the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Dec 2023
Conceptualising surface water-groundwater exchange in braided river systems
Abstract. Braided rivers can provide substantial recharge to regional aquifers, with flow exchange between surface water and groundwater occurring at a range of spatial and temporal scales. However, the difficulty of measuring and modelling these complex and dynamic river systems has hampered process understanding and the upscaling necessary to quantify these fluxes. This is due to an incomplete understanding of the hydrogeological structures which control river-groundwater exchange. In this paper, we present a new conceptualisation of subsurface processes in braided rivers based on observations of the main losing reaches of three braided rivers in New Zealand.
The conceptual model is based on a range of data including: lidar, bathymetry, coring, particle size distribution, groundwater, temperature monitoring, radon-222, electrical resistivity tomography, and fibre optic cables. The combined results indicate that sediments within the recently active river braidplain are distinctive, with sediments that are poorly consolidated and better sorted compared to adjacent deposits from the historical braidplain, which become successively consolidated and intermixed with flood silt deposits due to overbank flow. A distinct sedimentary unconformity, combined with the presence of geomorphologically distinct lateral boundaries, suggests that a “braidplain aquifer” forms within the active river braidplain through the process of sediment mobilisation during flood events.
This braidplain aquifer concept introduces a shallow storage reservoir to the river system, which is distinct from the regional aquifer system, and mediates the exchange of flow between individual river channels and the regional aquifer. The implication of the new concept is that surface water-groundwater exchange occurs at two spatial scales. The first is hyporheic and parafluvial exchange between the river and braidplain aquifer. The second is exchange between the braidplain aquifer and regional aquifer system. Exchange at both scales is influenced by the state of hydraulic connection between the respective water bodies. This conceptualisation acknowledges braided rivers as whole “river systems”, consisting of channels, and gravel aquifer.
This work has important implications for understanding how changes in river management (e.g., surface water extraction, bank modification and gravel extraction) and morphology may impact groundwater recharge, and potentially river flow, temperature attenuation, and ecological resilience during dry conditions.
- Preprint
(2647 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-2767', Anonymous Referee #1, 16 Feb 2024
Braided rivers can provide substantial recharge to regional aquifers, with flow exchange between surface water and groundwater occurring at a range of spatial and temporal scales. However, the difficulty of measuring and modelling these complex and dynamic river systems has hampered process understanding and the upscaling necessary to quantify these fluxes. This is due to an incomplete understanding of the hydrogeological structures which control river-groundwater exchange. This paper presents a new conceptualization of subsurface processes in braided rivers based on observations of the main losing reaches of three braided rivers in New Zealand. It is useful to understand the braided river systems.
Major issues:
- Figure 9 in page 22, please explain that why the “glacial outwash gravels” are unsaturated? And how about the braidplain aquifer? Saturated or unsaturated? Can you compares the ERT results with your borehole surveys in Fig. 4.
- In Fig. 10 at page 23, can you explain that what is the “DTS” and “A-DTS”?
Minor issues:
- Line 17:“including:” should be changed to “including”.
- Line 26:“hyporheic” should be changed to “the hyporheic”.
- Line 27:“exchange” should be changed to “the exchange”.
- Line 28:“Exchange” should be changed to “The exchange”.
- Line 30:“aquifer” should be changed to “aquifers”.
- Line 55:“scale” should be changed to “scales”.
- Line 95:“net” should be changed to “a net”.
- Line 100:“reach” should be changed to “reaches”.
- Line 147:“recognises” should be changed to “recognizes”.
- Line 190:“result” should be changed to “results”.
- Line 218:“enables” should be changed to “enable”.
- Line 282:“are” should be changed to “were”.
- Line 291:“underling” should be changed to “underlying”.
- Line 301:“on” should be changed to “in”.
- Line 304:“insufficient” should be changed to “an insufficient”.
- Line 331:“on Fig. 3” should be changed to “in Fig. 3”.
- Line 351:“on Fig. 4” should be changed to “in Fig. 4”.
- Line 351:“on Fig. 10” should be changed to “in Fig. 10”.
- Line 415:“are approximately” should be changed to “is approximately”.
- Line 416:“condition” should be changed to “conditions”.
- Line 441:“notable” should be changed to “a notable”.
- Line 446:“in the near surface” should be changed to “at the near surface”.
- Line 474:“in the 1960’s” should be changed to “in the 1960s”.
- Line 501:“on Fig. 1” should be changed to “in Fig. 1”.
- Line 515:“losing water” should be changed to “lose water”.
- Line 517:“therefore interpretation” should be changed to “ therefore an interpretation”.
- Line 550:“hyporheic exchange” should be changed to “the hyporheic exchange”.
