the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Oct 2025
Spatiotemporal patterns in CO2 fluxes and geochemical weathering in mountain glacial rivers
Abstract. Despite low temperatures that slow chemical reactions, geochemical weathering can be pronounced in glacial rivers due to large quantities of fresh comminuted sediments (glacial flour). We assessed the types and magnitude of geochemical weathering across multiple seasons and years in three proglacial rivers (Sunwapta-Athabasca, North Saskatchewan, and Bow) on the eastern slopes of the Canadian Rocky Mountains, as they meandered from their alpine glacial origins to the montane altitudinal life zone up to 100 kms downstream. To overcome the inherent ecological complexity of our study region, multiple lines of evidence were used to quantify geochemical weathering along river transects and across seasons. Carbon dioxide (CO2) was highly undersaturated and instantaneous CO2 fluxes mostly net consumptive at sampling sites nearest source glaciers. Basic geochemical parameters and a large suite of isotopes (87Sr/86Sr, δ34S-SO4, δ18O-SO4, δ13C-PIC, δ13C-DIC, and Δ14C-DIC) were used to dissect general trends in weathering geochemistry. These trends were supported by an inversion model and an inorganic-organic carbon mass balance model, which together found that while carbonate weathering dominated at all sampling sites and times, silicate weathering and organic carbon contributions to the dissolved inorganic carbon pool increased with distance downriver of glaciers regardless of season. Globally, we suspect these spatiotemporal patterns in the type and magnitude of geochemical weathering are common across glacierized watersheds. Therefore, as glaciers continue to retreat, we can expect to see an encroachment of downriver altitudinal life zones concurrent with glacier mass loss and an evolution of in-river geochemical weathering processes, with direct implications for present-day regional and global carbon budgets.
- Preprint
(7771 KB) - Metadata XML
-
Supplement
(3180 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-5020', Anonymous Referee #1, 26 May 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-5020/egusphere-2025-5020-RC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2025-5020-RC1 -
RC2: 'Comment on egusphere-2025-5020', Anonymous Referee #2, 17 Jun 2026
This manuscript provides valuable data from a relatively understudied glacial watershed. Focusing on three proglacial rivers in the Canadian Rockies, the study investigates seasonal and interannual variations in meltwater quality and chemical weathering processes. Importantly, it extends the analytical framework beyond the proglacial channels to a downstream distance of 100 km, allowing for an examination of CO2 dynamics along the flow path—a perspective that is commendable for its spatial breadth. The research clearly demonstrates that glacier-induced grinding enhances chemical weathering, positioning glacial meltwater as a notable sink for atmospheric CO2. By integrating river water chemistry, inverse modeling, coupled organic–inorganic carbon models, and multiple isotope tracers, the study offers a comprehensive characterization of weathering processes within the catchment. The findings carry significant implications for understanding chemical weathering in glacier-fed hydrological systems worldwide, and the methodological framework presented here holds strong potential for broader application and further development in diverse cryospheric environments.
Major concerns:
1. The MEANDIR model employed in this study relies on global average endmember values rather than locally measured endmembers. Although the authors acknowledge this limitation (Lines 467–468), the potential uncertainty introduced by the use of non-local endmembers is not sufficiently discussed, particularly in light of the near-equal partitioning of Na⁺ among atmospheric precipitation, evaporite dissolution, carbonate weathering, and silicate weathering (Fig. 7).To better constrain the model outputs and validate the source apportionment, the authors are encouraged to supplement the analysis with endmember values derived from 2–3 locally collected rock or soil samples. Such site-specific constraints would substantially strengthen the robustness of the weathering budget calculations.
2. In the absence of direct in situ velocity measurements, the estimation of FCO2 using hydraulic geometry and empirical rating curves is inherently subject to considerable uncertainty—a common challenge in proglacial and headwater environments. Substantial spatial variability in flow velocity was observed among the glacial river sampling sites, spanning contrasting channel types—including a high-gradient riffle, a plunge pool below a waterfall, a lake outlet, and a low-gradient wetland-influenced reach.
However, the authors applied a single nominal regression (Eq. 5) across all sites without stratifying by reach-scale hydraulic characteristics (e.g., slope, bed substrate, and cross-sectional morphology). This uniform treatment likely leads to systematic bias: at the same discharge (Q), the model assigns identical velocity (V) to both high-gradient torrents and low-gradient slow-flowing reaches, thereby underestimating V at steep sites and overestimating it at quiescent sites. Such aggregation may compromise the accuracy of reach-scale gas exchange estimates and warrants careful consideration.
Authors should revisit the velocity–discharge relationship by grouping sites according to their dominant hydraulic typology, or alternatively, incorporate reach-specific coefficients where feasible. If direct velocity measurements are unavailable, we recommend a sensitivity analysis to evaluate how the assumed velocity–discharge parameterization influences the overall FCO2 budget, and to explicitly discuss the associated uncertainties in the context of inter-site comparisons.
