Recorded 05 February 2026. Mohamed Iskandarani of the University of Miami presents "Uncertainty Quantification in Ocean-Atmosphere Simulations Using Polynomial Chaos Methods and Gaussian Process Regression" at IPAM's Mathematics and Machine Learning for Earth System Simulation Workshop. Abstract: Gaussian Process Regression and Polynomial Chaos methods have been used over the last decade to quantify uncertainties in simulations, and for data analysis. Both techniques offer the possibility of building a model surrogate but restrict themselves to exploring a small subset of the model parameter space. These surrogates are then used to perform the uncertainty analysis. We present an overview of these techniques, highlighting their strength and their limitations. We present two applications of these techniques: one involved the calibration of model parameters using observational data obtained during Typhoon Fanapi, and the other consists of using Gaussian Process Regression to reconstruct the velocity field obtained from a massive release of surface drifters. Learn more online at: https://www.ipam.ucla.edu/programs/workshops/mathematics-and-machine-learning-for-earth-system-simulation/?tab=overview