Evaluation of satellite-derived MODIS chlorophyll algorithms in the northern Antarctic Peninsula

Abstract

In this study, we used in situ spectral remote sensing reflectance data, Rrs (λ), coincident with chlorophyll-a concentration (Chla) measurements obtained by the Brazilian High Latitude Oceanography Group (GOAL, FURG) and provided by the SeaWiFS Bio-Optical Archive and Storage System (SeaBASS) in the northern Antarctic Peninsula (NAP). These data span the period between 2002 and 2010. The objective of this work was to evaluate the existing MODIS/Aqua chlorophyll algorithms and show an alternative atmospheric correction scheme based on the near-infrared (NIR) and shortwave near-infrared (SWIR) bands that could be used in the turbid coastal waters of the NAP. The current OC2M and OC3M empirical and GSM01 semi-analytical algorithms tend to underestimate Chla, which was determined by using Chla estimated by either high performance liquid chromatography (HPLC) or fluorometry. The GSM01 algorithm showed the best performance. The NIR-SWIR atmospheric correction scheme (Wang et al., 2009) was applied for the northern Antarctic Peninsula (NAP) in the presence of sea ice and deicing waters and showed an increase in Chla in the range from 0.75 to 2.5 mg m−3 and a slight decrease in the values above 2.5 mg m−3. Regional empirical chlorophyll algorithms based on the 3-band OC3M operational algorithm were derived and showed better performance in estimating Chla than the global algorithms.