Silicon isotopes in Arctic and sub-Arctic glacial meltwaters: the role of subglacial weathering in the silicon cycle

Abstract

Glacial environments play an important role in high-latitude marine nutrient cycling, potentially contributing significant fluxes of silicon (Si) to the polar oceans, either as dissolved silicon (DSi) or as dissolvable amorphous silica (ASi). Silicon is a key nutrient in promoting marine primary productivity, contributing to atmospheric CO₂ removal. We present the current understanding of Si cycling in glacial systems, focusing on the Si isotope (δ³⁰Si) composition of glacial meltwaters. We combine existing glacial δ³⁰Si data with new measurements from 20 sub-Arctic glaciers, showing that glacial meltwaters consistently export isotopically light DSi compared with non-glacial rivers (+0.16‰ versus +1.38‰). Glacial δ³⁰Si_ASi composition ranges from -0.05‰ to -0.86‰ but exhibits low seasonal variability. Silicon fluxes and δ³⁰Si composition from glacial systems are not commonly included in global Si budgets and isotopic mass balance calculations at present. We discuss outstanding questions, including the formation mechanism of ASi and the export of glacial nutrients from fjords. Finally, we provide a contextual framework for the recent advances in our understanding of subglacial Si cycling and highlight critical research avenues for assessing potential future changes in these environments.

Keywords: glaciers and ice sheets; silicon cycle; silicon isotopes; subglacial weathering.

Figure 1.

Conceptual model tracing subglacial processes impacting upon dissolved silicon isotope composition (δ³⁰Si_DSi) and silicon isotope composition in amorphous silica (δ³⁰Si_ASi). The hypothesized change in δ³⁰Si composition as a result of each subglacial processes is denoted by red arrows. The pH of the environment is high as a result of alkalinity from silicate hydrolysis reactions. (Online version in colour.)

Figure 2.

Location map of glacial rivers analysed for the silicon isotope (δ³⁰Si) composition for dissolved (DSi) and amorphous (ASi) phases. The proglacial rivers of the glaciers highlighted in red in the five smaller maps were sampled for δ³⁰Si composition. Clockwise from lower left; Iceland, Qeqertarsuaq (Disko Island), southeast Alaska, Svalbard and Norway. Graphs show the δ³⁰Si_DSi (black) and δ³⁰Si_ASi (red) composition measured for each glacier, organized by location. Error bars, mainly included in the symbol size, represent average external errors (0.08‰, 2 s.d.). Glaciers are organized by catchment size on each graph (acronyms defined in electronic supplementary material, table S2). The horizontal grey line represents the average non-glacial δ³⁰Si_DSi composition compiled from published data (see figure 3 for references). (Online version in colour.)

Figure 3.

Box plot showing the global average dissolved silicon isotope (δ³⁰Si_DSi) composition from glacial and non-glacial rivers. Data compiled from existing published sources [44,49,50], from this study for glacial rivers (+0.16‰) and from published sources for non-glacial rivers (+1.38‰) [,–,–,–118]. Boxes represent upper and lower quartiles, with the median represented by the middle line. (Online version in colour.)

Figure 4.

Relationship between suspended particulate material (SPM) concentrations and the dissolved silicon isotope (δ³⁰Si_DSi) composition of glacial meltwaters. Data are reported from glaciers measured in this study, Greenland [50,58] and Iceland [44]. Data have been trimmed to remove outliers (95th percentile), and the regression line has an R² value of 0.2542. Error bars on data from this study represent the average external error of 0.08‰ (2 s.d.), based on triplicate measurements of a subset of samples. (Online version in colour.)

Figure 5.

Relationships between catchment size and amorphous Si (ASi, wt.%) and the silicon isotope (δ³⁰Si_ASi) composition in suspended particulate matter of the new glacial meltwaters analysed in this study and from Greenland [58]. Error bars represent an average external error of 0.08‰ (2 s.d.), based on triplicate measurements of standards and a subset of samples, except for the Greenlandic samples in (b). These datapoints represent the discharge-weighted mean δ³⁰Si_ASi composition from measurements over the ablation season and error bars represent the range of values measured over the season. (Online version in colour.)