Review of Neilans and Urbano, 2007 from Journal of Hydrology

From GeoMod

Date: 17 Aug 2007 To: "Lensyl Urbano" lurbano@memphis.edu From: "J. Hydrology" hydrol-eo@elsevier.com Subject: HYDROL6090 - Editor decision - revise

Dear Dr. Urbano,

I can now inform you that the reviewers and editor have evaluated the manuscript "Assessing the meter-scale spatial and temporal variability of groundwater discharge using stream-bed thermal mapping" (Dr. Lensyl Urbano). As you will see from the comments below and on http://ees.elsevier.com/hydrol/, moderate revision has been requested.

Please consider the reviews to see if revision would be feasible. To facilitate further review, add line numbers in the text of your manuscript. Should you wish to resubmit you should explain how and where (i.e. by giving line numbers) each point of the reviewers' comments has been incorporated. For this, use submission item "Revision Notes" when uploading your revision. Also, indicate the changes in an annotated version of the revised manuscript (submission item "Revision, changes marked"). Please follow the order "Revision Notes", "Revision, changes marked", "Manuscript", etc.

Should you disagree with any part of the reviews, please explain why.

Please strictly follow the formatting requirements as presented in the Guide for Authors. When submitting your revised manuscript, please ensure that you upload the source files (e.g.Word). Uploading a PDF file at this stage will create delays should your manuscript be finally accepted for publication. If your revised submission does not include the source files, we will contact you to request them.

Given that the requested revisions are moderate the new version is required within three months.

To submit a revision, go to http://ees.elsevier.com/hydrol/ and log in as an Author. You will find your submission record under Submission(s) Needing Revision.

When resubmitting, please present any figures, tables etc. as separate files. See the Artwork Guidelines on the home page right menu for further file naming conventions, referencing and format issues.

I hope that you will find the comments to be of use to you and am looking forward with interest to receiving your revision.

Thank you for submitting your work to this journal.

With kind regards,

Journal Manager Journal of Hydrology

COMMENTS FROM EDITORS AND REVIEWERS

Contents 1 EDITOR'S COMMENTS: 2 Reviewer #1 3 Reviewer #2 4 Reviewer #3

EDITOR'S COMMENTS:

This manuscript is of potential interest to the Journal of Hydrology readership. All reviewers made useful and constructive comments, which the authors need to address in detail in a letter to the Editor and revise their manuscript as appropriate.

• The English language also needs improvement.

In quickly reading through the manuscript, the Editor also has a few additional comments as follows:

• In lines 148-149 please explain how "The temperature probe was insulated between itself and the plate."
• Common experience indicates that sediment temperature at a shallow depth, such as 20 cm, is variable through the seasons. You need to show that in your case the 20-cm temparature is constant over the field measurement season (not just over a day).
  • While we appreciate the importance of this question, our objective is to analyze the changes in groundwater discharge patterns and their interaction with bedform change over the field measurement season. As such, reorganization of shallow gw flow patterns is expected to change the temperature at 20 cm beneath the surface (see Conant, 2004). Indeed, our assumption is that only changes in sub-bed temperatures could result in the temporal changes in streambed temperature that we measure.
  • We have modified the text ??? to indicate this assumption.

Reviewer #1

Review: Assessing the meter-scale spatial and temporal variability of groundwater discharge using streambed thermal mapping. Neilans and Urbano

Submitted to J. Hydrology

This article discusses the use of streambed temperatures to characterize surface-water / ground-water (sw/gw) interactions in a stream within the Mississippi Embayment. Discharge inferred through temperature measurements are compared to discharge through seepage meters. Discharge is found to be highly heterogeneous within the streambed and is characterized by streambed springs. These streams affect the bed morphology in some cases.

Overall I think that the article presents some interesting data and will be of interest to J. Hydrology readers. In my opinion there are some important technical points that should be addressed. In addition, the authors have missed some important previous works that should be reviewed. One paper in particular, by Brewster Conant, has an impact on the novelty of the reported methods.