Citation: https://doi.org/10.5194/egusphere-2023-2767-RC1 -
AC2: 'Reply on RC1', Scott Wilson, 04 Apr 2024
Dear Reviewer
Thank you for your review of our article. We understand that we should improve some aspects of this article and are willing to address your comments in a revised version. Please find below answers to your questions and explanations on how we would address your comments. Our responses in blue refer to lines in the original document, and your comments are in black.
Scott Wilson (on behalf of the co-authors)
-
RC2: 'Comment on egusphere-2023-2767', Anonymous Referee #2, 13 Mar 2024
In this study, the authors present a revised conceptualisation of the sedimentary structure of braided river-aquifer systems as it is relevant for the quantification, upscaling and modelling of surface water-groundwater exchange fluxes on larger (regional to catchment) scales. The revised conceptualization builds on a range of different types of measurements, namely lidar, bathymetry, coring, particle size distribution, groundwater levels, water temperatures, radon-222 concentrations, electrical resistivity tomography and fiber optic distributed temperature sensing. The study is based on the analysis of three losing braided river reaches in New Zealand.
The authors demonstrate that the still active braidplain sections of a braided river system contain (or form) better sorted, largely unconsolidated sediments that can create a more conductive aquifer structure (which they call "braidplain aquifer"), while the formerly active, now inactive river sections develop more consolidated and clogged riverbeds or sediment layers due to the repeated deposition of fines and silty material during overbank flows. These "regional aquifers" are located underneath or adjacent to the braidplain aquifer. While these are essentially not new concepts, and riverbeds as separate hydrogeological units have long been considered are as the critical hydrogeological unit controlling exchanges in river-aquifer systems, the confirmation and extension of this concept to braided-river systems and a "braidplain aquifer" concept is still very valuable. What thus can be considered new here is the definition and characterization of an intermediate storage reservoir/aquifer layer which essentially extends the concept of riverbeds that act as modulating layers for the exchange between a (braided) river and an underlying aquifer to a system of larger vertical and horizontal extent and more spatial complexity. I believe that this concept is very helpful to better characterize, quantify and simulate the interactions between streams and aquifers in braided river systems, as it is, in principle, better suited to also address the highly preferential flow structures which one finds in such systems. These flow structures become slightly less important or more focused in deeper, older layers (buried paleochannels), but in the younger, shallow layers which lie close to the very active surface, the interactions and flow paths can be very complex and critically control exchanges not only of water but also of heat and mass.
Overall, the manuscript was nice to read. It is very clearly written and very well structured. I also really appreciated the clear definitions of the main terms in the introduction. It made the reading and understanding of the rest of the manuscript straightforward. Content wise, the updated conceptualization of a braided river-aquifer system can be adapted to many riverine environments and has a wide scientific relevance. I find that the following sentence of the discussion section nicely summarizes the applicability of the concept: "By identifying the base and margins of the BPA, and the process which forms it (reworking of bed material), the vertical and lateral extents to which hyporheic and parafluvial exchange occur can be identified by a change in sediment characteristics." In essence, the study shows that already relatively few key measurements are enough to identify the extent of the main layers as presented in their revised conceptualization, and that this knowledge can then be used to estimate water fluxes over much larger areas in these otherwise very complex braided river systems.
However, I also have some major points which I find have to be addressed first before the manuscript can be accepted. These as well as specific or minor points are listed in detail below. A major conceptual shortcoming is the fact that this intermediate aquifer layer between the river and the aquifer, i.e., what the authors call braidplain aquifer, is normally highly complex and can form highly conductive subsurface channels for preferential flow that, in case they are at some point covered by new sediments and not reworked anymore, can lead to buried paleochannels. The extended riverbed or "braidplain aquifer" concept therefore is primarily useful for simplified analyses on a very large scale, i.e., regional to catchment scale. On the reach or local scale, which was studied here, heterogeneity in the braidplain aquifer becomes extremely important for river-aquifer exchange fluxes, radon/residence time distributions, and flow paths in general. I think it is great to have a simplified conceptualization for regional scale assessments as it is presented in this generally excellent manuscript, but I miss an honest discussion of where this concept is NOT suited. Heterogeneity in these sediments is mentioned in the introduction, but the discussion section is void of any critical reflection of the new concept. I would highly appreciate some honest discussion about where this concept is usefull, and where it is too simplistic, as this will be very important information for practical implementations of the concept. I also must point out that the radon measurements and analyses are not sufficiently well described or their uncertainty properly addressed. The residence times derived from the radon measurements are fantasies, as the employed method doesn't allow measurements precise enough to derive residence times beyond the 3-half-lives mark of Rn-222. But as the uncertainties of the measurements are not presented, the reader is not able to see this easily.
