for state estimation in chaotic systems.
Get in touch
I’m always glad to discuss collaborations, talks, or student opportunities at the intersection of scientific ML, UQ, and geophysical fluid dynamics.
Email: ah1389@princeton.edu · Office: E-Quad, Princeton University