the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Mar 2024
On the Influence of Spatial Heterogeneity of Runoff Generation on the Distributed Unit Hydrograph for Flood Prediction
Abstract. The spatial scale mismatch between runoff generation and runoff routing is an acceptable compromise but a common issue in challenging hydrological modelling. Moreover, there is hardly any report available on whether unit hydrograph (UH) that is consistent with the spatial scale of runoff generation can be computed. The objective of this study was to explore the influence of spatial heterogeneity of runoff generation on the UH for flood prediction. To this end, a novel GIS-based dynamic time-varying unit hydrograph (DTDUH) was proposed based on the time-varying unit hydrograph (TDUH). The DTDUH can be defined as a typical hydrograph of direct runoff which gets generated from one centimeter of effective rainfall falling at a uniform rate over the saturated drainage basin uniformly during a specific duration. The DTDUH was computed based on the saturated areas of the watershed instead of the global watershed. The saturated areas were extracted based on the TWI. Finally, the Longhu River basin and Dongshi River basin were selected as two case studies. Results showed that the proposed method exhibited consistent or better performance compared with that of the linear reservoir routing method, and performed better than the TDUH method. Specifically, the DTDUH method indicated good performances for the flood events with low antecedent soil moisture, and it performed consistently with the TDUH when the global watershed is nearly saturated. The proposed method can be used for the watersheds with sparse gauging stations and limited observed rainfall and runoff data, as is the same with the TDUH method. Simultaneously, it is well applicable for the humid or mountain watershed where the saturation-excess rainfall is the dominant.
- Preprint
(2109 KB) - Metadata XML
-
Supplement
(139 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on hess-2024-51', Ilhan Özgen-Xian, 03 May 2024
Yi et al. propose a novel method for computing unit hydrographs that can better account for spatial heterogeneity. For this, they relate the unit hydrograph to the dynamics of the saturated area inside the catchment, computed through the topographic wetness index for unit precipitation. The exploration of unit hydrographs and hydrological response units is an important research direction in hydrological modelling. Here, especially the impact of spatial heterogeneity is---in my opinion---underexplored. The topic of the paper is thus timely and relevant to the readership of this journal.
My main issue is with the presentation of the results and subsequent discussion. I hope that this is not nitpicking, but the current organisation of the manuscript made reading, understanding, and evaluating the novelty and the methodology difficult to me. I therefore suggest revision as discussed below. Because the revision might become substantial, I recommend major revision of the current manuscript.
Minor comments
L11: "... challenging hydrological modelling" reads a bit awkward. Rethink or remove "challenging."
Table 6: The phrase "A typical hydrograph of direct runoff which gets generated from one centimetre of effective rainfall falling at a uniform rate over the saturated drainage basin uniformly during a specific duration." is not very clear to me. Does "uniformly" mean "spatially uniform" as implied in the Assumptions?
The discussion
It may always be debatable whether a certain part goes into the Discussion or somewhere else. Nevertheless, I have some suggestions. In my opinion, the Discussion section should focus on discussing the presented data and results and perhaps generalise some key insights if the data permits. The authors do this, for example, in L562-565 in the Conclusions. Based on this reasoning, I suggest a thorough rewrite of the discussion in this paper. The aim of this rewrite should be to support the discussed points with data generated in the test cases.
Specifically, I have the following comments:
L498: The title "Forecasting performance advantage analysis of the proposed DTDUH" sounds a bit strange. I think it is the advantage over the TDUH algorithm? If this is the case, it should be in the title. But the advantage is not very clear to me. It is related to how the water is redistributed and how it connects to the unit hydrograph, but there is no indication what the reference hydrograph should look like.
L501-509 including Table 6: This reads as introductory information rather than a discussion. Perhaps move it into the Introduction of the paper or make the connection to the test cases more obvious to the reader.
L512-530 including Figure 15: This entire discussion is disconnected from the test cases. I have no doubt the discussion is valid, but I wonder if this could be better connected to the test cases. Instead of the idealised basin with 24 cells, can you show and discuss these effects in the test cases you have shown? This would be the best option. Otherwise, include a modelling test case with the idealised basin at the beginning of your test cases and then discuss it in the Discussion section where you show these effects.
L531-549: This section is also not supported by data shown. These generalised shortcomings seem to belong in the conclusions. I suggest discussing these limitations in the context of the test cases you show. For example, if the unit hydrograph in your test case deviates from the reference hydrograph, you may show that this is most likely due to hybrid runoff generation. This would support the discussed points in this section.
