General comment
I believe that the authors improved the original version of the manuscript by clarifying some important aspects of their analysis. Still however I believe that there are some important issues that the authors should improve and clarify, including some requests of my previous review that were not addressed. In general, grammar, syntax, and equation notation should be carefully checked throughout the manuscript and I suggest that the structure of some sections should be revised. I believe that the analysis and the quality of the manuscript can be substantially improved in several aspects, as suggested in my specific comments below.
Introduction:
- Line 20: I would not mention riparian vegetation together with meteorological forcing. Rather, I would move it to point 3, after opportune adjustments. In addition, I believe that, given the focus of the study, anthropogenic effects should be explicitly mentioned in a specific point.
- Lines 26-28: please revise this sentence: it seems that fluxes and boundary conditions are two distinct entities, while in many cases boundary conditions are fluxes. This sentence confirms that riparian vegetation should be removed from point 1 (it is not a flux).
- Line 29: does the term "parameter" refer to the fluxes mentioned in the previous sentence? If so, since fluxes are not parameters, I suggest using e.g., the word "terms".
- Line 34: I do not understand the use of the term "analytic" here.
- Line 37: please start a new paragraph after "... to the Rhine river."
- Section 1.1: it is uncommon to have a subsection in the Introduction and I do not think that it is needed here, thus I strongly recommend to removing subsection 1.1. In doing so, a connection sentence between the first half of the Introduction and the second half will be required. As commented in my previous review, this second part of the Introduction reads more as a paragraph of Material and methods. I strongly recommend to fully revise the structure of this section focusing on presenting the objectives of the manuscript and an outline of the approach followed by the authors. Both points are not sufficiently addressed in the present version of the Introduction. Some specific comments are provided below:
* Lines 57-59: this sentence is unclear and the link between the concept of Tc and the assessment of the impact of industry, meteorology and hydrology is confused and unsettled. The syntax should be revised (e.g., "combine ideas from")
* Lines 61-62: I would not call the period 1979 to 2018 a "scenario"
* Lines 66-67: I do not agree with this sentence: Ta does not take into account the origin of water. This is particularly evident for ground water sources. This comment should be removed from the manuscript.
* Lines 70-72: this sentence is debatable since the hybrid model cited by the authors has been already used to evaluate the separate effects of anthropogenic and climate changes (Cai et al., 2018) and I have some concerns on affirming that a simple linear regression model (eq 1) can "allow for a clear distinction between meteorological, hydrological, and anthropogenic input".
* Figure 2 (the study site) should be probably cited here.
- As pointed out in the first revision round, I believe that the Introduction would benefit from mentioning existing literature on the assessment of anthropogenic impact on river water temperature (e.g., Cai et al., 2018; Gaudard et al., 2018; Råman Vinnå et al., 2018, just to cite some recent papers).
Section 2.1.
- Line 83-84: "reference" --> "data provider"/"data source" ?
- Lines 90-92: please, check the verb tenses. The use of "by us" is not recommended.
Section 2.3.
- As commented in the first revision round, the sentences in the first part of this section are too fragmentated and short.
The authors should explain in detail how they aggregated the heat input due to each NPP to obtain the overall heat input at each gauging station.
Section 2.4
- This section is disconnected from the previous, since the authors did not mention the use of GDP in the preceding part of the manuscript. This is something that should be mentioned in the Introduction, where the authors should properly (i.e., concisely but clearly and exhaustively) introduce objectives and approaches of their study.
- Figure 2: I would present the catchment closed at the most downstream gauging station. In this way the entire region analyzed in the study would be presented. Please substitute the symbol used for the Fessenheim NPP as it can be confused with a gauging station.
- Line 140: The use of subsections, subsubsections (e.g., 2.3.1) and paragraphs (e.g. Accumulation) is not harmonized. Please check it throughout the manuscript.
- Line 142: please revise the syntax ("... the the ..."; "... this very grid point")
Section 2.7
- In my view, part of subsection 1.1 (including equation 1) should be moved here. In the Introduction the authors can easily comment on their approach without showing the equation. Indeed, as commented above, in the Introduction the authors should focus on objectives and approaches used in their study in a concise and clear way.
More important, as commented in my previous revision, I do not agree that "The offset a1 (RBT) combines all other influences, which are controlled by anthropogenic sources". a1 accounts to much more than solely anthropogenic sources as it summarizes all contributions that are not directly linked to Ta and Q (groundwater inflow, geothermal flux, vegetation shading, tributary heat flux, upstream heat flux, etc). Please see Segura et al (2015) for an useful overview on the physical meaning of the intercept a1. Still, the authors can suppose that if all the aforementioned conditions are kept unchanged, changes in a1 can be linked to changes in anthropogenic sources (as they comment at lines 267-270). This is conceptually and formally different. The authors should carefully avoid any misunderstanding around the meaning of a1 (adjust also Introduction, lines 265-, Conclusions and all sentences where the meaning of a1 is commented).
- Line 50: "over the whole" --> "over the whole catchment"?
- Equation 4: I would have expected to see "Tc(t0-Delta t(x,y))" and "Q(x0,y0,t0)". Is this correct? As for the first point: since, as far as I understand, Tc is spatially averaged it should not depend on x and y. In addition, to predict Tw at time t0 at the closing section, Tc should be taken at time t0-Delta t(x,y) for the i-th cell located in (x,y) to account for the time delay due to water routing. Considering t0+Delta t is coherent with eq 8, where the authors define Delta t as a negative number, but this comes later in the text and in my view is misunderstandable.
