the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Organic and inorganic proxy development for advective heat flux in a basal-melt-dominated glacial system: Petermann Fjord, North Greenland
Abstract. Mass loss from the Greenland Ice Sheet directly impacts sea level rise and the freshwater balance of the ocean. Petermann Glacier, North Greenland, provides a unique setting to explore atmosphere-ocean-ice interactions in a basal melt–dominated system using proxy records. We analyse organic (sea ice-associated highly branched isoprenoids (HBIs) and open water phytoplankton sterols) and inorganic (10Be, 230Th) proxies from five sediment cores spanning the inner to outer Petermann Fjord to investigate oceanic advection and glacial dynamics, during times when the ice shelf occupied extended (after 0.6 ka BP) and reduced (pre-0.6 ka BP) positions. Results show higher fluxes of 10Be, HBIs, and sterols during the reduced ice shelf phase. However, 10Be, HBIs, and sterols are also present in sediments deposited when the ice shelf was extended, suggesting lateral advection and glacial meltwater contribute to their deposition. Whilst 10Be has previously been applied as a proxy for ice shelves and perennial sea ice, in Petermann Fjord our results indicate that lateral advection, enhanced by buoyant subglacial melt, is the dominant source of 10Be, with glacial melt providing secondary input. Our study highlights the potential to use a multi-proxy approach to reconstruct circulation regimes and advective heat flux patterns in basal-melt-dominated environments.
- Preprint
(2696 KB) - Metadata XML
-
Supplement
(1856 KB) - BibTeX
- EndNote
Status: open (until 23 Nov 2026)
- RC1: 'Comment on egusphere-2026-5321', Anonymous Referee #1, 13 Nov 2026 reply
-
RC2: 'Comment on egusphere-2026-5321', Anonymous Referee #2, 17 Nov 2026
reply
The manuscript by Davies et al. is well-written and both puts forward and tests intriguing hypotheses for how combinations of organic and inorganic proxies can be used to reconstruct advective water and possibly heat fluxes into glacial fjords from sedimentary proxy reconstructions. I found the discussion of the concentrations of organic proxies and 10Be to likely be robust, though the 10Be budget needs to be fixed, the general nature of their findings is likely to hold. However, there was severe lack of evaluation of a key proxy used in their study to determine sedimentary mass accumulation rates, 230Th, in an oceanographic and sedimentary environment where the application of 230Th as a constant flux proxy (CFP) is enormously complex. In addition, uncertainties were not routinely discussed throughout the paper and in figures, which made it difficult to discern whether the trends the authors reconstructed were robust.
My main concern with the manuscript is the lack of key evaluation or detail regarding 230Th-normalization, which is used to calculate sedimentary MARs. The sedimentary environment the authors are working in is not conducive at all to 230Th-normalization. It faces, as far as I can tell, five main issues:
1) The study is in a fjord, where salinities may deviate significantly from average seawater and alter the U concentration, and therefore the 230Th production rate. Salinities for Petermann Fjord were not discussed anywhere in this study.
1a) Regardless of salinity, U in the subglacial melt may alter U concentrations from the U-salinity relationship and generate different 230Th production rates on account of the variable U concentrations.
2) The cores are shallow, so there will be relatively less excess 230Th than deep cores.
3) The accumulation rates (50 cm/kyr) are extremely high relative to what is typically used in 230Th normalization in pelagic sediments, meaning each sample analyzed will have relatively less excess 230Th (whose concentration is inversely proportional to accumulation rate).
4) The bulk of the sediment accumulation is likely to be lithogenics, which carry non-excess 230Th, meaning there is a substantial input rate of non-excess 230Th that will drown out excess 230Th.
5) Because of the combination of 2), 3), and 4), the lithogenic U/Th ratio and its uncertainty will be critically important.
The authors acknowledge in section 3.4.2. that the errors in applying 230Th-normalization in the Petermann fjord environment will be large. But, they never quantify it, nor do they address the litany of potential issues with 230Th normalization here, nor is the U-series data available for review to determine how large the errors are. My suspicion is that even with a moderate uncertainty on their lithogenic U/Th ratio, the excess 230Th values may be within error of 0.
The MARs determined using 230Th normalization were also not available or presented anywhere during the review, making it difficult to evaluate whether the values were realistic.
I attempted to back calculate the MARs at their sites by combining the concentrations of parameters in Figure S1 with the fluxes in Figure 3. I got something like 2-4 g/cm2/kyr for the mass accumulation rates across the core sites and intervals. If you take the authors’ stated 50 cm/kyr linear sedimentation rates for these cores, assuming a dry bulk density of ~1 g/cm3, the stratigraphic MARs for these cores is 50 g/cm2/kyr – a factor of ~20 than the 230Th-based estimates. It’s not impossible that there are focusing factors of 20 in Petermann Fjord, it’s also not impossible that the lithogenic U/Th ratio in this region is higher than 0.45, and it is difficult to evaluate the plausibility of either scenario without knowing the uncertainties on the excess 230Th and associated MAR calculations.
