the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Flood frequency analysis using mean daily flows vs. instantaneous peak flows
Anne Bartens
Uwe Haberlandt
Abstract. In many cases flood frequency analysis needs to be carried out on mean daily flow (MDF) series without any available information on the instantaneous peak flow (IPF). We analyze the error of using MDFs instead of IPFs for flood quantile estimation on a German dataset and assess spatial patterns and factors that influence the deviation of MDF floods from their IPF counterparts. The main dependence could be found for catchment area but also gauge elevation appeared to have some influence. Based on the findings we propose simple linear models to correct both MDF flood peaks of individual flood events and overall MDF flood statistics. Key predictor in the models is the event-based ratio of flood peak and flood volume obtained directly from the daily flow records. This correction approach requires a minimum of data input, is easily applied, valid for the entire study area and successfully estimates IPF peaks and flood statistics. The models perform particularly well in smaller catchments, where other IPF estimation methods fall short. Still, the limit of the approach is reached for catchment sizes below 100 km2, where the hydrograph information from the daily series is no longer capable of approximating instantaneous flood dynamics.
- Preprint
(1960 KB) - Metadata XML
- BibTeX
- EndNote
Anne Bartens and Uwe Haberlandt
Status: final response (author comments only)
-
RC1: 'Comment on hess-2023-144', Anonymous Referee #1, 05 Jul 2023
The manuscript presents an interesting idea, but the paper is not presented well. There are main issues that have to be addressed. One issue is lack of consistency about the data and methods; e.g., it is unclear which timescale you worked on (daily, monthly annual maximum etc.). There is a lack of proper referencing in Methodology; this section is unclear and needs more information about the details and assumptions (e.g., the main reasons to apply different formulas in each section of it. The sole consideration of drainage area and topography for the analyses is questionable since there are other factors that can derive the mean/peak ratio. This can affect the findings too since a combination of other factors can affect the ratio. I have recommended a few edits and comments in the PDF. Here are additional comments:
- Overall, the writing is mediocre and needs improvements. This is not only about English but the way the story is told.
- Abstract needs to be revised to better discuss the methods and major results. Some quantitative results (numbers) are needed; as of now, everything has been discussed subjectively.
- In the Introduction section, it is not clear which approaches and methods you are addressing in different references. Please clarify. My comments in the PDF can clarify what I specifically refer to.
- Please explicitly state the objective and research gap in the Introduction section.
- In addition to drainage area and topography, what other factors affect the peak/mean ratio? Why do you solely study the two factors?
- Instead of using the three return periods, why not using the historical events for the analyses between flows and drainage area-elevation?
- Methodology section lacks proper references for the assumptions, methods and equations.
- There is a mention of the availability of IPF monthly maximum flows, but monthly is a too large timescale for instantaneous peak flow analyses. What is your justification for using that?
- How were IPFstat and MDFstat Please clarify.
- Equations 2-3 need references. Have they been developed by the authors or others (need references)?
- Any rational for using the GEV distribution?
- Why did you use a linear regression model in Equation 3?
- What is Qsuc in Equation 4?
- In line 92, what does “IPFand MDF do not necessarily overlap” mean? Do you mean in terms of their timing or magnitude? Please clarify.
- Line 143: What does “annual maxima from monthly maximum” mean?
- Line 148: To what extent, does the sample size change the uncertainty? Is 1,000 a sufficient sample size?
- Lines 165 and 189 are inconsistent in terms of the number of discharge stations (648 vs 653). Which one is right?
- Overall, there is a lack of connectivity between the subsections under Results section. This section needs a better flow.
- Section 4.3: It is expected that because the two databases (IPF and MDF) are different, their distribution parameters are different too. What is the main reason for comparing the parameters of distributions? A more proper comparison should be on quantiles (different return periods).
- Figure 13: What is the main reason for similarity among IPF-bs, MDF-bs, LM-bs-full and LM-bs-bs in different HQ years?