The velocity–discharge relationship could be refined by stratifying sites based on hydraulic typology or, where possible, by incorporating reach-specific coefficients. Given the lack of direct velocity measurements, a sensitivity analysis would be useful to assess how the current parameterization affects FCO2, and the resulting uncertainties merit explicit discussion across sites.
3. Beyond error and sensitivity analyses for FCO2 estimates, an alternative approach is to bypass the velocity-to-K conversion by using a mass balance model to estimate the FCO2 exchange potential of glacier-fed runoff, as demonstrated in recent studies (Kleber et al., 2024; Ragnoli et al., 2025). Such methods may provide independent constraints and reduce uncertainties inherent in hydraulic parameterization.
[Kleber G, Magerl L, Turchyn A, et al. Proglacial methane emissions driven by meltwater and groundwater flushing in a high Arctic glacial catchment[J]. Biogeosciences, 2024, 22: 659-674.
Ragnoli, M D, Martini J, Jechsmayr B, et al. Probing dissolved CO2 and CH4 in glacial streams of the European Alps[J]. Arctic, Antarctic, and Alpine Research, 2025, 57(1): 2580737.]
4. The comparison with other glacial catchments worldwide remains insufficient. Only a limited number of sites (e.g., Rhône, Greenland) are referenced, whereas systematic comparisons with typical glacierized basins in the Himalayas, the European Alps, and the Andes—in terms of weathering styles and fluxes—are lacking.
A comparative table summarizing weathering types and fluxes across global glacial catchments would be valuable, which could help highlight the distinctiveness of the present study, particularly its multi‑seasonal sampling and longitudinal gradient design.
5. A comparative analysis with existing studies on greenhouse gas dynamics in glacier-fed rivers is currently lacking. Numerous investigations have been conducted worldwide on the concentrations and diffusive fluxes of CO2 and CH4 in glacial meltwater streams, including those from the Greenland Ice Sheet, the Himalayan glaciers, the European Alps, and Alaska.
It would be beneficial to expand the discussions in Sections 3.1.1 and 3.1.2 by including a systematic comparison of CO2 concentrations and fluxes observed in the present study with those reported from other major glacial regions. Such a comparison would help highlight the distinctive CO2 dynamics of the studied glacial meltwater—encompassing concentration magnitudes, flux directions and amplitudes—and facilitate a broader discussion of the underlying controlling factors and mechanisms.
Minor Comments
1.Section 2.3 (Sample collection and analyses) is overly detailed in its current form. A streamlined version highlighting the key methodological approaches would improve readability, while supplementary details could be moved to the Appendix.
2.In Section 3.1.1, dissolved CO2 is discussed primarily in terms of percent saturation (Line 558). However, the more direct and interpretable metric would be the aqueous CO2 concentration (e.g., μM). It is suggested that concentrations be used as the primary variable, with the atmospheric equilibrium concentration also plotted in the figures to facilitate clearer visual interpretation.
3.The presentation of data in Figure 3 (L577–579) is not particularly reader‑friendly. It is suggested that the figure be restructured to display boxplots that distinguish among seasons, glacial catchments, and distance from the glacier terminus.
4. In Section 3.1.2 (L586), the units for FCO₂ should be expressed as –149 to 56 mg CO₂ m⁻² d⁻¹ to match the values shown in the figure. The statements in L587–592, particularly the comparison that "the most negative flux was nearly 12 times more CO₂ consumptive than tropical forests and a whopping 166 times more CO₂ consumptive than the tundra," require careful verification. Correspondingly, the relevant discussion in the Conclusions (L945–948) should also be reviewed and revised accordingly.
5. At L609–610, Table S4 is cited in support of seasonal contrasts in instantaneous CO₂ fluxes, but the required data are not presented in the table. Correction or supplementation is advised
6. In Figure 10b, silicate contributions show an increasing trend with distance from the glacier, but the correlation is extremely weak (R² = 0.109). Although the p-value is significant (p < 0.001), this significance is likely driven by the sample size rather than a meaningful relationship. In addition, it is suggested that for Figure 10a, the correlation statistics (R² and p-value) for the relationship between % of total DIC (from carbonate and atmospheric sources, and from organic matter) and distance from the glacier be included, so as to better quantify the downstream evolution of DIC sources.
Citation: https://doi.org/10.5194/egusphere-2025-5020-RC2
Data sets
Weathering dataset collected from climate-threatened glacial river headwaters on the eastern slopes of the Canadian Rocky Mountains (2019-2021) [dataset] J. A. Serbu et al. https://doi.pangaea.de/10.1594/PANGAEA.972842
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,611 | 1,400 | 147 | 3,158 | 456 | 143 | 114 |
- HTML: 1,611
- PDF: 1,400
- XML: 147
- Total: 3,158
- Supplement: 456
- BibTeX: 143
- EndNote: 114
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1