Data-Driven Probabilistic Air-Sea Flux Parameterization

Jiarong Wu

doi:10.1029/2025GL120472

Data-Driven Probabilistic Air-Sea Flux Parameterization

2026-03-19

0

OA

AI

Aneesh C. Subramanian

DOI：10.1029/2025GL120472

Original

Original request for help

Save

Abstract

En 中文

Accurately quantifying air-sea fluxes is important for understanding air-sea interactions and improving coupled weather and climate models. This study introduces a probabilistic framework to represent the highly variable nature of air-sea fluxes, which is missing in deterministic bulk algorithms. Assuming Gaussian distributions conditioned on the input variables, we use artificial neural networks and eddy-covariance measurement data to estimate the mean and variance by minimizing negative log-likelihood loss. The trained neural networks provide alternative mean flux estimates to existing bulk algorithms, and quantify the uncertainty around the mean estimates. A stochastic parameterization of air-sea turbulent fluxes can be constructed by sampling from the predicted distributions. Tests in a single-column forced upper-ocean model suggest that changes in flux algorithms influence sea surface temperature and mixed layer depth seasonally. The ensemble spread in stochastic runs is most pronounced during spring restratification.

Keywords:

air-sea turbulent fluxes

stochastic parameterization

machine learning

uncertainty quantification

bulk algorithm

single-column model

Data-Driven Probabilistic Air-Sea Flux Parameterization

Abstract

Journal

Organization