文献类型：期刊文献

英文题名：Enhancing the generalization of turbulent mixing parameterization by physics-informed machine learning

作者：Hu, Minghao[1];Xie, Lingling[1,2,3];Li, Mingming[1,2,3];Chen, Xiaotong[1]

机构：[1]Guangdong Ocean Univ, Coll Ocean & Meteorol, Lab Coastal Ocean Variat & Disaster Predict, Zhanjiang 524088, Peoples R China;[2]Resources & Environm Continent Shelf Sea & Deep Oc, Zhanjiang 524088, Peoples R China;[3]Guangdong Prov Observat & Res Stn Trop Ocean Envir, Zhanjiang 524088, Peoples R China

年份：2025

卷号：44

期号：12

起止页码：79

外文期刊名：ACTA OCEANOLOGICA SINICA

收录：SCI-EXPANDED(收录号：WOS:001714347500009)、、WOS

基金：The authors thank Quanan Zheng for his guidance on the manuscript. We would like to express our gratitude to the Archiving, Validation and Interpretation of Satellite Oceanographic data (AVISO) and the European Centre for Medium-Range Weather Forecasts (ECMWF) for providing the altimeter and wind data used in this study, which are available via their official websites (AVISO: http://www.aviso.oceanobs.com/; ECMWF: http://apps.ecmwf.int/datasets/).

语种：英文

外文关键词：microstructure observations; turbulent mixing; physics-informed machine learning; generalization

外文摘要：Using in-situ microstructure observations from 2010 to 2018, this study investigates the performance and generalization of machine learning models in parameterizing turbulent mixing in the northwestern South China Sea. The results show that the data-driven extreme gradient boosting (XGBoost) performs better than the other four models, i.e., random forest, neural network, linear regression and support vector machine regression. In order to further improve the generalization of machine learning-based parameterization method, we propose a physics-informed machine learning (PIML) that couples the MacKinnon-Gregg model (known as the MG model) and Osborn's formula to the XGBoost model. The correlation coefficient (r) and root mean square error (RMSE) between the estimated and observed lg(epsilon) (where epsilon denotes the turbulent kinetic energy dissipation rate) from the PIML are improved by 14% and 16%, respectively. The results also show that PIML effectively improves the generalization of the XGBoost-based parameterization method, enhancing r and RMSE by 35% and 75%, respectively.