Characterising hillslope–stream connectivity with a joint  event analysis of stream and groundwater levels

Beiter, Daniel; Weiler, Markus; Blume, Theresa

doi:https://doi.org/10.5194/hess-24-5713-2020

Articles | Volume 24, issue 12

https://doi.org/10.5194/hess-24-5713-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Linking landscape organisation and hydrological functioning:...

https://doi.org/10.5194/hess-24-5713-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 24, issue 12

Research article

|

30 Nov 2020

Research article |

| 30 Nov 2020

Characterising hillslope–stream connectivity with a joint event analysis of stream and groundwater levels

Daniel Beiter, Markus Weiler, and Theresa Blume

Download

Final revised paper (published on 30 Nov 2020)
Preprint (discussion started on 03 Feb 2020)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'review', Anonymous Referee #1, 06 Mar 2020
- AC4: 'Response to Anonymous Referee #1', Daniel Beiter, 28 Apr 2020
RC2: 'Review', Anonymous Referee #2, 06 Mar 2020
- AC2: 'Response to Anonymous Referee #2', Daniel Beiter, 28 Apr 2020
RC3: 'Review', Anonymous Referee #3, 08 Mar 2020
- AC3: 'Response to Anonymous Referee #3', Daniel Beiter, 28 Apr 2020
AC1: 'Response to Anonymous Referee #1', Daniel Beiter, 28 Apr 2020

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Publish subject to revisions (further review by editor and referees) (07 May 2020) by Hjalmar Laudon

AR by Daniel Beiter on behalf of the Authors (18 Jun 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (29 Jun 2020) by Hjalmar Laudon

RR by Anonymous Referee #1 (28 Jul 2020)

RR by Anonymous Referee #4 (06 Aug 2020)

Suggestions for revision or reasons for rejection

In this manuscript, the authors analyze a multi-year time series of shallow groundwater levels in 15 wells and stage at 5 stream locations in two catchments. They identified events throughout the study period and compared groundwater responses to stream response in several ways to determine how often hillslopes connected to the streams, whether or not connectivity depended on underlying geology, and if monitoring wells at a footslope can be informative of catchment connectivity. In their conclusions they describe how the mechanisms by which footslopes connect to the stream are likely different between the two geologies in the study (fill and spill vs.transmissivity feedback), while the frequency of connection didn’t vary significantly. Furthermore they conclude that a single well in a footslope in a catchment can describe catchment connectivity, and that many years of data are required to adequately describe the processes they investigated.

The data analysis and presentation of the results in the paper were very well done. I think this study is appropriate for HESS and will be of interest to its readers. However, I think some of the conclusions are overstated in their present form (detailed below). I think the reasoning behind a couple of the conclusions must be more clearly stated, or they should probably be taken out. I think there is still plenty of interest even if a couple of the conclusions were either hedged more clearly or taken out.

The first finding that I think needs work is that this study illustrates that you need many years of data to make the conclusions made in the study. I certainly don’t disagree! I think this is probably true. However, I don’t see in the paper where this is demonstrated quantitatively. I think the authors would need to show how the conclusions would differ if you showed only one year, compared to the entire time series. This might be accomplished by a version of figure 8, 9, or 10 that highlights one year of the data and how that differs from the total time period of the study. However, I wonder if the authors need to make this conclusion. I think there are plenty of other conclusions in the paper without having to make this one, which might necessitate more figures and analysis in an already figure-heavy paper.

The second finding that I struggled with is the notion that one piezometer can indicate the connectivity of the entire catchment. I agree with the authors that there are wells in their analysis that are tightly related to streamflow response, but I think the conclusions that result should maybe be a bit more nuanced. To me, the wells that are related to the stream response are likely somehow indicative of the storage state of the catchment. That could be because they are in a position where they are representing hillslope connectivity, as posited in the manuscript, but it could also be explained by other processes. How do we know the well isn’t just in a spot that is in the hyporheic zone and that is why it is so well linked to stream behavior? Or the well might response might be governed by deeper flowpaths that are more indicative of overall catchment wetness than the hillslopes, and that is why the relationship is so good? I think it would be good to mention other possibilities, and also to be careful saying that these wells represent hillslope connectivity. The spatial extent of the wells in the study was not that large, and the wells didn’t go very far up the hillslopes, so I don’t think the authors can say whether or not they were indicative of what was going on up the hillslope. All that being said, I think the authors’ analysis shows that some near stream wells CAN be good indicators of catchment wetness and be predictive of how the stream will respond to precipitation, which is definitely an interesting finding!