A secondary significant concern I have with the manuscript regards the authors’ 10Be model described in the supplement. The 10Be fluxes shown in Figure 3 are 1-5 x 108 atoms/cm2/kyr. In the supplement, they quote the atmospheric deposition flux in this region to be 5.23 x 105 atoms/cm2/year, which is equal to 5.23 x 108 atoms/cm2/kyr – equal to or slightly above the 10Be fluxes they calculate for their cores. Why are other sources needed to be invoked to generate such large 10Be fluxes if their sedimentary fluxes are nearly equal to the atmospheric fluxes? Of course, the stratigraphic MARs would give 10Be burial rates far in excess of their atmospheric fluxes..
There are also errors in their lateral advection terms that need to be fixed. They quote the density of AW to be 27.9 kg/m3 – this is probably the potential density anomaly? The density of this water must be 1027.9 kg/m3. No citation is given for the input rate of AW to the fjord, but the number they quote (3 Sv!!) is impossibly large. It implies a residence time for the whole Petermann Fjord of ~3-5 days, which cannot be the case. The authors must revisit and correct their model before publication.
Line by line and smaller comments:
Lines 53-54: These statements need citations.
Section 2.1: Salinities of the waters in Petermann Fjord need to be provided.
Line 253: More discussion of the lithogenic U/Th activity ratio is needed, and error on this must be propagated into their final excess 230Th and MAR values.
Figure 3: No errors are given on any of these fluxes! It is impossible to interpret the authors’ findings in terms of spatio-temporal trends without understanding the errors on the fluxes.
Lines 290-295: It is not clear that it has to be water column transport rather than sedimentary focusing of biomarkers that is responsible for this trend.
Line 344: Change amplify to amplified
Lines 376-384: This paragraph feels like it belongs in the conclusion.
Citation: https://doi.org/10.5194/egusphere-2026-5321-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 307 | 142 | 33 | 482 | 80 | 37 | 38 |
- HTML: 307
- PDF: 142
- XML: 33
- Total: 482
- Supplement: 80
- BibTeX: 37
- EndNote: 38
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review Comments
This manuscript presents an interesting multi-proxy approach to investigating ocean circulation and ice-shelf dynamics in Petermann Fjord. The manuscript is very well written and clearly structured, and the authors demonstrate a strong ability to communicate complex scientific concepts. However, I have several concerns regarding the limited dataset and the interpretation of the proxy records.
1. Limited dataset
The biomarker dataset is very limited, consisting of only ten samples from five sediment cores, with one sample representing each lithological unit. Such a small dataset raises concerns about the robustness and reliability of the observed patterns and the conclusions drawn from them. It is questionable whether these limited measurements provide sufficient evidence to support the proposed interpretations.
2. Lateral transport of biomarkers
The authors interpret the presence of IP25 and sterols beneath the extended ice shelf as evidence for lateral advection within the water column. While this is plausible, other processes, particularly sediment resuspension, redistribution, and redeposition, could also contribute to the observed biomarker signals. The authors should discuss these alternative transport mechanisms and clarify whether the available evidence can distinguish between them. The interpretation of biomarker fluxes in terms of ocean circulation and advective heat flux should therefore be more cautious.
3. Interpretation of β-sitosterol
In this study, β-sitosterol is classified as an open-water phytoplankton biomarker. However, β-sitosterol is commonly associated with terrestrial organic matter input. I also note that Detlef et al. (2021) interpreted β-sitosterol as an indicator of terrestrial input in the same fjord. Treating β-sitosterol primarily as an open-water productivity indicator without further justification may be misleading. I suggest that the authors reconsider its interpretation and discuss potential terrestrial contributions.
4. Interpretation and calculation of ¹⁰Be transport
The authors estimate the amount of ¹⁰Be transported into the fjord by seawater. However, not all of this ¹⁰Be will necessarily be deposited on the seafloor, as some may remain in the water column or be transported out of the fjord. The authors should clarify how these estimates support their conclusion that lateral advection is the dominant source of ¹⁰Be in the sediments.
In addition, the Atlantic Water density of 27.9 kg/m³ used in the Supplement appears to represent a potential density anomaly rather than the actual seawater density. The authors should verify this value and reassess the corresponding calculations.
5. Uncertainties in ²³⁰Th normalization
The authors acknowledge the uncertainties associated with applying ²³⁰Th normalization in fjord environments characterized by strong freshwater input and lateral transport. However, it would be helpful to further clarify how these uncertainties may affect the calculated proxy fluxes and the resulting interpretations.
6. Comparison with existing datasets
Given the limited biomarker dataset, a comparison with existing larger or higher-resolution records from Petermann Fjord and nearby regions would be valuable. This could help place the observed patterns in a broader context and assess their consistency with previous findings, although such comparisons would not fully compensate for the limited number of samples analysed in this study.