Parameterization of vertical chlorophyll a in the Arctic Ocean: impact of the subsurface chlorophyll maximum on regional, seasonal, and annual primary production estimates
- 1Takuvik Joint International Laboratory, Laval University (Canada) – CNRS (France), UMI3376, Département de biologie et Québec-Océan, Université Laval, Québec, Québec G1V 0A6, Canada
- 2Institut des sciences de la mer de Rimouski, Université du Québec à Rimouski, 310 Allée des Ursulines, Rimouski, Québec G5L 3A1, Canada
- 3BORÉAS and Département de biologie, chimie et géographie, Université du Québec à Rimouski, 300 Allée des Ursulines, Rimouski, Québec G5L 3A1, Canada
Abstract. Predicting water-column phytoplankton biomass from near-surface measurements is a common approach in biological oceanography, particularly since the advent of satellite remote sensing of ocean color (OC). In the Arctic Ocean, deep subsurface chlorophyll maxima (SCMs) that significantly contribute to primary production (PP) are often observed. These are neither detected by ocean color sensors nor accounted for in the primary production models applied to the Arctic Ocean. Here, we assemble a large database of pan-Arctic observations (i.e., 5206 stations) and develop an empirical model to estimate vertical chlorophyll a (Chl a) according to (1) the shelf–offshore gradient delimited by the 50 m isobath, (2) seasonal variability along pre-bloom, post-bloom, and winter periods, and (3) regional differences across ten sub-Arctic and Arctic seas. Our detailed analysis of the dataset shows that, for the pre-bloom and winter periods, as well as for high surface Chl a concentration (Chl asurf; 0.7–30 mg m−3) throughout the open water period, the Chl a maximum is mainly located at or near the surface. Deep SCMs occur chiefly during the post-bloom period when Chl asurf is low (0–0.5 mg m−3). By applying our empirical model to annual Chl asurf time series, instead of the conventional method assuming vertically homogenous Chl a, we produce novel pan-Arctic PP estimates and associated uncertainties. Our results show that vertical variations in Chl a have a limited impact on annual depth-integrated PP. Small overestimates found when SCMs are shallow (i.e., pre-bloom, post-bloom > 0.7 mg m−3, and the winter period) somehow compensate for the underestimates found when SCMs are deep (i.e., post-bloom < 0.5 mg m−3). SCMs are, however, important seasonal features with a substantial impact on depth-integrated PP estimates, especially when surface nitrate is exhausted in the Arctic Ocean and where highly stratified and oligotrophic conditions prevail.