the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Mar 2025
Formational Conditions of Ribbed Moraine in Norway: A Distribution Analysis and Ribbed Moraine Inventory
Abstract. Ribbed moraines are common landforms in regions formerly glaciated by the Fennoscandian, British and Laurentide ice sheets. Their process of formation is disputed, as formation conditions are hard to reconstruct. With this work, we address this issue through a combination of a comprehensive ribbed moraine inventory of mainland Norway, mapped topographic information, and modelled glacial and hydrological information. First, a detailed 10 metre resolution ribbed moraine dataset is produced for the entire mainland Norway, which is used in combination with spatial statistics to isolate common distributions for ribbed moraines in nine series of data. These values include (a) topographic (elevation, slope and curvature), (b) glacial (basal temperature, ice thickness and ice velocity), and (c) hydrological values (flow accumulation, hydraulic head and hydraulic gradient). We derive mean conditions at 21000 years before present and compare values for areas of ribbed moraines and an equal area of land where ribbed moraines are not present. Our findings show that (a) ribbed moraines typically form in flat, low gradient and low curvature depressions with a low hydraulic gradient, (b) hydraulic head, hydraulic gradient and ice velocity are globally important for ribbed moraine formation, while factors such as elevation and ice thickness are too spatially variable for a wide-scale link to be drawn, but they show a strong local relationship, (c) ribbed moraines are present in areas where ice flow was relatively slow, (d) occurrence of ribbed moraines in areas of high hydraulic head, low hydraulic gradient and low ice velocity suggests that ribbed moraines formed in transitional areas between slow and fast ice flow, which may resemble a “patchwork” of slippery and sticky spots of high and low frictional resistance. However, these relationships are not definite, as we simply note relationships rather than process observations, and as such we conclude the possibility of an “equifinality” theory explanation for the formation of ribbed moraines.
- Preprint
(1426 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-108', Henry Patton, 08 Apr 2025
Barnes et al. analyse a compiled inventory of ribbed moraine mapped across mainland Norway to identify potential commonalities in formation conditions. Various topographic and glaciological parameters are investigated, of which the latter set is derived from previous numerical modelling.
Common settings are found for ribbed moraines both at the regional and global scale in Norway, e.g., in flat, low-gradient depressions or areas of high hydraulic head. Given these various settings, the authors conclude on the likelihood for an equifinality explanation for ribbed moraine formation.
The data and results are presented clearly and are discussed well within the context of recent literature and existing hypotheses. The study is limited by the low spatial and temporal resolution of the palaeo model data employed, with only ice velocity playing a strong role in its value as a predictor. Given the general unavailability of such model data for Fennoscandia, though, this is not necessarily a critique but opens avenues for future research. In the interests of partiality, it should be noted that the model data used comes from a previous reconstruction I published.
I think the study fits well within the scope of ESurf, and presents a useful framework for furthering insight into this enigmatic landform. I provide some specific comments and suggestions below. Aspects in particular that I feel need more clarity are the construction of the ribbed moraine inventory and the limitations of the modelling comparison in relation to temporal variations.
31: And Iceland, for the first time described by Nína Aradóttir et al. 2024
52: There appears some confusion here, for example Linden et al 2008 subscribe to the 'polygenetic' school of thought.
55: And generally linked to compressive ice flow?
90: Check numbering of sections
100: I think you should briefly explain the methodology from the 2024 paper e.g., it's not clear that it's a machine-learning algorithm.
105: If there exists a national database of landforms from Kartverket, I don't immediately follow what the role of the machine-learning approach is? Are there gaps in the Kartverket database? Does your algorithm find ribbed moraines missed by Kartverket? What is the detection accuracy in this study? It sounds like the algorithm could be a useful tool to apply elsewhere too, so some additional information on its performance for creating an inventory at this scale would be relevant for future users.
110: Will this dataset be made available? I think it would be a useful resource for any future workers wanting to build on this work.
130: UiT model is more commonly used reference rather than Patton model.
234: Maybe the Shackleton paper is not the best citation here - there is a lot of literature on the subglacial thermal organisation of ice sheets where this 'patchy' ice resonates e.g., Kleman & Glasser 2007.
256: Define what a murtoo is as it is probably not widely known.
341: While reading the results it may be worth considering using acronyms for the data subsets e.g., southern Norway under 900 m elevation = SN<900, just to make it a bit easier to digest.408: I think this is an important section on the limitations regarding the comparison with model data. But I think there is also a temporal aspect to consider. For example, you are comparing data to a snapshot in deglaciation, and quite early too. It is not realistic to assume all ribbed moraine in Norway are contemporaneous so the glaciological conditions examined are likely to have some inherent uncertainty linked to temporal variations. From other modelling it has been shown that basal conditions can vary significantly over short timescales e.g., Patton 2022. I think you briefly touch upon this on line 227, and it's an interesting point that you see similar trends at 14 ka even with a reduced dataset.
424: To clarify, the figure referred to here shows the potential routing of subglacial drainage, even if no subglacial water was present (check also line 467). Otherwise, I agree that subgrid anomalies are to be expected, particularly in areas of high relief.
525: Misformatted citation.
578: I'm not sure that this section is necessary as it repeats what you have already discussed earlier in the discussion.
600: This should probably be more specific on what pressure and gradients - hydraulic.
609: And evolving glaciological conditions through deglaciation.
616: Why is the flood hypothesis dismissed?
