Bayesian Data Assimilation of Density Uncertainty for Estimating the Impacts in Subaqueous Gravity Flows

Key Points

• The aim is to estimate the impacts of density uncertainty in subaqueous gravity flows using Bayesian data assimilation techniques.
• Integration of data from 50g centrifugal model experiments with fluid analysis.
• Application of a Gaussian Process Regression surrogate model for estimating density parameters.
• Bayesian framework used for data assimilation of density parameters.
• Probabilistic estimation of density parameters improved overall prediction accuracy of fluid analysis.
• Effectiveness was notable in high-impact areas but limited in low-impact regions.
• Highlighted potential dominance of turbulence and viscosity as fixed parameters in low-impact scenarios.