- Figure 14: What is the difference between the median of the three HQ-years in each part of the confidence interval? They seem similar in the boxplot median.
- Are all methods and approaches sensitive to the database type? Can those be generalized to other catchments? If so, what are some considerations?
- All acronyms and abbreviations should be spelled out in the keywords, figures, tables and headings.
- Please italicize all parameters and coefficients throughout the text.
I hope the authors find these comments useful in their research. If the authors decide to submit a revision, both sets of my comments, including the above and in the PDF, should be addressed.
-
RC2: 'Comment on hess-2023-144', Anonymous Referee #2, 03 Aug 2023
The manuscript “Flood frequency analysis using mean daily flows vs. instantaneous peak flows” by Anne Bartens and Uwe Haberlandt investigates the difference in using mean daily flow series (MDF) instead of instantaneous peak flow series (IPF) for flood frequency analyses and proposes an approach to correct MDF series and flood statistics, based on the ratio between flood peak and flood volume. The analysis is carried out for several gauges in five German regions. The study is interesting for practical applications. Language is sometimes not precise and unclear, and, in my opinion, it can be improved. I have some major and specific comments to the manuscript. Please find my comments below.
Major comments:
- Eq. 3 vs Eq. 5: in Eq.3 the IPF of the events is obtained by dividing the MDF by the ‘linear model’ while the MDF statistics is multiplied by the linear model in Eq.5. is it a typo (in line 103 the authors say the two models are analogous) or there is a reason behind this difference in the structure of the two corrections? Why do the authors use this correction type? Is the linear regression an appropariate model?
- Terminology: throughout the manuscript the authors refer to the proposed correction method as “linear models” or “linear regression models”. This is somehow confusing. I suggest finding another name for the correction method.
Specific comments:
- Line 145: I disagree that with the bootstrapping/resampling we measure the ‘uncertainty due to distribution fitting’. In my opinion it is the sampling uncertainty / parameter uncertainty. Same in section 4.5.
- Lines 233-236: the authors use both ‘(a)synchronous occurrence’ and ‘temporal overlap’. Do they refer to the same thing or not? If yes, please use consistent terminology. How is the temporal overlap measured/identified?
- Lines 237-238: “…may belong to significantly different events and thus to different populations”. This is not clear in this context and clarification.
- Line 256: I suggest adding “percentage” in front of “change”.
- In the results section there are often reported considerations that would better fit into the discussion section (e.g. lines 274-280).
- It is not always clear which tables / figures refer to the calibration or validation set of gauges. II suggest clarifying this in the figure captions and in the relative text.
- Figure 8: it would be interesting to see a similar figure for IPF vs the corrected MDF
- Figure 9: please specify in the figure caption the type of error shown (i.e. error in the parameters of the distribution)
- Lines 390-391: it is unclear to me.
- Table 8: it is unclear what “mixed-models” stands for. Does it refer to all year?
- Line 406: “light blue points represent 100 resampled model estimates”. It is not clear what is resampled exactly. Do the authors resample the original MDF and then they apply the correction or the other way around? Also in the following line “permutation in the linear models” and in figure 15 is not clear what you resample exactly.
- Terminology: resampling, permutation and bootstrapping are used as synonyms (as far as I understand) but they are not exactly the same. Please clarify and homogenize the terminology throughout the entire manuscript.
- Lines 424-425: “At many stations there is a significant overestimation of the true IPF quantile..”. I am not sure what the authors refer to exactly. Instead I see in figure 14 that the median is rather centred on 0.
- Line 425: what does “linear model transpositions” mean?
- Line 482: “even when equalizing the other factors catchment size and elevation”. What does it mean?
Citation: https://doi.org/10.5194/hess-2023-144-RC2
Anne Bartens and Uwe Haberlandt
Anne Bartens and Uwe Haberlandt
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 419 | 109 | 11 | 539 | 5 | 6 |
- HTML: 419
- PDF: 109
- XML: 11
- Total: 539
- BibTeX: 5
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1