• Indeed the paper by Conant (2004) was overlooked. As it uses a very similar method to our own the comment on the novelty of the method is well taken. Schmidt et al. (2006) took a similar approach, although they measured temperatures at different sub-streambed depts across 2 cross-sectional transects. Our method however does have an important difference; we show that it is possible to use temperature measurements at the sediment-water interface in high discharge reaches to identify spatial patterns of groundwater discharge, whereas Conant (2004) used measurements at 20cm beneath the interface. Given the importance of Conant (2004) and Schmidt et al. (2006) we have extensively incorporated their results into our introduction and discussion sections.

As detailed below, I recommend that the article undergo a revision and be resubmitted for publication. Previous work needs to be better documented. In light of previous work, I think the introduction and conclusions should be refocused on the measurements and interpretation, rather than the measurement technique which is not novel. The title should be revised appropriately. If these changes are made, the article will be a significant contribution to the literature.

• We greatly appreciate the recommended literature and have extensively incorporated the suggested previous works into the manuscript, and particularly recognize Conant's work, we do believe that our work shows that it is possible to determine the general spatial patterns of groundwater discharge using temperature measurements at the water-sediment interface, which has not been previously demonstrated. This greatly speeds the measurements even when compared to Conant and Schmidt et al., and demonstrates that under high discharge conditions the mixing of stream and gw discharge is inefficient enough to allow for the efficacy of this method. Our work, Conant and Schmidt et al. all show discharge foci with high rates of groundwater discharge; we demonstrate that it at least under certain conditions, it is necessary only to measure temperature at the water-sediment interface in order to ascertain the general spatial pattern of gw dishcarge.

Major Comments:

• 1. I suggest that the authors take a look at a recent review (Kalbus et al., 2006) of measurement of gw/sw measurements to place their article in better context. The literature review needs to be expanded to include the recent flurry of activity in this area.
  • We acknowledge the accuracy of this comment and thank the reviewer for bringing to our attention the references cited at the bottom of this review. This literature has been extensively integrated into the literature review and discussion.
• 2. In particular, Brewster Conant (Conant, 2004) performed high resolution temperature sensing in a stream and found similar behavior to that reported in the submitted article. There were regions of subsurface springs, for example. He compared his temperature measurements to those of mini-piezometers. The authors do not mention this method as an alternative to seepage meters.
  • The Conant (2004) reference is particularly appreciated, as its results comport so well with ours. Our inclusion of this reference in the manuscript is described above. We did not use mini-piezometers as opposed to seepage meters because the latter provide data closest to the water-sediment interface for calibration of the spatial thermal measurements. Given the spatial distribution of temperature (and thus discharge) we could not make the common assumption that all groundwater flow was vertical, which is necessary for the use of point piezometer measurements.
• 3. In light of the history of using streambed temperatures as an indicator of inflow e.g. (Becker et al., 2004; Conant, 2004; Constantz, 1998; Kalbus et al., 2006; Keery et al., 2007), I do not think it is appropriate to focus the paper as streambed temperature as novel technique for measuring sw/gw interactions. The authors should avoid misleading statements in the conclusions such as "Bed-surface temperature can be used to determine relative rate of groundwater discharge seepage in a gaining stream". This has already been established in the literature.
  • We are in slight disagreement with this comment, perhaps because of the differing uses of the term "streambed". We use the term from the perspective of stream hydrology where it refers to the sediment-water interface. The reviewer uses the term to refer to the sediments below the interface, as does the authors in the references cited here. These references, as is typical for the literature on thermal methods, use the assumption of the mathematical models (eg. Keery et al., 2007) that the water-sediment interface can be treated as a specified boundary condition, where the temperature at the interface is equal to the stream temperature. Conant (2004) for example uses measurements from 20cm below the surface. We demonstrate that this is not entirely the case. We detect the thermal signature measured at 20cm at the water-sediment interface. We conclude that the high discharge rates at the discharge foci create a thin boundary layer above the water-sediment interface. This boundary layer small however, so it may not materially affect the specified temperature assumption of the mathematical models.