Abstract:
Lines 13-17: Is this finding detailed in the conclusions?

Line 23: I think the authors should be careful saying they provided insight into catchment scale connectivity with such a small spatial distribution of wells.

Main text
Should the intro be past tense?

Page 3 line 11: can spatial connectivity be addressed with the study design?

Hypothesis 2: can you add some discussion in your lit review about bedrock controls?

Hypothesis 3: again, catchment scale connectivity just seems like a stretch. I wonder if a surrogate for this is just catchment wetness or storage state.

Figure 1: I know there is a lot going on in this figure already, but some indication of land cover could be good

Figure 2: and all figures: since the bedrock comparison is important for the study, it could be good to indicate that more clearly in these figures, so maybe a different color in the tabs where the site labels are for the two bedrock types, or just labelling them with the name of the bedrock types… just a suggestion.

Page 6 line 3: “were taking” = took

Page 10 line 9: “allow us” = allowed us

Page 12 line 23: “also” could be removed

Page 15 line 5: “values not necessarily” , should this be “values did not necessarily”

Figure 9: Very cool plot.
Figure 9: For this and figures 10 and 11: I wonder if there is a way to indicate in these, whithout them getting too cluttered, how far the wells are from the stream. While they are all considered “near stream”, some are much further away than others, and seeing that information might aid in interpretation

Page 18 line 18: Are the thresholds in water level all at the same stream stage across wells? That could be interesting to discuss.

Page 21 line 9: Can the authors comment on whether or not the particle size distribution in the shallow soils is consistent with the ability to have a thick capillary fringe?

Page 23 line 15: I’m not sure what “infiltration distance” means in this context

Page 23 line 26: “on the one hand” and “on the other” could be removed

Page 24 line 13: “variability of possible flow paths compared to conditions of high groundwater levels” - I’m not sure what this is saying

Page 24 line 18: is = are

Page 25 line 14: I don’t see how, with this experimental setup, conclusions can be made about how many hillslopes are connected or how far that connection extends up slopes

Page 26 lines 6-7: According to what analyses can you say how many events you need to characterize a response?

Page 26 line 12: It’d be good to see an example of this capillary fringe transition… could this just be explained by higher winter-time water levels? It’d be interesting to see just how fast this transition occurs. (I may be misunderstanding this though)

Page 27 line 9-13: I don’t think much can be said about hillslope travel distances/connectivity in the framework described in the Klaus and Jackson paper. Sure, there is a level at which there is a suggestion that a lot of water is moving into the stream, but is that just due to a much higher potential energy gradient? Does that really give any specific indication of from how far up the hillslope that water is coming? It just doesn’t seem that relevant to the rest of the paper.

Conclusions
First paragraph: First, I think a better first paragraph would be a very succinct summary of what the authors did in the study, just as a lead in and transition. Second, as detailed above, I’m not sure the conclusions made quite strongly in this first paragraph are backed up by the study. I’d consider adding some sort of detailed discussion or analysis of how the findings of the study would change with a more limited time series, or just taking this stuff out.

Hypothesis section: I like the very straightforward presentation of how the hypotheses were addressed. However, I think it would be useful to restate each one. Otherwise you have to flip back and forth in the paper to see which is which. Also, I think it would be good to go through these and be sure that there is only discussion of conclusions directly about each hypotheses in each paragraph. Right now H1 has discussion of geology that doesn’t seem related to H1 as stated in the beginning of the paper.

Page 28 Lines 15-19: I had to read this sentence several times to figure it out, I think it should be split into two or more.

Page 28 line 32: as mentioned above, the footslope can be gatekeeper, but there are other explanations for why water levels there might be well synced up with streamflow.

Page 29 line 5: “can be identified a-priori” if this is going to be included in the conclusions, I think there should be discussion of at least a proposed method for how this could be done in the discussion. Were the topographic similarities between the wells that worked well for the analysis?

Page 29 line 7: “novel way of visualizing” Definitely! The figures in this manuscript are super well done.

Hide

ED: Publish subject to minor revisions (review by editor) (07 Aug 2020) by Hjalmar Laudon

AR by Daniel Beiter on behalf of the Authors (02 Sep 2020) Author's response Manuscript

ED: Publish as is (29 Sep 2020) by Hjalmar Laudon

Short summary

We investigated the interactions between streams and their adjacent hillslopes in terms of water flow. It could be revealed that soil structure has a strong influence on how hillslopes connect to the streams, while the groundwater table tells us a lot about when the two connect. This observation could be used to improve models that try to predict whether or not hillslopes are in a state where a rain event will be likely to produce a flood in the stream.