Citation: https://doi.org/10.5194/hess-2024-51-RC1 - AC2: 'Reply on RC1', Bin Yi, 14 Jul 2024
-
RC2: 'Comment on hess-2024-51', Anonymous Referee #2, 28 May 2024
This paper deals with rainfall-runoff modelling and proposes a new routing scheme adapted from the principle of geomorphological and distributed unit hydrographs (Rodriguez-Iturbe and Valdès, 1979; Maidment et al., 1996). The new proposed scheme is called Dynamic Time Varying Unit Hydrographs, since it combines i) a variation of flow velocities depending on the rainfall intensities and the saturation level of the basin, and ii) a variation of the flow contributing areas considered to generate the unit hydrograph, which are assumed to correspond to the saturated areas. This basic assumption of runoff dominated by saturated areas of course limits the potential of application of the method (contributions from saturated areas should dominate the other flow generation processes), but this limitation is clearly stated in the conclusions of the manuscript, and this assumption seems to be valid in the presented case studies (humid basins).
The topic of this paper is very relevant to HESS, and the proposed method brings significant novelty. Unfortunately, I think that the presentation of the methods and results should still be largely improved. Moreover, the methodology proposed for evaluation does not appear relevant to me, and does not sufficiently support the conclusions of this work, in my opinion. Particularly, the manuscript should be improved on the following aspects:
- the description of the hydrological model is not sufficiently clear (consistency and links between equations, variables and parameters, description of parameters to be calibrated, ..). See my detailed remarks below
- the principle to replace only the routing scheme for the surface runoff part of the model is presented too late. This should be stated at the beginning, and the routing schemes used for other (subsurface) components should be at least quickly presented.
- the methodological choice to use only the XAJ+LR calibration for the evaluation of the XAJ+TDUH and XAJ+DTDUH models does not appear relevant to me. This leads logically to better performances of the XAJ+LR models, and may also advantage one of the two other XAJ+TDUH and XAJ+DTDUH models. The calibration results of the XAJ+TDUH and XAJ+DTDUH models (appendix) suggest that DTDUH does not perform as well as expected. I think that to achieve robust conclusions all the models should be calibrated and evaluated in validation (with a comparison of calibrated parameters).
- Some of the presented results appear to be inconsistent: for instance, the time to peak of DTDUHs presented on figures 6 and 8 do not vary with the saturation level alpha, which does not seem consistent with eq.9 ; also, the DTDUHs on figures 6 and 8 show areas under curves which largely vary with the level of saturation alpha of the basins, whereas on figures 9 and 10 the areas under curves do not vary anymore with the saturation level.
Considering all these weaknesses, I would not recommend to publish this paper in its current form.
Detailed comments:
- l. 10 rather « a common challenging issue in ..”
- l. 25 “as for the TDUH method”
- l. 48 specify here “the watershed response to efficient rainfall”, since the response to rainfall (including the transformation to efficient rainfall) is obviously not linear..
- l. 55 “high intensity of rainfall”
- l. 66 “have attracted much attention”
- l. 83 “the approximations”
- l. 93 “This raised the question whether ..”
- l. 103-105 I think a reference to the work of Andrieu et al. (2021), who proposed e-GUIHs accounting for the spatial variability of rainfall (and thus indirectly the spatial variability of efficient rainfall) would be relevant here.
- l.110 rather “the runoff generating areas”
- l.111-112 I have some difficulties with this sentence: what do you mean by “unify the spatial scales of the runoff generation and the confluence method” ? Please reformulate
- l. 113 “Finally, .. “ Please remove the word finally since the case studies are directly linked to the questions and evaluation presented before.
- l.115-116 A short presentation of the structure of the paper would be good here.
- l.118-120 Please moderate this statement, since subsurface flows can also be a significant source of efficient rainfall, when infiltration capacities are larger than rainfall intensities.
- l.128 – 153 This description of the production part of the hydrological model is not clear. Please define more clearly each notion and variable (Aps, Ap, Af and As, free water storage and tension water storage, AU, FR, BU, S, SM, ..), and the way they are used and related, if necessary by including additional figures. Figure 2 could be placed here, but it does not provide a sufficient level of detail to understand how the model’s variables and parameters are related. Finally, I do not understand how the areas Aps, Ap, Af, As are used in the computation, and how eq. 3 and equations 4, 5, 6 can be combined to ensure a conservation of volume (I guess R should be equal to RS+RI+RG but this does not seem consistent with the equations)
- l. 173 “the total number of grid cells”
- l.175 “The depth of the excess rainfall occurs only in the saturated areas when the entire basin does not reach a global saturated state”: this appears as a theory that is only rarely valid.
