engleski

Application of Multivariate Statistical Techniques to Relate Phytoplankton Optical Signatures and Taxonomy

Phytoplankton is the base of the aquatic food web, providing a direct food source for zooplankton and fish and plays a major role in the global carbon cycle. Satellite ocean color remote sensing allows us to observe the spatial and temporal variability of Chlorophyll a (Chl a) as a proxy for phytoplankton biomass on a global scale. Ocean color observations reach time and space beyond our capabilities with research vessels and, therefore, may fill in data gaps where observations are lacking. While Chl a concentration remains the most commonly used product as a proxy of phytoplankton biomass, recent and ongoing algorithm developments are moving beyond its retrieval to more descriptive indices of phytoplankton community structure in the ocean by exploiting optical signatures of different phytoplankton groups. During the Sea-to-Space Particle Investigation project on the Schmidt Ocean R/V Falkor, scientists collected multiple data types of phytoplankton taxonomy including classical and imaging flow cytometry, microscopy, pigments and DNA. We will present the application of two multivariate statistical techniques, Cluster analysis and Principal Component Analysis (PCA), to in situ measured phytoplankton taxonomic and radiometric data. Cluster analysis is applied to reveal the distribution patterns of the phytoplankton taxonomic and radiometric features over the sampling grid. PCA, an exploratory ordination technique, is used to reveal the radiometric wavelengths that best explain phytoplankton groups detected with each method, thereby creating a predictive capability of in situ and satellite-derived radiometry for different phytoplankton groups.

Phytoplankton, Satellite ocean color, Sea-to-Space Particle Investigation

nije evidentirano