- Equations 5 and 6: similarly to above, the time lag is not correctly accounted for: "T(x0,y0,t0) = Ta(x0,y0,t0 -Delta t)" and in the second equation the integral should be from i=t0-Delta t to i=t0. Delta t can be confused with the time lag due to routing while here it refers to the lag due to water inertia. Moreover, here and in the following equations “T” is not defined. The authors should avoid any misunderstanding, be precise in the use of notation, and carefully check all equations.
- Line 165: "reason reason". There are several other points where the syntax should be checked. I will not comment further on this, but strongly encourage the authors to thoroughly check the entire manuscript.
- Lines 165-176: this part is confused. This is what has been already said in eq 4 and the reference to the 8-day lag is unclear and probably unnecessary: from the text one understands that only the routing lag has been considered, while the time lag due to thermal inertia has not. Overall, this whole section requires significant improvement and clarification.
- Lines 183-186: the authors should better explain how the flow speed has been estimated. Do they mean RMSE between observed and simulated Tw using the Time lag + weight model (or just considering the Time lag, or Time lag + weight + ACC)? Is this value of flow speed confirmed also when analyzing the other gauging stations? The authors should show (at least in the supplementary material) the relationship between RMSE and flow speed for all gauging stations. Is the minimum in RMSE clear and unequivocal? Did the authors optimize independently the flow speed when testing the different definitions of Tc?
- Table 4: "which are statistical significant only if R2 > 1.99" there should be an error here. Please, specify that R2 refer to the linear trends. As suggested in my previous revision, I believe that the authors could add the Pearson coefficient between Ta and Tw to further (and more robustly) support their reasoning.
- Lines 238-241: this paragraph is chaotic. Some concepts are repeated and not necessary in this context (e.g., in Base Ta and Tw show similar behavior).
Section 3.2
- Line 246: "catchment-wide Delta t" the use of catchment-wide is probably not appropriate here as it could be understood as a constant Delta t for the entire catchment.
- Line 250: see my comment relative to Lines 183-186. Here and below: NCS-->NSC
- Line 255: I would not say that a figure is the reason of the results, but that the content of the figure can explain the results.
Section 3.3
- Lines 275-276: actually RBT seems to decline some years in advance (in 1995 looking at the running mean in Fig. 7). Can the authors add a comment on this?
- Table 6: here the authors do not show that the two series of Delta RBT have "similar trends" since the results in the table only depend on initial and final values (which has clear side effects and limitations). To test if the trends are similar, they should calculate Delta RBT in continuous with eq 2 and using the time series of the heat input, and compare it with the time series of RBT shown in Figure 7. This is something that I already suggested in my first review and that I strongly recommend adding to the analysis. While the authors replied that they wanted to pick the largest Delta HI to avoid influences by short term trends, I believe that analyzing the time series is needed to properly show that the two time series have comparable medium- to long-time trends (short term trends can be easily filtered e.g., with a moving average if required).
- Lines 285-317: as commented above, the term RBT summarizes several contributions besides anthropogenic effects. This should be properly recalled in this section, since this is the most likely reason of the differences between RBT and GBT trends (including the specific cases commented by the authors), although I agree that a significant correlation is visible in figure 9.
As commented in my previous review, the comment on the effect of lakes is somehow disconnected from the rest of the paragraph. I perfectly understand and agree that "finding the reason [of the peculiar trend in Basel] is not in the scope of this paper", but the sentences at lines 289-293 should be better contextualized. The reasoning on stratification is fine, the fact that deep water temperature is somehow decoupled from Ta is fine as well (but only for deep lakes), however this does not explains the trend shown in Basel nor the reader knows if such lakes contribute to the Rhine through surface water (natural lakes) or deep water (hydropower reservoirs with deep intake). Better than trying to draft a possible explanation without effectively providing the proper information and hypothesis, would be to simply write that the trend in Basel cannot be explained in this analysis.
Conclusions:
- Line 340: the term "reanalysis" and “forecast" are probably not the most appropriate here.
- Line 342: fluxes are not parameters. I would say that fully physical models requires all meteorological data in input.
- Lines 343-348: the comment on tropical and subtropical rivers is somehow disconnected from the study presented by the authors and personally I do not believe that it is appropriate here. I do not believe that the study by Morril et al (2005) suggests that "this case study of the Rhine can be applied globally". On the contrary the comment on the possible coupling with catchment-wide hydrological models is more significant but poorly examined (please improve it, adding a comment also on the limitations of the analysis).
References
Cai H., Piccolroaz S., Huang J., et al. 2018. Quantifying the impact of the Three Gorges Dam on the thermal dynamics of the Yangtze River. Environ Res Lett.;13(2018):054016.
Gaudard, A, Weber, C, Alexander, TJ, Hunziker, S, Schmid, M. 2018. Impacts of using lakes and rivers for extraction and disposal of heat. WIREs Water. 5:e1295.
Råman Vinnå L., Wüest A., Zappa M., Fink G., Bouffard, D. 2018. Tributaries affect the thermal response of lakes to climate change, Hydrol. Earth Syst. Sci., 22, 31–51
Segura, C., Caldwell, P., Sun, G., McNulty, S., and Zhang, Y. 2015. A model to predict stream water temperature across the conterminous USA. Hydrol. Process., 29, 2178– 2195. |
- Printer-friendly version
- Supplement