- l.186 the presence of the alpha (soil moisture) variable in eq.9 corresponds to the assumption that velocities vary with the global soil moisture. This also suggests that flow generation does not only correspond to saturation excess.
- l.213-215 “To compare the differences …” Again here I feel this sentence not very clear, please could you reformulate this.
- l.215-216 I do not see any application of the muskingum method in the presented routing schemes.
Figure 2. According to this figure and the text, it seems that the routing schemes are applied only on RS (surface runoff) . Could you explain what happens with subsurface and groudwater components (RI and RG): are they also routed and how?
- l.224 – 230 Again here, it is suggested that the discharge at the basin outlet is computed by routing only the surface runoff and is resumed to QS. Can you at least explain if Qi and QG are neglected and if the validity of this assumption has been verified for the presented case studies?
- l.237 – 238 “aimed at maximizing flow characteristics”: not clear , please reformulate.
- l.248 – 255 Could you justify here the choice of different parameters for evaluation and for calibration?
- l.260 - 261 According to the mean slopes, the basins do not appear to be very mountainous.
- l.262 – 263 Please mention here the names of institutions providing data, in addition to URLs.
- l.269 – 271 Here again, could you mention the data providers for rainfall and discharge data?
- l.271 – 272 Do you mean here that the model was calibrated only on flood periods?
- l.271 – 272 and table 1 Could you mention which events were used for calibration and which ones for verification?
- l.295 – 286 It is rather surprising here that the routing parameters Ic and gamma do not vary from one event to another, and seem to have been calibrated formerly, i.e. independently of the calibration of the production part of the model. Could you better justify this choice?
- l.297 – 297 This statement is very important to understand the methodological choices. This should have been mentioned at the beginning of this paper.
- l.300 – 301 “and thus the parameters of the runoff generation module kept unchanged” This statement is not consistent with explanations provided at lines 306-316, which suggest that the models were calibrated separately. Maybe it would be better to include this statement in lines 317-324 where the explanation is provided.
- l. 306-324 The methodological choice to keep only the XAJ+LR calibration results, for the evaluation of the two other models (XAJ+TDUH and XAJ+DTDUH) seems very surprising to me, since the calibration results resented in appendix suggest that these two model may perform similarly to the XAJ+LR model, when appropriately calibrated. Moreover, the choice to calibrate the models on the whole dataset (without preserving validation data) is also curious. I think it would have been more relevant to calibrate the three models, and to preserve a validation datset and/or to provide cross validation results.
- l.337-347 Could you explain here how do you retrieve the value of alpha_t based on the structure of the model presented in figure 2? Could you also mention how the value of alpha_t is converted to a saturated surface: assumption that alpha_t=x corresponds to a x% of saturated cells in the watershed (corresponding to the x% cells having the largest TWI values) ?
- l.356-365 and figures 5 to 8: it is rather surprising here that the time to peak of DTDUHs do not vary, depending on alpha_t values, since the eq. 9 used for the computation of velocities still integrates the value of alpha_t. I guess here that for DTDUH the velocities are computed from an equation that differs from eq. 9. Could you clarify this?
- Figure 9 and 10 these figures appear inconsistent with figure 6 and 8, since the shape of DTDUHs do not significantly vary here between S1 and S4, whereas large variations in the areas under the curves are observed on figures 6 and 8 (which is justified in lines 360-361 by the fact that DTDUHs are derived only from saturated areas)
- l.404—407 I feel significant differences are observed for both basins, even if more limited for Dongshi basin
- Figure 11 I understand here that the three flow components of the model (surface and two subsurface ones) were kept for the comparison to observed hydrographs. Could you provide information on the routing scheme used for subsurface flows? Could you also provide information about the alpha_t values corresponding to antecedent soil moisture conditions in figure 11.b?
