Afonso Ferreira, Ana C. Brito, Carlos R.B. Mendes, Vanda Brotas, Raul Costa, Catarina Guerreiro, Carolina Sá and Thomas Jackson

Remote sensing of chlorophyll a (chl-a; a proxy of phytoplankton biomass) is as an essential tool for monitoring marine ecosystems, particularly for polar regions where in-situ data availability is very limited. For the Southern Ocean, however, the current global chl-a algorithms tend to underestimate chl-a, preventing accurate retrievals of chl-a. This work sought to develop a new chl-a algorithm for the Antarctic Peninsula and to assess potential drivers of chl-a underestimation in this region. OC4-SO, the algorithm developed and presented here, exhibits the following advantages: i) developed using a 20+ years and spatially thorough in-situ HPLC-derived chl-a dataset; ii) based on a multi-sensor product, which are expected to be increasingly popular as the number of operating satellites increases; iii) adequately corrects for chl-a underestimation, now allowing for the assessment of the Antarctic seasonal cycle which was not previously possible using global algorithms; and v) exhibits a good performance for Antarctic waters outside of the Antarctic Peninsula. Low particulate backscattering coefficients in the green region of the light spectrum were seen to be main factor linked to chl-a underestimation in high chl-a waters, corroborating previous works. Overall, the OC4-SO can be a valuable tool for ocean colour remote sensing in Antarctic waters, enabling more accurate estimates of chl-a and contributing to a better understanding of this ecosystem’s role to the global biogeochemical cycle.