Characterizing spatial patterns of satellite-derived chlorophyll-a in the Bohai and Yellow Seas of China using self-organizing maps (SOM) approach

Dynamic of chlorophyll-a (Chl-a) concentration is essential information to understand the status and trends of marine ecosystems. In this study, a Self-Organizing Map (SOM) was applied to delineate space-in-time patterns of Chl-a from satellite dataset during 2002–2022 over the Bohai and Yellow Seas of China (BYS). Six typical Chl-a spatial patterns were discerned through a 2 × 3 nodes SOM, while temporal evolutions of dominant spatial patterns were analyzed. The Chl-a spatial patterns were characterized by different concentrations and gradients, and obviously changed over time. The Chl-a spatial patterns and their temporal evolutions were mainly shaped by joint effects of nutrient level, light availability, water column stability, and other factors. Our findings provide first glimpse of space-in-time Chl-a dynamics in the BYS, and complement to the traditional time-in-space Chl-a pattern studies. The accurate identification and classification of the Chl-a spatial patterns are of great significance to marine regionalization and management.