- Figures 12 and 14: what appears clearly here is the advantage given to the LR model that has been calibrated. I think this reflects the limits of the chosen methodology. It would have been better to split the dataset in calibration/validation, and to compare the calibrated versions of the three models
- Table 6 It should be mentioned more clearly here what is meant by “current theory”
- l.512-522 and figure 15: this development should be found in the methodology section since it illustrates the computation of DTDUHs
- l.562-565 Unfortunately, these results are not shown (appendix not present)
- l.566-568 I think keeping the same parameters set for the runoff generation module (calibrated with another routing scheme) is not adapted here, since it corresponds more or less to an absence of calibration. This choice may advantage one of the both routing schemes. Providing calibration / validation results for both models (together with a comparison of calibrated parameters sets), would be more relevant in my opinion.
Citation: https://doi.org/10.5194/hess-2024-51-RC2 - AC1: 'Reply on RC2', Bin Yi, 14 Jul 2024
- AC3: 'Reply on RC2', Bin Yi, 14 Jul 2024
Status: closed
-
RC1: 'Comment on hess-2024-51', Ilhan Özgen-Xian, 03 May 2024
Yi et al. propose a novel method for computing unit hydrographs that can better account for spatial heterogeneity. For this, they relate the unit hydrograph to the dynamics of the saturated area inside the catchment, computed through the topographic wetness index for unit precipitation. The exploration of unit hydrographs and hydrological response units is an important research direction in hydrological modelling. Here, especially the impact of spatial heterogeneity is---in my opinion---underexplored. The topic of the paper is thus timely and relevant to the readership of this journal.
My main issue is with the presentation of the results and subsequent discussion. I hope that this is not nitpicking, but the current organisation of the manuscript made reading, understanding, and evaluating the novelty and the methodology difficult to me. I therefore suggest revision as discussed below. Because the revision might become substantial, I recommend major revision of the current manuscript.
Minor comments
L11: "... challenging hydrological modelling" reads a bit awkward. Rethink or remove "challenging."
Table 6: The phrase "A typical hydrograph of direct runoff which gets generated from one centimetre of effective rainfall falling at a uniform rate over the saturated drainage basin uniformly during a specific duration." is not very clear to me. Does "uniformly" mean "spatially uniform" as implied in the Assumptions?
The discussion
It may always be debatable whether a certain part goes into the Discussion or somewhere else. Nevertheless, I have some suggestions. In my opinion, the Discussion section should focus on discussing the presented data and results and perhaps generalise some key insights if the data permits. The authors do this, for example, in L562-565 in the Conclusions. Based on this reasoning, I suggest a thorough rewrite of the discussion in this paper. The aim of this rewrite should be to support the discussed points with data generated in the test cases.
Specifically, I have the following comments:
L498: The title "Forecasting performance advantage analysis of the proposed DTDUH" sounds a bit strange. I think it is the advantage over the TDUH algorithm? If this is the case, it should be in the title. But the advantage is not very clear to me. It is related to how the water is redistributed and how it connects to the unit hydrograph, but there is no indication what the reference hydrograph should look like.
L501-509 including Table 6: This reads as introductory information rather than a discussion. Perhaps move it into the Introduction of the paper or make the connection to the test cases more obvious to the reader.
L512-530 including Figure 15: This entire discussion is disconnected from the test cases. I have no doubt the discussion is valid, but I wonder if this could be better connected to the test cases. Instead of the idealised basin with 24 cells, can you show and discuss these effects in the test cases you have shown? This would be the best option. Otherwise, include a modelling test case with the idealised basin at the beginning of your test cases and then discuss it in the Discussion section where you show these effects.
L531-549: This section is also not supported by data shown. These generalised shortcomings seem to belong in the conclusions. I suggest discussing these limitations in the context of the test cases you show. For example, if the unit hydrograph in your test case deviates from the reference hydrograph, you may show that this is most likely due to hybrid runoff generation. This would support the discussed points in this section.
Citation: https://doi.org/10.5194/hess-2024-51-RC1 - AC2: 'Reply on RC1', Bin Yi, 14 Jul 2024
-
RC2: 'Comment on hess-2024-51', Anonymous Referee #2, 28 May 2024
This paper deals with rainfall-runoff modelling and proposes a new routing scheme adapted from the principle of geomorphological and distributed unit hydrographs (Rodriguez-Iturbe and Valdès, 1979; Maidment et al., 1996). The new proposed scheme is called Dynamic Time Varying Unit Hydrographs, since it combines i) a variation of flow velocities depending on the rainfall intensities and the saturation level of the basin, and ii) a variation of the flow contributing areas considered to generate the unit hydrograph, which are assumed to correspond to the saturated areas. This basic assumption of runoff dominated by saturated areas of course limits the potential of application of the method (contributions from saturated areas should dominate the other flow generation processes), but this limitation is clearly stated in the conclusions of the manuscript, and this assumption seems to be valid in the presented case studies (humid basins).
The topic of this paper is very relevant to HESS, and the proposed method brings significant novelty. Unfortunately, I think that the presentation of the methods and results should still be largely improved. Moreover, the methodology proposed for evaluation does not appear relevant to me, and does not sufficiently support the conclusions of this work, in my opinion. Particularly, the manuscript should be improved on the following aspects:
- the description of the hydrological model is not sufficiently clear (consistency and links between equations, variables and parameters, description of parameters to be calibrated, ..). See my detailed remarks below
- the principle to replace only the routing scheme for the surface runoff part of the model is presented too late. This should be stated at the beginning, and the routing schemes used for other (subsurface) components should be at least quickly presented.
- the methodological choice to use only the XAJ+LR calibration for the evaluation of the XAJ+TDUH and XAJ+DTDUH models does not appear relevant to me. This leads logically to better performances of the XAJ+LR models, and may also advantage one of the two other XAJ+TDUH and XAJ+DTDUH models. The calibration results of the XAJ+TDUH and XAJ+DTDUH models (appendix) suggest that DTDUH does not perform as well as expected. I think that to achieve robust conclusions all the models should be calibrated and evaluated in validation (with a comparison of calibrated parameters).
- Some of the presented results appear to be inconsistent: for instance, the time to peak of DTDUHs presented on figures 6 and 8 do not vary with the saturation level alpha, which does not seem consistent with eq.9 ; also, the DTDUHs on figures 6 and 8 show areas under curves which largely vary with the level of saturation alpha of the basins, whereas on figures 9 and 10 the areas under curves do not vary anymore with the saturation level.
Considering all these weaknesses, I would not recommend to publish this paper in its current form.
Detailed comments:
- l. 10 rather « a common challenging issue in ..”
- l. 25 “as for the TDUH method”
- l. 48 specify here “the watershed response to efficient rainfall”, since the response to rainfall (including the transformation to efficient rainfall) is obviously not linear..
- l. 55 “high intensity of rainfall”
- l. 66 “have attracted much attention”
- l. 83 “the approximations”
- l. 93 “This raised the question whether ..”
- l. 103-105 I think a reference to the work of Andrieu et al. (2021), who proposed e-GUIHs accounting for the spatial variability of rainfall (and thus indirectly the spatial variability of efficient rainfall) would be relevant here.
- l.110 rather “the runoff generating areas”
- l.111-112 I have some difficulties with this sentence: what do you mean by “unify the spatial scales of the runoff generation and the confluence method” ? Please reformulate
- l. 113 “Finally, .. “ Please remove the word finally since the case studies are directly linked to the questions and evaluation presented before.
- l.115-116 A short presentation of the structure of the paper would be good here.
- l.118-120 Please moderate this statement, since subsurface flows can also be a significant source of efficient rainfall, when infiltration capacities are larger than rainfall intensities.
- l.128 – 153 This description of the production part of the hydrological model is not clear. Please define more clearly each notion and variable (Aps, Ap, Af and As, free water storage and tension water storage, AU, FR, BU, S, SM, ..), and the way they are used and related, if necessary by including additional figures. Figure 2 could be placed here, but it does not provide a sufficient level of detail to understand how the model’s variables and parameters are related. Finally, I do not understand how the areas Aps, Ap, Af, As are used in the computation, and how eq. 3 and equations 4, 5, 6 can be combined to ensure a conservation of volume (I guess R should be equal to RS+RI+RG but this does not seem consistent with the equations)
- l. 173 “the total number of grid cells”
- l.175 “The depth of the excess rainfall occurs only in the saturated areas when the entire basin does not reach a global saturated state”: this appears as a theory that is only rarely valid.
- l.186 the presence of the alpha (soil moisture) variable in eq.9 corresponds to the assumption that velocities vary with the global soil moisture. This also suggests that flow generation does not only correspond to saturation excess.
- l.213-215 “To compare the differences …” Again here I feel this sentence not very clear, please could you reformulate this.
- l.215-216 I do not see any application of the muskingum method in the presented routing schemes.
Figure 2. According to this figure and the text, it seems that the routing schemes are applied only on RS (surface runoff) . Could you explain what happens with subsurface and groudwater components (RI and RG): are they also routed and how?
- l.224 – 230 Again here, it is suggested that the discharge at the basin outlet is computed by routing only the surface runoff and is resumed to QS. Can you at least explain if Qi and QG are neglected and if the validity of this assumption has been verified for the presented case studies?
- l.237 – 238 “aimed at maximizing flow characteristics”: not clear , please reformulate.
- l.248 – 255 Could you justify here the choice of different parameters for evaluation and for calibration?
- l.260 - 261 According to the mean slopes, the basins do not appear to be very mountainous.
- l.262 – 263 Please mention here the names of institutions providing data, in addition to URLs.
- l.269 – 271 Here again, could you mention the data providers for rainfall and discharge data?
- l.271 – 272 Do you mean here that the model was calibrated only on flood periods?
- l.271 – 272 and table 1 Could you mention which events were used for calibration and which ones for verification?
- l.295 – 286 It is rather surprising here that the routing parameters Ic and gamma do not vary from one event to another, and seem to have been calibrated formerly, i.e. independently of the calibration of the production part of the model. Could you better justify this choice?
- l.297 – 297 This statement is very important to understand the methodological choices. This should have been mentioned at the beginning of this paper.
- l.300 – 301 “and thus the parameters of the runoff generation module kept unchanged” This statement is not consistent with explanations provided at lines 306-316, which suggest that the models were calibrated separately. Maybe it would be better to include this statement in lines 317-324 where the explanation is provided.
- l. 306-324 The methodological choice to keep only the XAJ+LR calibration results, for the evaluation of the two other models (XAJ+TDUH and XAJ+DTDUH) seems very surprising to me, since the calibration results resented in appendix suggest that these two model may perform similarly to the XAJ+LR model, when appropriately calibrated. Moreover, the choice to calibrate the models on the whole dataset (without preserving validation data) is also curious. I think it would have been more relevant to calibrate the three models, and to preserve a validation datset and/or to provide cross validation results.
- l.337-347 Could you explain here how do you retrieve the value of alpha_t based on the structure of the model presented in figure 2? Could you also mention how the value of alpha_t is converted to a saturated surface: assumption that alpha_t=x corresponds to a x% of saturated cells in the watershed (corresponding to the x% cells having the largest TWI values) ?
- l.356-365 and figures 5 to 8: it is rather surprising here that the time to peak of DTDUHs do not vary, depending on alpha_t values, since the eq. 9 used for the computation of velocities still integrates the value of alpha_t. I guess here that for DTDUH the velocities are computed from an equation that differs from eq. 9. Could you clarify this?
- Figure 9 and 10 these figures appear inconsistent with figure 6 and 8, since the shape of DTDUHs do not significantly vary here between S1 and S4, whereas large variations in the areas under the curves are observed on figures 6 and 8 (which is justified in lines 360-361 by the fact that DTDUHs are derived only from saturated areas)
- l.404—407 I feel significant differences are observed for both basins, even if more limited for Dongshi basin
- Figure 11 I understand here that the three flow components of the model (surface and two subsurface ones) were kept for the comparison to observed hydrographs. Could you provide information on the routing scheme used for subsurface flows? Could you also provide information about the alpha_t values corresponding to antecedent soil moisture conditions in figure 11.b?
- Figures 12 and 14: what appears clearly here is the advantage given to the LR model that has been calibrated. I think this reflects the limits of the chosen methodology. It would have been better to split the dataset in calibration/validation, and to compare the calibrated versions of the three models
- Table 6 It should be mentioned more clearly here what is meant by “current theory”
- l.512-522 and figure 15: this development should be found in the methodology section since it illustrates the computation of DTDUHs
- l.562-565 Unfortunately, these results are not shown (appendix not present)
- l.566-568 I think keeping the same parameters set for the runoff generation module (calibrated with another routing scheme) is not adapted here, since it corresponds more or less to an absence of calibration. This choice may advantage one of the both routing schemes. Providing calibration / validation results for both models (together with a comparison of calibrated parameters sets), would be more relevant in my opinion.
Citation: https://doi.org/10.5194/hess-2024-51-RC2 - AC1: 'Reply on RC2', Bin Yi, 14 Jul 2024
- AC3: 'Reply on RC2', Bin Yi, 14 Jul 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 389 | 83 | 41 | 513 | 42 | 35 | 30 |
- HTML: 389
- PDF: 83
- XML: 41
- Total: 513
- Supplement: 42
- BibTeX: 35
- EndNote: 30
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1