• 4. To me, it seems the more significant contribution is the dataset and interpretation. Conant's work is the only other study with which I am familiar that maps discharge at this spatial resolution. His stream was in a very different hydrologic environment. It is interesting to see similar behavior in such a different geomorphologic setting, and it calls into question the "textbook" conceptualization of stream recharge as a primarily bank process. This is a significant finding.
• 5. I think the authors have some problems with the resolution of the Type-K thermocouple. These thermocouples have a resolution of 0.1 C, and in most cases the difference between measurements are used so the error is +/- 0.2C. In Figure 3, for example, a regression is made between discharge rate and the temperature difference between river bed and water column. These temperature differences are between 0.6 and 0.2 deg so with error bars of 0.2 deg C is this significant? Also, in the maps shown in Figure 4 contour intervals are 0.2 deg C, which seems inappropriate given the error. Has the precision of the thermocouple been tested in the field? Perhaps that would lend more confidence to the data. The error of the thermocouple measurements need to be expressed as error bars on the plots and the impact of the measurement error discussed in the text.

Minor/Specific comments:

• 74. should be a mention of the use of mini-piezometers as a method of measuring seepage
• 158. Spelling: interference
• 162 Orphan paragraph
• 173 Spelling: potential
• 196 "probe was set to record"…
• 216 The text here does not match the data in the figure. The figure 3 is entirely mislabeled and needs to be reworked.
• 225 "It should be possible to eliminate this effect (different regressions) if the temperature of the bed is measured below the effects of diurnal heating…" Isn't the confidence in the regression highly impacted by the difference in temperature between ground water and surface water temperature, especially in light of the limited resolution of the thermocouple? You should see better signal to noise in the summer where the surface temperature is warmer relative to ground water temperature. Is the surface water temperature affected by precipitation events?
• 256 Since you raise the subject of spatial correlation you should describe the kriging parameters (variogram) used to krige the data. Was the experimental variogram linear? I would think not given the localized influence of the springs. A spherical model, for example, would be more appropriate
• 289 Was the bank discharge ground water or bank storage?
• 296 Why did you suspect that the lineaments in temperature were artifacts and how were you able to reject this possibility?
• 306 It seems highly unlikely that you would have transmissive fractures in a saturated clay.
• 324 One of your primary conclusions in the paper is that bank and middle-stream discharges are independent. The fact that the bank and mid-river discharge areas line us seems to contradict this conclusion. This contradiction needs to be carefully explained.
• Fig2 Symbols should be large, hard to distinguish plots.
• Fig3 Figure 3 is almost entirely mislabeled and does not match text.
• Fig4 Contours seem too fine given error in temp probe. Perhaps 0.4 C would be more appropriate. I think the grayscale levels will not support 0.2 anyway.

References

• Becker, M.W., Georgian, T., Ambrose, H., Sinscalchi, J. and Fredrick, K.C., 2004. Estimating ground-water discharge using stream temperature and velocity. J. Hydrology, 296(1-4): 221-233.
• Conant, B., 2004. Delineating and quantifying ground water discharge zones using streambed temperatures. Ground Water, 42(2): 243-257.
• Constantz, J.E., 1998. Interaction between stream temperature, streamflow, and groundwater exchanges in alpine streams. Water Resources Research, 34(7): 1609-1615.
• Kalbus, E., Reinstorf, F. and Schirmer, M., 2006. Measuring methods for groundwater - surface water interactions: a review. Hydrology and Earth System Sciences, 10(6): 873-887.
• Keery, J., Binley, A., Crook, N. and Smith, J.W.N., 2007. Temporal and spatial variability of groundwater-surface water fluxes: Development and application of an analytical method using temperature time series. Journal of Hydrology, 336(1-2): 1-16.

Reviewer #2

Manuscript Number: HYDROL6090

Title: Assessing the meter-scale spatial and temporal variability of groundwater discharge using stream-bed thermal mapping Authors: D.A. Neilans and L. Urbano Article Type: Research Paper

REVIEWER BLIND COMMENTS TO AUTHOR

ABSTRACT

• 1. The last sentence is unclear and should be rephrased. What is meant by "intermediate scale variability"?