Given these shortcomings in an otherwise very nice manuscript, I recommend minor revisions before accepting the manuscript for final publication.
Major comments
Radon: Unfortunately, the entire radon aspect needs a thorough revision. The method is not sufficiently well described to allow an assessment of the validity/quality/uncertainty of the data (e.g., which water amount was analysed per sample? How long were samples stored before being measured? How long were the samples measured for,i.e., how many counting cycles were employed?). Moreover, the (instrument reported) measurement uncertainties are not provided, nor is any assessment of the uncertainty of the radon measurements made. The residence times that are calculated based on the measured radon activities are fantasies, as the precision of the employed method is not good enough (it is impossible to use the Rad H2O method and get radon concentration measurements at a precision that would allow a residence time detection beyond around the 3-half-lives mark of radon (ie. about 12 days), and this without even considering the fact that most likely the samples were not measured directly but first stored for a day or two, leading to a further loss in precision). Last but not least, the secular equilibrium for one of the two measured sites is in my opinion not chosen appropriately, as the regional aquifer shows much more elevated background activity concentrations. I don't see why the braidplain aquifer should be used to define the background values instead of the regional aquifer, unless the sediments are very very different, which is very unlikely. Even though this aspect of the study needs a major overhaul, looking at the data tells me that the main conceptual conclusions from the data will most likely remain valid.
Lidar-Bathymetry data: Although mentioned as the first two data types in the abstract, this data was not presented in the paper. It is only used to discuss the concept and derive conclusions. Since this data is important for the assessment of braided river-aquifer systems, and critical for the derivation of the updated conceptualization of braided river-aquifer systems in this manuscript, it must be provided somehow.
Specific comments:
L 103-106: a more recent modelling approach for braided river-aquifer systems with their buried paleochannels that should be referred to here is Schilling, O. S., Partington, D. J., Doherty, J., Kipfer, R., Hunkeler, D., & Brunner, P. (2022). Buried paleo-channel detection with a groundwater model, tracer-based observations, and spatially varying, preferred anisotropy pilot point calibration. Geophys. Res. Lett., 49(14), e2022GL098944. https://doi.org/10.1029/2022GL098944
L 107-109: Indeed, changing bed morphologies create problems for the representation of braided river systems (and this problem isn't even restricted to braided river systems but occurs also in less complex river corridors). One way to deal with this problem when one simulates SW-GW interactions in braided river systems is using data assimilation, where whenever new information on the bathymetry of the river system becomes available, that information gets incorporated into the forward model. The concept was demonstrated by Tang, Q., Schilling, O. S., Kurtz, W., Brunner, P., Vereecken, H., & Hendricks Franssen, H.-J. (2018). Simulating flood induced riverbed transience using unmanned aerial vehicles, physically-based hydrological modelling and the ensemble Kalman filter. Water Resour. Res. https://doi.org/10.1029/2018WR023067. Although not necessarily feasible or required on the regional scale, this way forward should nevertheless be mentioned here. Another approach that can be useful relies not on assimilation of riverbed bathymetry information but on a moving pilot points approach, which can be used to "follow" changing structures simply via inversion of hydraulic- and tracer-based data: Khambhammettu, P., Renard, P., & Doherty, J. (2020). The traveling pilot point method. A novel approach to parameterize the inverse problem for categorical fields. Adv. Water Resour., 138, 103556. https://doi.org/10.1016/j.advwatres.2020.103556. Also this study should be mentioned here.
l192-211: It is worth mentioning here that this sedimentological and SW-GW interaction concept of braided, alluvial gravel rivers was already presented a rather long time ago, and moreover gave the different types of gravel-sand sediments a full nomenclature. The key reference to mention here is: Huggenberger, P., Hoehn, E., Beschta, R., & Woessner, W. (1998). Abiotic aspects of channels and floodplains in riparian ecology. Freshwater Biol., 40, 407-425. https://doi.org/10.1046/j.1365-2427.1998.00371.x
l226: don't superscript the "s" in the unit
l228-334: Unfortunately, the findings that the electromagnetic methods do not help in identifying the structure of braided river sediments is not surprising and only confirms what could be observed previously. I disagree however that the magnitude of the resistivity is the main cause for these methods to not work. The main reason is rather that the resistivity of the sediments are too similar, as they are in fact all more or less the same type of gravel and sand, simply differing in whether they are more washed or aligned in different ways. In order to be able to differentiate these types of sediment structures one has to resort to ground penetrating radar. The effectiveness of the method has, for example, been demonstrated in the publication I mentioned above: Huggenberger, P., Hoehn, E., Beschta, R., & Woessner, W. (1998). Abiotic aspects of channels and floodplains in riparian ecology. Freshwater Biol., 40, 407-425. https://doi.org/10.1046/j.1365-2427.1998.00371.x
or in more detail in the following publication: Huggenberger, P., Meier, E., & Pugin, A. (1994). Ground-probing radar as a tool for heterogeneity estimation in gravel deposits: advances in data processing and facies analysis. J. Appl. Geophys., 31(1-4), 171-184. https://doi.org/10.1016/0926-9851(94)90056-6
I would prefer it if the author's revise the statement on why EM methods do/did not work in their case and highlight somewhere that ground penetrating radar could be a way forward in this respect.l 429, and following: unclear what 'BPA source' is supposed to be. Moreover, the following sentence is repeated twice.
