A source-target transfer learning framework with physics priors for subsurface chlorophyll reconstruction in the Northwest Pacific Ocean
Abstract
Three-dimensional (3D) chlorophyll-a (Chl-a) distributions provide a foundation for understanding upper-ocean ecosystem dynamics and biogeochemical variability. We developed the source-target coupled network (ST-CoupleNet), a physics-guided transfer learning framework that combines Earth system model simulations with observation-constrained reconstructions to estimate subsurface Chl-a in the Northwest Pacific Ocean. The framework first learns large-scale spatiotemporal vertical representations from long-term CMIP6 simulations and then adapts them using reconstructed fields constrained by BioGeoChemical-Argo (BGC-Argo) observations, reducing simulation biases while preserving regional variability. By integrating dual-attention, residual learning, and convolutional long short-term memory (ConvLSTM) architectures, ST-CoupleNet substantially improves reconstruction accuracy over a conventional convolutional neural network (CNN). This framework demonstrates how transferable knowledge from climate simulations can be combined with autonomous observations to generate continuous 3D Chl-a fields, enabling large-scale investigations of subsurface marine ecosystems and biogeochemical change.
Related
- Geochemical evaluation of subsurface hydrogen-brine-mineral interactions and hydrogen loss
- Integrating high-resolution magnetic surveys and metaheuristic optimization for quantitative characterization of subsurface magnetic sources in a municipal landfill
- Optical mediation of chlorophyll-a trends and radiative heating in coastal waters
- Asymmetric bloom lengthening drives high-latitude ocean chlorophyll accumulation imbalance
- A cross-regional machine learning transfer prediction framework for heat risk based on environmental characteristics of mature new towns
BibTeX
@article{gong2026sourcetarget,
title = {{A source-target transfer learning framework with physics priors for subsurface chlorophyll reconstruction in the Northwest Pacific Ocean}},
author = {Xiang Gong and Xiaoling Li and Xiaoyu Liu and Jinxuan Wan and Tao Zou and Xun Gong and Xinyu Guo and Wenming Yin},
journal = {iScience},
year = {2026},
month = oct,
doi = {10.1016/j.isci.2026.117109},
url = {https://www.cell.com/iscience/fulltext/S2589-0042(26)02487-9},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - A source-target transfer learning framework with physics priors for subsurface chlorophyll reconstruction in the Northwest Pacific Ocean
AU - Xiang Gong
AU - Xiaoling Li
AU - Xiaoyu Liu
AU - Jinxuan Wan
AU - Tao Zou
AU - Xun Gong
AU - Xinyu Guo
AU - Wenming Yin
PY - 2026
DA - 2026/10/12
JO - iScience
DO - 10.1016/j.isci.2026.117109
UR - https://www.cell.com/iscience/fulltext/S2589-0042(26)02487-9
AB - Three-dimensional (3D) chlorophyll-a (Chl-a) distributions provide a foundation for understanding upper-ocean ecosystem dynamics and biogeochemical variability. We developed the source-target coupled network (ST-CoupleNet), a physics-guided transfer learning framework that combines Earth system model simulations with observation-constrained reconstructions to estimate subsurface Chl-a in the Northwest Pacific Ocean. The framework first learns large-scale spatiotemporal vertical representations from long-term CMIP6 simulations and then adapts them using reconstructed fields constrained by BioGeoChemical-Argo (BGC-Argo) observations, reducing simulation biases while preserving regional variability. By integrating dual-attention, residual learning, and convolutional long short-term memory (ConvLSTM) architectures, ST-CoupleNet substantially improves reconstruction accuracy over a conventional convolutional neural network (CNN). This framework demonstrates how transferable knowledge from climate simulations can be combined with autonomous observations to generate continuous 3D Chl-a fields, enabling large-scale investigations of subsurface marine ecosystems and biogeochemical change.
SN - 2589-0042
ER -