Hyperspectral Satellite-Derived Phytoplankton Diagnostic Pigments in the East China Sea: Spatiotemporal Patterns and Ecological Response to Coastal Fronts

Abstract

Phytoplankton diagnostic pigments are critical for identifying phytoplankton functional types, yet their retrieval is limited by sparse in situ data and low spectral resolution. This study retrieved phytoplankton diagnostic pigments in the East China Sea using hyperspectral PACE ocean color data with 5 nm resolution in the visible range. Fine spectral features from synchronous in situ HPLC–hyperspectral remote sensing reflectance datasets from the ECS and global ocean were extracted via Gaussian decomposition to construct an Optuna-optimized CatBoost model. The model performance was validated against multispectral MODIS-derived results, and its robustness was further verified across independent random and two out-of-distribution generalization scenarios. Results showed PACE/OCI test set ${{R}^2}$ exceeded 0.80 with MAE of 0.2–0.4 and RMSE of 0.3–0.7, comparing to MODIS’s test set ${{R}^2}$exceeded 0.69 with MAE of 0.2–0.7 and RMSE of 0.4–1.1. Monthly average pigment distributions from PACE revealed a spatial pattern of nearshore/offshore high, outer sea low and seasonal trend of winter-spring high and summer low, with regional variations. Case analyses of the Shandong Peninsula and Min-Zhe coastal fronts confirmed that seasonal ocean fronts regulated pigment distribution by acting as nearshore accumulation barriers in winter and promoting blooms in spring. Our study provides a reliable framework for ECS pigment monitoring, supporting coastal ecological assessment, and offers a reference for hyperspectral satellite applications in marginal seas.