Quantifying thermal refugia connectivity by combining  temperature modeling, distributed temperature  sensing, and thermal infrared imaging

Dzara, Jessica R.; Neilson, Bethany T.; Null, Sarah E.

doi:https://doi.org/10.5194/hess-23-2965-2019

Articles | Volume 23, issue 7

https://doi.org/10.5194/hess-23-2965-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-23-2965-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 23, issue 7

Research article

|

12 Jul 2019

Research article |

| 12 Jul 2019

Quantifying thermal refugia connectivity by combining temperature modeling, distributed temperature sensing, and thermal infrared imaging

Jessica R. Dzara, Bethany T. Neilson, and Sarah E. Null

Download

Final revised paper (published on 12 Jul 2019)
Supplement to the final revised paper
Preprint (discussion started on 07 Sep 2018)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Quantifying Small-scale Temperature Variability comments', Anonymous Referee #1, 17 Oct 2018
- AC1: 'Response to RC1', Sarah Null, 07 Jan 2019
RC2: 'major revisions', Martijn Westhoff, 17 Oct 2018
- AC1: 'Response to RC1', Sarah Null, 07 Jan 2019
RC3: 'Review of HESS-2018-441', Anonymous Referee #3, 22 Oct 2018
- AC2: 'Response to RC', Sarah Null, 07 Jan 2019
RC4: 'Review of "Quantifying Small-scale Temperature Variability using Distributed Temperature Sensing and Thermal Infrared Imaging to Inform River Restoration"', Anonymous Referee #4, 02 Nov 2018
- AC2: 'Response to RC', Sarah Null, 07 Jan 2019

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (08 Jan 2019) by Sally Thompson

AR by Anna Mirena Feist-Polner on behalf of the Authors (20 Feb 2019) Author's response

ED: Publish subject to revisions (further review by editor and referees) (01 Mar 2019) by Sally Thompson

ED: Referee Nomination & Report Request started (03 Apr 2019) by Sally Thompson

RR by Martijn Westhoff (02 May 2019)

Suggestions for revision or reasons for rejection

The authors have improved their manuscript according to all reviews, although I sometimes still have difficulties to distinguish the different statistical measures. Besides that, a couple of new issues have been raised as well, while every now and then I would like to see some more explanation about what exactly is done.

The main ‘new’ issue comes from the newly mentioned literature listed in the introduction on P2, L4-7. Although I am not familiar with this literature, it is stated here that already a lot is known about how, where and when refugia are needed. This means that this data could be applied to the results presented in this manuscript, but this is unfortunately not done. Instead, the authors state that (p16,L10-13) “Future research is needed to validate temperature ranges by river feature at the watershed-scale, evaluate how fish use thermal refugia, and improve understanding of the resiliency of thermal refugia with anticipated climate change (Fullerton et al. 2018; Frechette et al. 2018; Ficklin et al. 2018; Stevens and DuPont 2011; McCullough et al. 2009).”

While the second part of this statement has apparently being done in the cited literature, the first part (“to validate temperature ranges by river feature at the watershed-scale”) is done in this research. By connecting the two, you may get very valuable information about which restoration efforts are required to maintain a save passage for LCT. And such a quantitative analysis is also required to back-up all the suggested efforts listed just below this statement (P16, L14-24).

A second, slightly minor issue is that the author state that it is not possible to come up with maximum temperatures for the 50 and 300 m reaches of the TIR data due to the fact that part of the TIR data resembles the riparian zone. However, they also have the TIR summary points, which do report a maximum value for the 300 m reaches. So how are these maximum values obtained? At the same time I also wonder what causes the very small range in the TIR data compared to the DTS data (which is clearly visible in Fig. 7).

Line by line comments:

P1, L18: The abbreviation of DTS has not been defined yet

P2, L22: rephrase: you cannot have fine spatial scales at point locations

P6, L6: How much time did it take to measure the whole stream? And how much would the temperature change over such a time period (you may get such an estimate from the temperature model).

P6, L18: Was the average flow, the average over the TIR collection time?

P6 L32: Define which boundary conditions are needed

P7, L19: Make clear that r refer to a 300m model reach and not to one of the two locations where DTS has been employed.

P8, L1: “One m extends…” ???

P8, L12: I guess you mean the 'mean' instead of 'median'?

P8, L22-23: Do you mean outside the measured temperature range?

P10,L3: Explain what you mean with ‘consistent temperatures’

P10,L5-6: This is Ti,r, isn't it? I suggest mentioning these parameters every time you report them, so the reader can easily go back to the methods to see which formula is used. Please do this throughout the manuscript. This will also help to see if all statistics parameters mentioned in section 3 are indeed used. Although I did not double check it, I don’t recall to have seen values of Td,r.

