What question did this study set out to answer?

This research aims to develop a framework for accurately estimating bottom-water dissolved oxygen using remote sensing data and machine learning.

March 12, 2026Remote Sensing0 citationsOpen Access

Inversion and Interpretability Analysis of Bottom-Water Dissolved Oxygen in the Bohai Sea Using Multi-Source Remote Sensing Data

TLTao LiChinese Academy of Sciences JGJie GuoNingbo University SLShanwei LiuChina University of Petroleum, East China

Key Points

This research aims to develop a framework for accurately estimating bottom-water dissolved oxygen using remote sensing data and machine learning.
Developed an inversion framework integrating multi-source remote sensing data
Used decision tree-based machine learning models for predictions
Evaluated various lag intervals, identifying a 14-day lag as optimal
Employed genetic algorithm to optimize feature selection for the XGBoost model
Conducted interpretability analysis to determine key factors affecting bottom DO.
Achieved high predictive performance for bottom-water DO estimates with R2 = 0.86 and RMSE = 0.79 mg/L
Identified sea surface temperature as the main driver of bottom-water DO variation in the Bohai Sea
Model accuracy improved to R2 = 0.89 with the inclusion of bottom-layer variables
Successfully reproduced the spatiotemporal variability of bottom DO from 2022 to 2024
Reconstructed long-term distribution of bottom DO from 2014 to 2025 using satellite data.

Abstract

Seasonal hypoxia in bottom waters of the Bohai Sea poses an escalating threat to marine ecosystems, yet monitoring it via satellite remote sensing continues to be challenging due to the inaccessibility of bottom layers. However, surface bio-optical signals do not instantaneously reflect variation in bottom-water dissolved oxygen (DO); instead, a distinct temporal lag exists between surface biological activity and its influence on bottom DO. Leveraging this insight, an inversion framework was established, integrating multi-source remote sensing data with decision tree-based machine learning models to estimate bottom-water DO concentration. We evaluated multiple lag intervals for satellite-derived bio-optical variables and adopted a 14-day lag as representative of the delayed impact of surface processes on bottom DO. An optimized feature set selected via a genetic algorithm (GA) was used to train the XGBoost model, which achieved high predictive performance (R2 = 0.86, RMSE = 0.79 mg/L, MAPE = 8.89%). Interpretability analysis identified the sea surface temperature as the dominant driver of bottom-water DO variation in the Bohai Sea. The framework successfully reproduced the spatiotemporal variability in bottom DO from 2022 to 2024 in the Bohai Sea and captured the locations of summer hypoxic zones. Further analysis demonstrated that incorporating physically based bottom-layer variables substantially enhances model accuracy (R2 = 0.89, RMSE = 0.68 mg/L, MAPE = 7.85%), underscoring their critical role in regulating bottom-water DO concentrations. Building on the established inversion framework and integrating extended in situ and satellite observations, we reconstruct the long-term temporal distribution of bottom DO in the Bohai Sea from 2014 to 2025, revealing the considerable potential of satellite data for monitoring bottom-water DO conditions in coastal seas.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper