Eye in the Sky: Using UAV Imagery of Seasonal Riverine Canopy Growth to Model Water Temperature

Round  1

Reviewer 1 Report

This is a sound piece of scientific work. The originality is somewhat limited, although there are fewer papers on riverine canopy than on streambank shading. There are a few points that could be addressed:

What type of smoothing is applied to shade data between time steps?

What are the implications of increased riverine canopy on salmonid habitat, other than temperature regulation?

I have found a few minor spelling errors and made minor suggestions in the attached file.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer 1,

Thank you for your thoughtful comments and revisions on our submission. We have addressed both the overall comments that you provided, as well as the in-line revisions, and included our point-by-point response below. We have tracked our changes in the manuscript to clearly identify our in-text revisions. We appreciate you time and attention to detail, and feel your comments have greatly improved our submission.

Point 1: What type of smoothing is applied to shade data between time steps?

Shade and roughness values were updated to new values as a discontinuous step function; no smoothing was applied. Language was added to the manuscript to clarify this part of the modeling method.

Point 2: What are the implications of increased riverine canopy on salmonid habitat, other than temperature regulation?

Implications of increased riverine canopy on salmonid habitat may extend to habitat usage influenced by changes to cover and hydraulic properties in streams. While an in-depth review of those implications are beyond the scope of this study, language was added to the discussion to note the potential implications and contrasts to existing work.

Point 3: Line 30: Remove “will”

Revision accepted; change made in text.

Point 4:  Line 30: Add an “s” to “affect”

Revision declined. “Water temperature changes” is a plural subject, making “affect” the correct verb conjugate.

Point 5: Line 117: Change “was” to “were”

Revision accepted; change made in text.

Point 6: Line 118: How so?

This statement refers to a general observation about selection of model performance metrics in Dugdale et al. (2017), a review of process-based approaches for river temperature modeling: “Because of the strong seasonal component present in river temperature series, the use of the model's root mean-squared error (RMSE) is generally preferred to the Nash-Sutcliffe model efficiency coefficient (NSE; see Janssen and Heuberger, 1995) or other similar measures due to the fact that RMSE remains unbiased by seasonal cyclicity.” (p.109)

Point 7: Line 195: Any requirement for ground truthing?

Because of our limited access to the field site, lack of funding for additional surveying resources, and because we saw extensive changes in the distribution of growth from month to month, we used survey data of the control points as our ground-truthing, and compared analysis to the images rather than a comparison to field conditions. With more resources and longer access to the field site, we’d like to supplement this approach by surveying random patches in the stream for groundtruthing. Language was added to the discussion to reflect this area of improvement.

Point 8: Table 2: possible underestimation of Open to Canopy?

Given the 16% area couldn’t be analyzed, it’s possible that the proportion of open to canopy is underestimated. However, we qualitatively observed that as vegetation patches seemed to become more established, and flow was channelized between these patches, the higher velocities seemed to entrain and remove some vegetation. Without additional data describing the hydraulic dynamics, we did not feel that we could defensibly speculate one way or another about this trend.

Point 9:  Line 217:Change “steading” to “steady”

Revision accepted; change made in text.

Point 10: Line 242: ??dates??

Text corrected to include dates.

Point 11: Line 255: Replace "near" with "for"

Revision accepted; change made in text.

Point 12: Line 355: Add a d to “require”

Revision accepted; change made in text.

Reviewer 2 Report

I find the article interesting and suitable for publication. It refers to one of the elementary problems of modern hydrological research (but also other scientific disciplines such as biology, ecology, water management, etc.), namely changes in thermal conditions of inland waters. The issues are particularly important in the context of the observed climate changes. In my opinion, the methodology adopted in the paper has a broad applicative potential, and can be employed in measures aimed at limiting the effect of global warming on the thermal regime of rivers. The paper has a transparent structure, and its layout constitutes a logical coherent whole. The paper should be slightly modified before publishing by:

a)    Fig. 1. adding coordinates or a fragment in another scale (currently

this is unclear)

b)    Fig. 2. no legend

c)     Line 82 supplementing literature confirming the statement:

Global river temperatures and sensitivity to atmospheric warming and changes in river flow, Water Resources Research, 2011

Long-term water temperature fluctuations in coastal rivers (Southern Baltic) in Poland, Bulletin of Geography. Physical Geography Series, 2016

Changing river temperatures in northern Germany: trends and drivers of change, Hydrological Processes, 2016

d)    Part of the discussion should be expanded by other research concerning the effect of forest cover on water temperature:

Integrating process-based flow and temperature models to assess riparian forests and temperature amelioration in salmon streams, Hydrological Processes, 2018

Effects of catchment area forestation on the temperature of river waters, Forest Research Paper, 2017

The role of riparian vegetation density, channel orientation and water velocity in determining river temperature dynamics, Journal of Hydrology, 2017

Author Response

Dear Reviewer 2,

Thank you for your comments and revisions on our submission. We have addressed each point in the text, and include our responses to each point below. Thank you for your time and attention to detail; we feel your changes have greatly improved our manuscript.

Point 1: Figure 1: adding coordinates or a fragment in another scale (currently this is unclear)

Revision accepted. Approximate centroid coordinates were added to the caption.

Point 2: Figure 2: No legend

Revision accepted; legend added to Figure 2

Point 3: Line 82: supplementing literature confirming the statement

Revision accepted; also included Null et al. (2013), as that study helps anchor the point in this study’s general geographic context.

Point 4: Part of the discussion should be expanded by other research concerning the effect of forest cover on water temperature:

Revision accepted; discussion expanded in section 4.2 to include concepts such as micro-climate effects, climate change vs. canopy, and broader implications for salmon/cold-water species recovery.

Reviewer 3 Report

The major scientific content of this work has been published previously in different forms, couple good examples of this:

the author refer for 15 times to the reference number 15 “Willis, A.D.; Nichols, A.L.; Holmes, E.J.; Jeffres, C.A.; Fowler, A.C.; Babcock, C.A.; Deas, M.L. Seasonal aquatic macrophytes reduce water temperatures via a riverine canopy in a spring-fed stream. Freshwater Science 2017, 36, 508-522.” , this reference is belong to the same main Author for the current proposed work. imagine 15 times referring to a published paper , in another word the reference number 15 is almost the same current work with slight modification.

The reference 28 has been mentioned 5 times which is related to the same main author.

The reference number 27 is available in the “reference section” but never available at the content. which is also belong to the same author 

The Fig.1 has no reference , this figure has been mentioned many times in several online published paper most of them by the same author ,however the author even didn’t justify how and what is the reference for this figure.

other finding regarding to the funding the research , the Author claimed that no fund has been recieved for this paper , while this work has been originally funded by several partners.  the Author has been published a report in October 24, 2017 https://watershed.ucdavis.edu/files/biblio/Willis%20et%20al.%202017%20Riverine%20canopy.pdf

holding again the same figures and similar content mentioned the funder and partners for the same current work as followings:
• The Nature Conservancy
• California Trout
• U.S. Bureau of Reclamation
• State Water Resources Control Board
• National Fish and Wildlife Foundation
• National Oceanic and Atmospheric Administration • Wildlife Conservation Board 

• Research Partners:
• The Nature Conservancy
• California Trout
• Watercourse Engineering, Inc. 

Scientifically, the methods, figures, tables, references all have been repeated and published several times. 

Author Response

Dear Reviewer 3:

Thank you for your comments. Respectfully, we disagree with your review, and appreciate the opportunity to respond. We provide a point-by-point rebuttal to each comment below:

Point 1: 

The major scientific content of this work has been published previously in different forms, couple good examples of this:

the author refer for 15 times to the reference number 15 “Willis, A.D.; Nichols, A.L.; Holmes, E.J.; Jeffres, C.A.; Fowler, A.C.; Babcock, C.A.; Deas, M.L. Seasonal aquatic macrophytes reduce water temperatures via a riverine canopy in a spring-fed stream. Freshwater Science 2017, 36, 508-522.” , this reference is belong to the same main Author for the current proposed work. imagine 15 times referring to a published paper , in another word the reference number 15 is almost the same current work with slight modification.

The major scientific content of this work is the use of UAV technology to enhance monitoring of the seasonal trends in riverine canopy growth, as indicated by the title. This methodology is as-yet unpublished in any work, as we discuss in detail in 4.1 UAV survey methods. As we and other reviewers of this manuscript have observed, the methodology presented in this paper has broad applications, particularly in the area of thermal process management to mitigate for climate change. Willis et al. (2017) merely identifies the riverine canopy as a novel contributing factor to the overall thermal process,  but fails to explore both the seasonal dynamics of this process and modeling considerations to sufficiently simulate the effects of canopy on heat flux for management applications. Those recommendations for future work formed the basis for this study, but do not replicate the current work in any way. Other studies, such as Garner et al. (2017), commonly reference previous studies extensively through the manuscript and cite them as a source of either materials or data. These citations are critical to cite the source of data not collected during the course of this work, and to clearly distinguish between the goals of previous work versus current studies.

