Usability of Sentinel-3 satellite ocean colour data for examination and chlorophyll gradient classification of submesoscale island wake formations

Abstract

Submesoscale turbulent formations present challenges to sampling efforts due to their scale and transient nature. The development of successful sampling methods is imperative to achieving an understanding of the biological and ecological implications of these turbulent formations. This study assesses the viability of using Sentinel-3 OLCI multispectral satellite data for the study of submesoscale island wake formations. This research subsequently samples the eddies and filaments with virtual transects of satellite chlorophyll products, calculates metrics such as formation gradient and FWHM, identifies key environmental drivers of wake characteristics, and presents a novel framework by which to classify these formations. The study is subject to limitations regarding cloud cover of a small area of interest, but nonetheless is able to identify seasonal occurrence of formations, and delineate environmental variables driving formation metrics (i.e. eddy gradient is significantly correlated with current velocity (MLR; slope = 1.490, p-value = 0.001) and current angle (MLR; slope = 1.660 x 10-02, p-value = 0.016), and eddy FWHM is significantly correlated with: a formation’s distance from island (MLR; slope = 0.015, p-value = 0.024), island diameter (MLR; slope = 0.361, p-value = < 0.001), and current velocity (MLR; slope = 10.205, p-value = < 0.001)). An eddy and filament characterisation framework is proposed regarding chlorophyll gradient, in anticipation of the biological implications of such gradients: weak gradient (< 25th percentile of all gradient values of the formations sampled within the AOI), moderate gradient (> 25th percentile and < 75th percentile of all gradient values), and strong gradient (> 75th percentile of all gradient values) formations. The results regarding environmental drivers of wake characteristics align with previous studies of the physical implications of island wakes, as well as provide new insight. This research presents a missing link between a physical understanding of submesoscale island wakes and a more extensive understanding regarding the biological signatures – expanding the potential for further study regarding the biological and conservation/management implications.