Evaluating the Biogeochemistry of Dissolved Inorganic Carbon in the Canadian Arctic Archipelago Using Stable and Radiocarbon Isotopes

Abstract

Human activity is unequivocally contributing to climate change and global warming (IPCC 2023). The oceans act to moderate climate by removing CO₂ (a greenhouse gas) from the atmosphere via air-sea gas exchange and sequestration as dissolved inorganic carbon (DIC). DIC is the largest actively cycling pool of carbon on Earth (38,000 Gt) and approximately half of all anthropogenic carbon emissions have already been absorbed by the world's oceans (IPCC 2023). Despite this, the Arctic is still warming at rates up to four times faster than the global average (Rantanen et al., 2022). Research suggests the Canadian Arctic Archipelago (CAA) is an important region for the marine carbon cycle (e.g. Papakyriakou & Miller, 2011; Zeidan et al., 2022) but its response to climate change remains poorly constrained. The CAA connects the Pacific, Arctic, and Atlantic Oceans. Melting permafrost, sea ice, changing biogeochemistry and increased freshwater flux within have the potential to impact local and global processes, such as deepwater formation in the North Atlantic and global thermohaline circulation (e.g. Serreze et al., 2006). Stable carbon (δ¹³C) and radiocarbon (Δ¹⁴C) in marine DIC are powerful tools for determining carbon sources and ageing. These isotopes can be used as tracers for processes such as water mass advection, biogeochemical modification, and anthropogenic carbon penetration. In this thesis, we present n = 151 DIC samples from the CAA (Beaufort Sea to Lancaster Sound) combined with n = 110 samples from Baffin Bay sampled by Zeidan and coworkers (2022) to make broad, pan-arctic carbon cycling interpretations. We offer the most complete, pan-arctic DIC isotope dataset available.

CAA DIC concentrations, δ¹³C and Δ¹⁴C values ranged between 1.79 to 2.65 mmol L⁻¹, -0.68 to +1.86 ‰, and -90.7 to +49.5 ‰ respectively. A general trend from West to East of decreasing DIC concentrations, decreasing δ¹³C values and increasing Δ¹⁴C was observed in advecting Pacific water masses throughout the CAA. These isotopic trends suggest heterotrophic remineralization of "bomb" carbon-containing particulate organic matter from riverine influx is translated to the marine DIC pool in the CAA. The shallow, fresh Kitikmeot region is especially dominated by riverine influx, leading to low DIC concentrations (0.10 to 2.20 mmol L⁻¹) and positive Δ¹⁴C (+5.0 to +6.8 ‰) with positive δ¹³C (+0.30 to +0.99 ‰) from additional physical mixing processes unique to the region. In contrast, highly negative surface DIC Δ¹⁴C values of -44.7 ‰ and -51.9 ‰ near the Mackenzie River suggest thawed permafrost carbon from this great Arctic river's expansive watershed is incorporated into marine DIC in the Beaufort Sea. Further North, we examine Parry Channel, an important region between M'Clure Strait and Lancaster Sound through which all CAA waters must flow before entering Baffin Bay. We hypothesize that our observed enrichment of both δ¹³C and Δ¹⁴C across the channel is either indicative of two distinct biogeochemical provinces separated by the shallow (~125 m) Barrow Strait, or that underlying Atlantic waters are mixed vertically into shallower Pacific waters by the local bathymetry (as interpreted by Lehmann et al., 2022). In the Beaufort Sea, we use nearby historical data to propose anthropogenic carbon concentrations have increased in Canadian Arctic water masses since 2009, reaching new depths as low as 1500 m, but that rates of uptake are decreasing by as much as eight times less than in 2012. This suggests the ocean's natural sink for CO₂ is weakening significantly on decadal timescales. Lastly, the stark contrast in DIC δ¹³C and Δ¹⁴C values between the Beaufort Sea and Baffin Bay suggests the two basins are characterized by different water mass endmembers (Pacific and Atlantic origins, respectively), with deep water DIC Δ¹⁴C in Baffin Bay being significantly older than that of the Canada Basin.