Finally, as other reviewers have similarly noted, while literature on riparian canopies is extensive, literature reviewing riverine canopies is limited. We are aware of only a handful of papers, cited in this submission, that explore the effects of aquatic plants on thermal process; of those papers, only Willis et al. (2017) studies the effects of emergent plants. As this area of research expands, we look forward to reading the research and analysis completed by other scientists as a welcome development beyond our own work.

Point 2: The reference 28 has been mentioned 5 times which is related to the same main author.

Nichols et al. (2014) refers to reach-scale temperature dynamics that are controlled by upstream, local thermal regimes. This reference provides landscape context for the current manuscript, illustrates potential broader applications, and contains results critical to the interpretation of the current study. We similarly reference other works unrelated to our publications with similar broad, distinct relevance to this study (e.g., references 2, 3, 8, 9, 31)  Referencing previous studies with relevant implications for current work is common, as shown by myriad of other works, including Garner et al. (2017)., Isaak et al. (2017), Hannah and Garner (2015), and is critical to appropriately crediting other work.

Point 3: The reference number 27 is available in the “reference section” but never available at the content. which is also belong to the same author

Reference 27, Willis et al. (2015), has been updated to reference 32 given revisions to the literature review. This reference is cited on line 92, in which we summarize previous studies that have included water temperature modeling and analysis in the region: "Extensive temperature modelling and analysis has occurred in the lower Shasta watershed to explore past and current stream conditions as they relate to salmon [15,31-33]."

Point 4: The Fig.1 has no reference , this figure has been mentioned many times in several online published paper most of them by the same author ,however the author even didn’t justify how and what is the reference for this figure.

Figure 1 is referenced twice in the text of the manuscript (lines 115 and 181) to illustrate the spatial array of monitoring work as well as analysis points for the modeling work. While the style of this figure is similar to others produced in previous work (i.e., black and white line illustrations), each figure is unique to their respective manuscript as the monitoring and analytical work are unique to each manuscript. 

Point 5: 

other finding regarding to the funding the research , the Author claimed that no fund has been recieved for this paper , while this work has been originally funded by several partners.  the Author has been published a report in October 24, 2017 https://watershed.ucdavis.edu/files/biblio/Willis%20et%20al.%202017%20Riverine%20canopy.pdf

holding again the same figures and similar content mentioned the funder and partners for the same current work as followings:

• The Nature Conservancy

• California Trout

• U.S. Bureau of Reclamation

• State Water Resources Control Board

• National Fish and Wildlife Foundation

• National Oceanic and Atmospheric Administration • Wildlife Conservation Board

• Research Partners:

• The Nature Conservancy

• California Trout

• Watercourse Engineering, Inc.

Previous studies in the same study area were supported by outside funders, who were appropriately credited for their support. The work developed in this study followed the completion of those studies, and was not financially supported by any third party. We used results from previous studies that are publicly available, including modeling files, and have been similarly provided to other entities upon request - an open source requirement of almost all peer-reviewed research, but one that does not constitute financial support to those utilizing the publicly available data. To clearly indicate (and credit) where previous work was used, we have been careful to cite all instances extensively through the manuscript. As well as these detailed citations, we have credited all entities who provided either access, materials, or technical support to this work in our acknowledgements section to clearly show how this work was supported beyond our own personal efforts.

Point 6: Scientifically, the methods, figures, tables, references all have been repeated and published several times.

Respectfully, we and the other anonymous reviewers of this study disagree with that characterization. The use of UAV technology for areal monitoring of aquatic plants in mid-sized streams is as-yet unpublished methodology, and constitutes the main original contribution of this paper. While process-based temperature modeling itself is not new, the analysis of the resolution needed to accurately replicate riverine-canopy-based thermal regimes is novel and a valid exploration of the monitoring methodology. The methods, figures, and tables are unique to this manuscript; considering the authors personally collected and completed the analysis ourselves, and have not shared even preliminary findings, we would be alarmed to find our work replicated in any other publication prior to the publication of this manuscript. The references have been published several times as they constitute important, foundational, publications in their field. Many papers have comprehensively reviewed previous work on thermal regimes in streams, process-based temperature modelling, and management implications to build a foundation for their own, unique work, and have similarly cited the references included in this manuscript.