l435-437: Looking at table 3, the regional aquifer shows much larger Rn-222 activities than the BPA. Why do the authors believe that it is justified to use the largest measured BPA Rn-222 activity concentrations as the secular equilibrium values? It makes much more sense to me to use the regional aquifer values, assuming that the regional aquifer also consists of sediments from the Waikirikiri river, albeit sediments that were deposited a longer time ago. Using the regional aquifer values as secular equilibrium indicators would result in much lower residence times for the BPA water in Waikirikiri. I believe that the RT of 15.5 days (essentially the upper detection limit of the RAD H2O based Rn-222 method) is too old and in reality would lie closer to the values observed for the Wairau AQ.
l437-438: What is the basis for the assumption that the secular EQ for the Wairau AQ is 4800 Bq/m3 ? For the Waikirikiri site it is explained how the secular EQ was derived, but for the Wairau aquifer it is not. From table 3 I understand that the 4800 Bq/m3 was the highest value measured in the regional aquifer, which makes sense to me. But please explain.
table 3: What are the measurement uncertainties of the Radon measurements? The RAD H2O method is notoriously uncertain if one follows the standard protocol of the manufacturer. Which steps were taken to reduce the measurement uncertainty of the standard measurement protocol? If the default procedure to measure Rad7 in the 250ml Rad H2O bottles was used, then it is highly likely that the instrument reported measurement uncertainties were an order of magnitude larger than the measured values shown for the river water samples. In other words: In order for the reader to be able to assess the quality and validity of the presented radon-222 data with the specific method that the authors used, the authors must provide information on the measurement protocol that was employed and report the instrument-reported measurement uncertainties. This is especially important as the uncertainties in the activities have a huge effect on the uncertainty of the residence time estimates.
table 3: estimating a residence time of 25 days is not possible with the Rn-222 method, unless maybe one has an extremely precise radon-222 measurement system, for example a very well tuned liquid scintillation counter. With the Rad7 instrument, and particularly with the Rad H2O grab sampling technique, uncertainties are much too large to estimate Rn-222 based residence times beyond the 3 half-lives mark, i.e. beyond about 11.5 days.... The fact that the authors provide such values without a proper discussion of the employed method or the uncertainties and detection limits again tells me that a thorough re-thinking and revision of the radon-222 methodology is necessary.
L471: "The hydrogeologic structure..."
L535-540: Lidar and bathymetric data were not presented and therefore can't be used here to discuss aspects/the functioning of the system... or the other way around: if such data are used to derive insights/conceptualisations of braided river-aquifer systems functioning, the respective data must also be presented and discussed properly.
Citation: https://doi.org/10.5194/egusphere-2023-2767-RC2 -
AC1: 'Reply on RC2', Scott Wilson, 04 Apr 2024
Publisher’s note: this comment is a copy of AC3 and its content was therefore removed.
Citation: https://doi.org/10.5194/egusphere-2023-2767-AC1 -
AC3: 'Reply on RC2', Scott Wilson, 05 Apr 2024
Dear Reviewer
Thank you for your comprehensive and constructive review of our article. We understand that we should improve some aspects of this article and are willing to address your comments in a revised version. Please find below answers to your questions and explanations on how we would address your comments. Our responses in blue refer to lines in the original document, and your comments are in black.
In response to your request for an honest discussion of where this concept is NOT suited, we refer you to lines 558-573 of the discussion, where we have suggested settings where the concept may not apply. However, we acknowledge that some additional discussion around heterogeneity is required here and will suggest an improvement to the discussion at the end of this document.
Scott Wilson (on behalf of the co-authors)
-
AC1: 'Reply on RC2', Scott Wilson, 04 Apr 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 360 | 114 | 28 | 502 | 19 | 17 |
- HTML: 360
- PDF: 114
- XML: 28
- Total: 502
- BibTeX: 19
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1