浮游植物
海洋学
环境科学
卫星
地理
生物
生态学
地质学
天文
营养物
物理
作者
Deyong Sun,Yuhang Chen,Zhenghao Li,Shengqiang Wang,Peng Chen,Yu Huan
摘要
Abstract Documenting the variations of phytoplankton pigments is essential for understanding phytoplankton assemblages' dynamics and physiological characteristics in the global oceans. Advanced chlorophyll a satellite product has provided an overview of global phytoplankton biomass, yet deeper taxonomic biomass monitoring demands more detailed pigment information. In this study, we have collected 1562 in situ measured sea surface samples (depth < 10 m) from 1997 to 2017 covering the global ocean, where the in situ dataset consists of pigment concentration and phytoplankton absorption coefficient ( a ph ). A global semi‐analytical pigment model was developed using the Gaussian decomposition method and was evaluated by the leave‐one‐out cross validation method. Model accuracy verification based on in situ measurement (e.g., mean absolute percentage errors below ~ 67% for most pigments) and satellite synchronization dataset (e.g., most pigments have mean absolute percentage errors below ~ 70%) showed the reliable performance of the pigment model in the global oceans. We reconstructed the global hyperspectral phytoplankton absorption by the three‐component absorption model of the phytoplankton size class, enabling satellite application of the pigment model and obtaining the nearly 20‐yr spatiotemporal variations of 14 phytoplankton pigments. Global pigment concentrations exhibited a generally relatively unchanging pattern at the beginning of the 21 st century, as evidenced by stable geographic distribution and temporal trends, while local environmental disturbances created anomalies in 1‐yr period. Our findings provide an effective model to detect global pigment concentration and essential satellite products for further analysis of phytoplankton assemblages.
科研通智能强力驱动
Strongly Powered by AbleSci AI