INTRODUCTION

• 2. The first para. of the Intro should set the study in a wider context and identify clearly the key issues. The current text provides an account of a sinlge study but then draws some wider implications. In general, there is over-reliance on Becker et al. 2004 in the Intro.
• 3. The Intro misses some key papers and, thus, the study is not set in an appropriate context. There are several fairly recent journal papers (in Water Resources Research, Journal of Hydrology, Hydrological Processes etc.) on hyporheic thermal dynamics plus linking seepage rates and riverbed temperature that are not cited. There is also some very nice, recent work on the effects of bedforms on hyporheic flow and thermal patterns in Water Resources Research and Journal of Hydrology. Reference to basic texts (such as Knighton 2002) could be avoided by reference to such journal articles.
• 4. Some text on the controls on/ processes determining riverbed temperature would provide useful background.
• 5. The aim of the paper could be me more clearly specified at the in the last para. of the Intro, with subsidiary objectives also stated.

STUDY AREA

• 6. A map of the study area would be a useful addition, with a detailed inset of the study reach and its sampling framework.
• 7. Some of the study area description detail seems tangential to the current study (or its importance is not highlighted to the reader?).
• 8. It needs to be explained why a gaining section was needed (line 124-). Who identified the reach as gaining (and how)?
• 9. Data are required to support the assertions about sediment (line 125-)
• 10. The year of data collection is not given (line 140-)
• 11. How representative a low flow conditions of system dynamics (line 104-)? How many sampling dates? Perhaps this information should be included in the Methods.

METHODS

• 12. What are type-K thermocouples (line 143-)? How was the sensor insulated from the plate (line 149-)? The 2 sec response time seems rather short for a typical thermocouple (cf. fine-wire thermistor). Further details are required about the sensor.
• 13. What brand and model of datalogger was employed (line148-)?
• 14. Details of the seepage meter array are required (line 162-).
• 15. "temperatures at greater depth" (line 162-): this text is too unspecific.
• 16. The rationale for section 3.1 needs to be provided at the outset of the methods and in the wider context of the study objectives.
• 17. Line 173-178: this text is very difficult to follow both in terms of scientific rationale (i.e. potentially flawed argument and lack of detail) and writing.
• 18. Line 179-186: this text is too vague and again very difficult to follow. Other studies on riverbed temperature fluctuations at depth should be cited.
• 19. Line 187-189: this text is far, far too vague and raises serious questions about the robustness of the data set.
• 20. Section 3.2 seems to report results rather than methods. It seems odd there is no thermal lag with depth into the riverbed.
• 21. There are 2x section 3.2.
• 22. Line 208: why were seepage meters installed at random locations. Line 209-210: why were 3 and 6 sites used on different dates? The sampling framework does not seem very systematic.
• 23. Figure 3 and lines 212 and 222-: it is inappropriate to draw a regression line through 3 points (09/06/06). By omitting the circled point (see annotated manuscript), all data seem to fall along a single regression line.
• 24. Section 3.3 (renumbered) seems to report results rather than methods.
• 25. Line 227-228 does not make sense.
• 26. How was bed topography derived?

RESULTS AND DISCUSSION

• 27. It would be preferable to have separate results and discussion sections to differentiate facts from inference.
• 28. Line 232-: This statement is only partially true as the method may work in temperate region winter when the water column is cooler than warmer groundwater.
• 29. The first two para. of the Results report on methods; hence, the paper is poorly organised.
• 30. Line 252: How were weekly measurements derived? The methods suggest spot sampling. Are these one-off measurements (one per week) or weekly averages based on what resolution of original observations?
• 31. What other factors (than seepage) may have influenced the water column-bed thermal gradient (channel hydraulics, riparian shading etc.)? Have these influences been factored out?
• 32. The interpretations on lines 271-278 seem highly speculative in the absence of independent data.
• 33. What are the likely processes causing the banks and 'springs' to be "separate discharge regimes" (line 293)? Another mid-chhannel discharge regime is later introduced (line 310).
• 34. Line 295-297 need further explanation.
• 35. Could thermal differences be due to hyporheic recirculation of channel water and/or preferential locations for upwelling driven by pressure differentials over the bed topography?
• 36. Line 316-: a text citation is needed.
• 37. Data to link to precipitation events (line 321 and 328) are not presented; hence, this inference is very speculative.

