Simultaneous measurement of O2, Mn, Fe, I−, and S(—II) in marine pore waterswith a solid‐state voltammetric microelectrode

Abstract

A solid‐state Au/Hg voltammetric microelectrode was used to measure, simultaneously and with millimeter spatial resolution, the vertical distributions of O2, Mn2+, Fe2+, HS−, and I− in the pore water of sediments from the Canadian continental shelf and slope. The electrode was used shipboard to analyze undisturbed sediment cores and in a sediment mesocosm to determine the three‐dimensional distribution of redox species in the sediment surrounding an actively irrigated worm burrow. In cores from Emerald Basin and Cabot Strait, O2 disappeared and Mn2+, Fe2+, and HS− appeared in the vertical sequence predicted on the basis of thermodynamics. In a core from the Scotian slope, O2 disappeared at ~14‐mm depth but Mn2+ was not detected over the 50‐mm depth examined with the probe. In all three cores, I− was detected in the pore water below the oxygen‐penetration depth. In none of the cores did the distributions of O2 and Mn2+ overlap. The thickness of the layer within which neither O2 nor Mn2+ could be detected ranged from 8 to 40 mm; we suggest that nitrate, not oxygen, is used to oxidize Mn2+. The upwarddirected Mn2+ gradients of 3.5 and 10 mol mm4 in the Emerald Basin and Cabot Strait sediments, respectively, drive manganese fluxes that are more than one order of magnitude lower than the O2 fluxes, which are estimated at 4‐S mmol mm−2 d−1. The three‐dimensional distributions of solutes in the pore water in the sediment surrounding an irrigated worm burrow demonstrate that both the O2‐penetration depth and the depth where Mn2+ is first detected are deeper in the vicinity of the worm burrow than away from it. This is consistent with the notion that irrigation brings oxygenated water into the worm tube, allowing oxygen to diffuse across the burrow wall into the sediment and react with Mn2+ and other reduced porewater constituents. Three‐dimensional heterogeneity in porewater composition cannot be detected by conventional one‐dimensional coring and slicing techniques. Because these latter techniques average the porewater composition in a large volume of sediment, they may indicate that the distributions of individual species overlap when in fact they do not.