Macronutrient and carbon supply, uptake and cycling across the Antarctic Peninsula shelf during summer

Abstract

The West Antarctic Peninsula shelf is a region of high seasonal primary production which supports a large and productive food web, where macronutrients and inorganic carbon are sourced primarily from intrusions of warm saline Circumpolar Deep Water. We examined the cross-shelf modification of this water mass during mid-summer 2015 to understand the supply of nutrients and carbon to the productive surface ocean, and their subsequent uptake and cycling. We show that nitrate, phosphate, silicic acid and inorganic carbon are progressively enriched in subsurface waters across the shelf, contrary to cross-shelf reductions in heat, salinity and density. We use nutrient stoichiometric and isotopic approaches to invoke remineralization of organic matter, including nitrification below the euphotic surface layer, and dissolution of biogenic silica in deeper waters and potentially shelf sediment porewaters, as the primary drivers of cross-shelf enrichments. Regenerated nitrate and phosphate account for a significant proportion of the total pools of these nutrients in the upper ocean, with implications for the seasonal carbon sink. Understanding nutrient and carbon dynamics in this region now will inform predictions of future biogeochemical changes in the context of substantial variability and ongoing changes in the physical environment.This article is part of the theme issue 'The marine system of the West Antarctic Peninsula: status and strategy for progress in a region of rapid change'.

Keywords: Antarctic Peninsula; Circumpolar Deep Water; carbon cycling; nitrate isotopes; nitrogen cycle; nutrients.

Figure 1.

Map of the study area showing all 11 stations and the locations of Marguerite Trough, Marguerite Bay, Ryder Bay and the WAP mainland. Grey shading depicts bathymetry, according to the colour bar shown. (Online version in colour.)

Figure 2.

Section plots of (a) temperature, (b) salinity and (c) δ¹⁸O and derived fractions of (d) meteoric water and (e) sea ice meltwater along the transect from T01–T10. CH1 is excluded due to its topographic isolation from CDW.

Figure 3.

Section plots of concentrations of (a) nitrate, (b) phosphate, (c) silicic acid, (d) DIC and (e) nitrite and (f) N* and (g) Si* along the transect as for figure 2.

Figure 4.

Depth profile plots of (a) [NO₃⁻ + NO₂⁻], (b) and (c) for all stations, as per legend.

Figure 5.

Depth profile plots of (a) DIC concentration, (b) pCO₂ and (c) pH for all stations, as per legend.

Figure 6.

Depth profile plots of (a) weight percent POC, (b) chlorophyll, (c) weight percent PN and (d) δ¹⁵N_PN for all stations, as per legend, which applies to all plots. Note different y-axis scale for d.

Figure 7.

Plots of (a) nitrate versus phosphate, (b) silicic acid versus nitrate, (c) salinity versus nitrate and (d) DIC versus nitrate for all stations. Note different colour shading and legends; stations for plots (a–c) as per the legend next to (b), temperature for (d). In (a), the dashed line depicts uptake according to the Redfield ratio (16 : 1); the solid line is the linear regression for our data with a [NO₃⁻]/[PO₄³⁻] uptake ratio of 14.6 ± 0.2. In (b), the dashed line depicts the linear regression for the upper 40 m where biological uptake occurs, with a [Si(OH)₄⁻]/[NO₃⁻] uptake ratio of 1.0 ± 0.1. In (c), the dashed line shows a mixing trend between the high-nitrate high-salinity subsurface waters and the upper ocean. In (d), the dashed line shows the linear regression from the surface to the Winter Water layer.

Figure 8.

Plots of (a) versus ln[NO₃⁻], (b) versus ln[NO₃⁻] and (c) versus , and (d) a depth profile plot of Δ15–18 for all stations, as per legend. In (a) and (b), black lines depict the modelled relationships using ε values of 4‰ (solid) and 5‰ (dashed); grey lines depict the standard errors of modelled values. In (c), the dashed line depicts a 1 : 1 enrichment ratio of  : . Error bars for show the uncertainty associated with the correction for nitrite interference, which arises from the range of values measured in the Southern Ocean [64]. Error bars for depict analytical error. In (d), dashed grey lines depict the expected range of Δ15–18 in CDW.