CONCLUSION

• 38. Line 348: there is a potential chicken-and-egg problems re: low points and springs.
• 39. Delete last line.

FIGURES

• 40. I could not find any figure captions. This makes interpretation of Figures 4 and 5 particularly difficult.
• 41. Figure 1 is not cited in the text.
• 42. Figure 2: line styles cannot be differentiated (and high-res TIF image file is corrupted). The sample site numbers are not explained.

REFERENCES

• 43. Cushman et al. 1964 and Hosman and Weiss 1991 are missing from the reference list
• 44. The reference list is not formatted in JoH house-style. It is inconsistent in format and should not be bulleted.

WRITING

• 45. The paper is rather poorly expressed and organised, so it reads like an early draft rather than final manuscript for journal submission. Suggestions for rephrasing of unclear sentences are made (underlined) on the annotated manuscript. Numerous typos are also noted on the manuscript.
• 46. Please use SI (Système International d'Unités) units consistently throughout the paper.

Reviewer #3

HYDROL6090

Assessing the meter-scale spatial and temporal variability of groundwater discharge using streambed thermal mapping By Neilans and Urbano

General comments: In this paper, the authors address the issue of assessing the variability in groundwater discharge by using streambed thermal mapping. The paper describes the application of a very interesting method ( thermal mapping) which improves our understanding about spatial variability in groundwater discharge. In addition, the study identifies stable discharge foci that persist over long time periods.

Obviously, the authors put a lot of work and effort into the experimental work and also in writing this manuscript which I definitely would like to acknowledge here. My main point of concern is that the paper suffers from that a wider context missing. It is difficult to put this study into any context as information about scales (temporally and spatially), climatic conditions at the study basin etc. are not given. The method section is very detailed and well written, however, the introduction and discussions (and cited literature) needs in my point of view large expansion (i) with regards literature which was published in recent years and (ii) with regards to studies which deal with the actual implications of groundwater recharge / hyporheic processes / groundwater / surface water interactions (e.g. ecological implications). A number of groups published very interesting results with regards to this recently, none of them is cited here.

The paper for the most part is clearly written - although the result section contains large parts which would be method section - and is acceptable for publication in JoH after moderate revision. I do have, however, specific comments which I outline below for the authors to consider as part of improving this piece of work prior to publication.

Specific comments: Introduction:

P. 2. Within the introduction (in particular at its start) I find a short section missing which would explain why it is actually important know about seepage and seepage rates and why we still need to increase our process understanding with regards to this. Please expand on this and what some of the possible implications about groundwater recharge.

P4. Please define clear objectives of this paper.

Study area: p.5 Please give details about scales (spatial and temporal) to put this study site into context. E.g. could you provide a map or similar about scale of investigated river / river reach?

P5. What are general climatic conditions at this site? Which years were investigated? sWhat is annual precipitation (for which periods?), annual discharge etc? Is there any background information available about groundwater contribution in this catchment / site?

P5, L. 119; Urbano et al 2005? Or change in reference list

P5, L. 126: scale information?

P 6, L. 140, which year? What were climatic conditions during this year?

Methods: P. 6 L143: which provider? P. 9, L. 206 change "if the" to "if they"

Results and discussion: p. 10: The section 4.1. is in my point of view a section for the methods (no results)

p.10, L. 232: "based on the correlation" - there is no real correlation. How is the described uncertainty incorporated and considered?

p. 10, L. 234: "well during warmer months" which are?

p. 11, L.250-255: this para should be part of the method section.

p. 11, L. 253: why kriging and not other interpolation methods?

Conclusion: The conclusion is not well structured and doesn't read smoothly. Again, please mention briefly why knowledge about groundwater recharge is actually important and what are the implications (e.g. the whole m/s doesn't relates to studies which have shown ecological implications of groundwater recharge etc…)

References: Expand on literature about importance of understanding hyporheic exchange flow, gw/surface water interactions, gw recharge etc. Malcolm et al, Gooseff et al…

Figure captions are missing, hence not able to comment on their clarity.