631: It would be useful to say what is needed to move this forward (future directions) e.g., higher-resolution model outputs, a time-transgressive analysis of glacial parameters, examination of other domains...Aradóttir, N., Benediktsson, Í.Ö., Helgadóttir, E.G., Ingólfsson, Ó., Brynjólfsson, S., Farnsworth, W.R., 2025 Ribbed moraines formed during deglaciation of the Icelandic Ice Sheet: implications for ice-stream dynamics. Boreas n/a. https://doi.org/10.1111/bor.12690
Kleman, J., Glasser, N.F., 2007. The subglacial thermal organisation (STO) of ice sheets. Quaternary Science Reviews 26, 585–597. https://doi.org/10.1016/j.quascirev.2006.12.010
Citation: https://doi.org/10.5194/egusphere-2025-108-RC1 -
RC2: 'Comment on egusphere-2025-108', Anonymous Referee #2, 23 Apr 2025
Apologies for my review being late.
This paper presents some interesting observations of subglacial rib occurrence across Norway. However, I find the scope of this paper, and the insights it provides to be rather limited. The authors derive no real insight into rib formation, and the dataset is not particularly useful for constraining rib formation in models.
Major comments
The distribution of ribs and their characteristics was documented by Hättestrand and Kleman in 1999, - how does the database here (Figure 1) improve on this?
I think throughout, the authors set up a straw man, a false dichotomy, between equifinality and equicausality. This conflict has been resolved and shown to be of little use by Fowler (2018), who shows how many models account for a finite amount of physics but infinite amounts of potential history through "equimorphology" essentially resolving this debate.
Furthermore, I think the term "ribbed moraine" is misleading, with many now using "subglacial ribs" as they are not moraines in the traditional end of a glacier sense.
My final point is that the authors use a single time-slice from a model for understanding rib formation. However, we have absolutely no idea whether these conditions are those in which ribs formed. Indeed, I would expect much of the area studied to be cold-based at the last glacial maximum, and thus no subglacial movement would have occurred. This approach is therefore fundamentally flawed, and a better model-data comparison is required. Perhaps deriving flow direction from the ribs and the model, and looking at conditions when the flow direction and the ribs align with with each other might be a good start.
Minor comments
Abstract: Poorly written. essential says "we didn't find any relationships." Also, reflects the false fight between equifinality and equicausality mentioned above.
L44: See comment above about how this dichotomy is false.
L90:Something wrong with the numbering (two section 1s).
L72: Misrepresentation of Ely et al. (2023) - rib formation is it is not linked to ice velocity, but effective pressure at the base. The point of this paper, and indeed the models beforehand from Fannon, Fowler and Hindmarsh, is that an instability occurs which makes linking the formation to ice flow conditions intractable. Thus, the attempt to do so in this paper is perhaps a bit misguided. The Barchyn paper has also been discredited as being physically unrealistic (see Fowler, 2018). I suggest rewriting after reading these papers properly.
L94: How realistic is it to interpolate ice sheet model conditions to this scale? How was this interpolated?
143: Now you say you do every timestep - why didn't you analyse every timestep?
Section 1.3.1 - I don't get why you subdivide at all. It doesn't appear logical to me as stated. Surely this introduces bias into your analysis?
L228 - As mentioned above, I don't think it is reasonable at all to use the 21 ka BP period solely. Why only use these two times when comparing? I am a bit baffled. Of course some ribs will be deglaciated by the latter time slices, but that doesn't mean that the ice flow conditions at 14 for those that were covered don't relate to the formation time. This reads as "we picked 21ka because it was easy, not because it was scientifically correct"
L235 - what is Shackletons theory? I am pretty sure this paper is an analysis of model output, not a theoretical proposal (i.e. one based in maths).
Results - it is interesting that ribs seem to favour some places. But these seem to be basins - presumably just controlled by where sediment is available? For example, Figure 3, you aren't going to get ribs on the really steep slopes, as the sediment would fall off. This also impacts the hydraulic head calculation. As noted above, the model comparison is flawed.
Section 3.1.1 - I agree there are problems with the resolution comparison. One could have offset this by calculating the difference between the model grid elevation, and the DEM elevation of the ribs, then applying this to the basal temperature calculation. The bigger flaw is taking one timeslice, as mentioned above.
L484 - What is glacial status?
L580: There are no velocities in Ely et al., 2023.
Throughout the discussion there is some mention of a "cavity system" - what scale is this? Glaciologists would assume a 1-10 m scale as described by theories such as Kamb and Lliboutry. But it seems the authors are refering to km scale patches. Also, wouldn't the water be held in the sediment and thus mostly flowing via Darcian flow? I think this needs rethinking with sound glaciological basis.
L611 - Both models show stable ribs, and don't account for cold-warm based transitions yet.
Throughout the discussion of different processes is limited and needs thinking about. The instability theory for example does not prohibit shear-stacking from happening. It's just a process of sediment movement to this model.
The conclusion - specifically conclusion F - seems to be that we can't rule out any of the hypotheses, therefore they are all likely correct. This reasoning is flawed. It should be that we can't rule any of the hypotheses, they could all be correct or all could be wrong, we need a better way of testing them.
Additional References:
Fowler, A.C., 2018. The philosopher in the kitchen: the role of mathematical modelling in explaining drumlin formation. Gff, 140(2), pp.93-105
Citation: https://doi.org/10.5194/egusphere-2025-108-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 57 | 24 | 5 | 86 | 8 | 9 |
- HTML: 57
- PDF: 24
- XML: 5
- Total: 86
- BibTeX: 8
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1