To better understand the marine phytoplankton variability on different scales in both space and time, this study proposed a machine learning based scheme to provide continuous and consistent long-term observations of various phytoplankton groups from space on a global scale, which enables time series analysis for further trend and anomaly investigations. This study provides an essential ocean variable to help assess the ocean health in the biogeochemical aspect.

Continuous monitoring of phytoplankton groups using satellite data is crucial for understanding global ocean phytoplankton variability on different scales in both space and time. This study focuses on four important phytoplankton groups in the Atlantic Ocean to investigate their trend, anomaly and phenological characteristics both over the whole region and at subscales. This study paves the way to promote potentially important ocean monitoring indicators to help sustain the ocean health.

In this paper, we will show how a proxy for particulate composition (PPC), classifying the suspended particulate matter into its organic, mineral, or mixed fractions, can be estimated from remote-sensing observations. The selected algorithm will then be applied to MERIS observations (2002–2012) over global coastal waters to discuss the significance of this new product. A specific focus will be on the English Channel and the southern North Sea.