P10,L25: cooling effect on what? It is indeed cooler in the drain than outside, but due to the limited length of observations downstream of the drain, it is hard to see any cooling effect here.

P11,L9: Do you mean that the temporal (e.g. daily) range of these features were large, or that they are locations with a distinctive lower/higher temperature than the mean spatial temperature of that specific range?

P11,L13: “for one hour”: I guess you mean "for a single point in time"?

P11,L24-25: Such a firm statement requires some proof, which is missing here. A few lines before it was stated that it MAY be due to such shallow groundwater contributions.
In fact, don't these shallow groundwater contributions, which are caused by irrigation, consist of the same water as the return flows (and thus with a similar temperature)?

P12,L10: Maybe I misunderstood what has been compared here, but this statement implies that the minimum temperatures for all six 50 m reaches within a 300m reach should be the same. When looking at Fig. 5, this seems not to be the case with differences in minimum temperatures between the six 50m reaches of 1 or maybe 2 degrees C

P13,L24-26: Please quantify this effect! In other words: what is the accuracy of this method?

P13,L31-32: I understand that this is outside the scope of this paper, but with some simple back-of-the-envelope calculations (e.g. a simple diffusion equation) it is possible to give an estimate or an upper limit of this stratification. This may also help to get an idea about the accuracy of the TIR data.

P14,L7: “Future studies could collect data specifically to overlap in time and space”: Please make clear what the gain is of doing so!

P14,L16-17: “DTS could measure temperatures that varied spatially over short distances”. Although I understand that the authors only have a couple of ‘summary points’, but the raw TIR data also has this high resolution data, doesn’t it?

P14,L18-19: “indicating that these methods complement each other”: But it could also be that different periods result in different temperature distributions along the complete stream...

P14,L32: “has poor aquatic habitat as a function of streamflow and stream temperature”: What do you mean with this statement?

P15,L1-2: I am not familiar with those studies, but does this conclusion arises from results presented in this manuscript?
Or stated differently: Your results show that although the modelled stream water temperature may be too high, there are still places within each model reach that are colder (or cold enough). Can you subsequently use the findings of the studies listed here or in Line 4-7 of the introduction to indicate if these location for refugia are sufficient for LCT to survive?

P15,L11-12: Also here: Is it possible to connect your quantitative results with the studies described in L4-7 of the introduction. The same for L23-24 of this page

P15,L19-20: I still don't understand what you mean: Is it a spatial temperature range that covers a 300 m modelling grid cell or is it a temporal range comparing day and night temperatures of the specific beaver dam?

P15,L30-32: Are these values compared to the mean temperature of the 300m reach, or do they reflect the maximum range? In case of the latter you cannot simply say that the coldest temperature within a model reach is this much colder, while in case of the former you have to make explicit that in Fig. 9 you assume that the modelled temperature is the 'correct' average of the whole stream segment.

P16,L10-13: I don’t understand why future research is needed for this: In the introduction you stated that this literature studied this effect. So why can you not use their results to say something about the survival changes of LCT for the Walker stream. Eventually you may come up with advice on where extra refugia are needed.
And to be more strict: such a quantitative analysis should be done first before you can suggest the list of restoration efforts listed in the next paragraph (P16,L14-24)

Fig. 2: In section 3.1, it is stated that ~400 m of cable is situated on either side of the river. So that means that the upper half of the plot should be more or less a mirror image of the lower half. So I think it is helpful if the flow direction is indicated in the graph, where the water is flowing to (or from) ~550m.

Fig. 3: The purple dots indicate the borders of the 300m model reaches. However, the reach covered by the DTS cable is 400 (or 450). I understand there can be some kind of sinuosity in the cable, but a difference of 100 or 150 m seems rather large to me. To me it seems that the modelled stream reaches are too short and I am wondering which effect this has on the simulated stream water temperature.

Fig. 6: I don’t understand the phrase “with the upstream-most river km on the left side of the x-axis”. The same phrase is present in the caption of Fig. 7, and there I have the feeling that the authors mean that in the graphs the water is flowing from left to right.

Hide

ED: Reconsider after major revisions (further review by editor and referees) (08 May 2019) by Sally Thompson

AR by Anna Wenzel on behalf of the Authors (24 Jun 2019) Author's response

ED: Publish as is (27 Jun 2019) by Sally Thompson

Short summary

In Nevada's Walker River, stream temperatures nearly always exceed optimal temperature thresholds for adult trout. We used high-resolution measured data to verify simulated stream temperatures and estimate the spatial distribution of cold-water pockets for fish. Irrigation return canals, beaver dams, and groundwater seeps were river features with cold-water, and the average distance between pockets of cold-water in this river was 2.8 km.