Transport and reaction of iron and iron stable isotopes in glacial meltwaters on Svalbard near Kongsfjorden: From rivers to estuary to ocean

Abstract

Glacial meltwater has been suggested as a significant source of potentially bioavailable iron to the oceans. However, the supply of dissolved iron (dFe) in glacial meltwaters is poorly constrained as few sites have been studied, and because the chemical processing of Fe during transport from glaciers to the adjacent coastal ocean is not well understood. In order to better constrain glacial fluxes of dFe to the ocean, iron concentrations, iron stable isotopes (δ 56 Fe), and other supporting chemical and physical measurements were made along a ~4 km long glacial meltwater river on Svalbard and in estuarine waters that it flows into. Dissolved iron concentrations in the Bayelva River decreased from a maximum of 734 nM near the glacier to an average value of 116 nM near the mouth of the river. Measurements in the Kongsfjorden estuary suggest that 3 to 10 nM of dFe from the Bayelva River is stabilized in glacial waters by the time it mixes into the ocean. Incubation of Bayelva River waters over two weeks in both the light and dark show similar results, with the majority of dFe being quickly precipitated and 4 to 7 nM Fe stabilized in the dissolved phase. Evidence suggests that Fe is most likely lost from the dissolved phase by aggregation and adsorption of nanoparticulate and colloidal Fe to particles. Dissolved δ 56 Fe was between -0.11‰ and +0.09‰ for all river samples and did not vary systematically with dFe concentrations. We infer that the Fe is lost from the dissolved phase by a process that fractionates Fe isotopes by less than 0.05‰, indicating that the Fe bonding environment does not change during precipitation. This is consistent with DOC loss that is much faster than predicted photo-oxidation rates, suggesting that DOC is also lost through adsorption and precipitation. Dissolved Fe concentrations in the Bayelva River (15-734 nM), and Fe concentrations which are stabilized in the dissolved phase (4-7 nM) are much lower than some previous estimates of Fe in glacial meltwaters, with roughly 80 % of dFe lost during transit in the Bayelva River and roughly 90% of the remaining dFe lost in the estuary. This may mean that glaciers are a less significant source of dissolved Fe to the global oceans than has been previously hypothesized, that cold base glaciers of the type studied here do not contribute significantly to the dissolved Fe flux, or that the flux of reactive particulate Fe to the oceans is more important than the dissolved flux. In Arctic regions with similar proglacial environments, bedrock composition, weathering intensity, and as precipitation of colloidal and nanoparticulate Fe may all play an important role in regulating the glacial meltwater iron flux to the ocean.

The iron isotopic composition of subglacial streams draining the Greenland ice sheet

“…2), a feature that is also observed in glacially fed rivers such as the Copper River in Alaska Escoube et al, 2015). This observation agrees with previous d 56 Fe measurements in glacial outflow from Bayelva River near Kongsfjorden Svalbard (500 m downstream of the terminus; (Zhang et al, 2015). The measured d…”

“…As has been demonstrated for other Fe sources to the North Atlantic (Conway and John, 2014;Zhang et al, 2015), constraining the Fe isotopic composition of glacial melt waters (expressed as d 56 Fe or d 57 Fe in the literature) may facilitate the tracing of glacial Fe in proximal oceans. Such constraints are not straightforward to obtain, as there is reason to suspect that the Fe isotopic composition of subglacial streams may vary considerably.…”

“…Likewise, non-redox chemical reactions involving Fe have also been associated with large isotopic effects (e.g. Anbar et al, 2000;Skulan et al, 2002;Welch et al, 2003;Thompson et al, 2007;Kiczka et al, 2010;Zhang et al, 2015). For example, isotopic effects as large as 4‰ have been associated with organic-Fe complexation (e.g.…”

Seasonal Changes in Fe along a Glaciated Greenlandic Fjord

“…Whilst the size fractions used to define Fe phases vary between studies, our concentrations are within the range reported for freshwater elsewhere around Greenland (Statham et al, 2008;Bhatia et al, 2013b;Hawkings et al, 2014). The non-conservative decline in DFe along estuaries has been well characterized for many river systems worldwide (Boyle et al, 1977;Sholkovitz, 1978), including some in glaciated catchments (Schroth et al, 2014;Zhang et al, 2015). The dominant feature of DFe and TdFe distribution in this fjord however appears to be the particle plume by GF11 which causes some deviation from the expected decline in DFe concentration with increasing salinity (Figure 7).…”

“…Fe behavior in the glacially fed Arctic River Bayelva supports this hypothesis. Eighty percent of dissolved Fe (<0.40 µm) was found to aggregate along a 4 km stretch of river prior to additional removal during estuarine mixing (Zhang et al, 2015). This resulted in a final dissolved Fe removal of around 98% between a glacial outflow and the sea, which is at the upper end of the range observed in temperate estuarine systems (Boyle et al, 1977;Sholkovitz, 1978).…”