Comparison of primary production algorithms for GOCI

Abstract

Geostationary Ocean Color Imager (GOCI) is the first ocean color remote sensing satellite on a geostationary orbit providing ocean color imagery around the Korean Peninsula with high spatial and temporal resolution. This study compares three primary production algorithms to estimate primary production using Geostationary Ocean Color Imager (GOCI) satellite data. Three algorithms are in the form of vertical generalized production models: the original model and two variants with different parameterization of chlorophyll-a specific carbon fixation rate. We evaluated the performance of three primary production algorithms using simulated in-situ primary production datasets observed from 1991 to 2010 at 117 stations in the Korean waters (Yellow Sea, East Sea, East China Sea, and Yeosu Bay). The comparison of the three algorithms shows that the regionally-tuned variants give better estimates than original formulation. Original formulation largely overestimates primary production over the Korean waters. There were also differences in the goodness of fit among the regions. Especially, the East Sea shows the lowest fit while Yeosu Bay shows the highest fit.three primary production algorithms to estimate primary production using Geostationary Ocean Color Imager (GOCI) satellite data. Three algorithms are in the form of vertical generalized production models: the original model and two variants with different parameterization of chlorophyll-a specific carbon fixation rate. We evaluated the performance of three primary production algorithms using simulated in-situ primary production datasets observed from 1991 to 2010 at 117 stations in the Korean waters (Yellow Sea, East Sea, East China Sea, and Yeosu Bay). The comparison of the three algorithms shows that the regionally-tuned variants give better estimates than original formulation. Original formulation largely overestimates primary production over the Korean waters. There were also diff