Spatio-Temporal Variation of Trophic Status and Water Quality with Water Level Fluctuation in a Reservoir

Round 1

Reviewer 1 Report

In the reviewed Manuscript (thereafter, MS) the Authors analyzed the results of monitoring of Shihmen Water Reservoir (thereafter, SWR) in relation to fluctuations of its water level. The main objective of MS formulated as follows: “…the main purpose of this study is to understand the spatio-temporal difference in water quality with WLF of Shihmen Reservoir from 2011 to 2021.”.

Methodologically, the water quality (WQ) of SWR was quantified by the Trophic State Index (TSI) calculated from equations suggested by Carlson (1977) using three WQ parameters: Total Phosphorus (TP), Secchi Depth (SD) and concentration of Chlorophyll (Chl). Average values of the three TSIs provided integrated estimate of WQ of SWF, and thus, of water quality.

Principal comments

The CTSI is a tool of classification (i.e., it classifies water ecosystem according to its trophic state: “oligotrophic” or “eutrophic”). The trophic state estimates say nothing about the WATER QUALITY of the ecosystem under consideration. The problem of quantification of WQ should solve the task of quantification, i.e., it should say is WQ “good” or “bad”. The founders of trophic classification system wrote (Carlson & Simpson 1996): “An unfortunate misconception concerning trophic state is that the term is synonymous with the concept of water quality. Although the concepts are related, they should not be used interchangeably.” The procedure of the WQ quantification should include the following stages (Parparov, A., K. D., Hambright, L. Hakanson, A. P. Ostapenia. 2006. Water quality quantification: basics and implementation. Hydrobiologia, 560:227-237; Parparov, A., G. Gal, D. Hamilton, P. Kasprzak, and A. Ostapenia. 2010. Water quality assessment, trophic classification and water resources management. J. Water Res. Prot. 2: 907-915.).

The Authors indirectly assume that the eutrophication represents the main threat for SWR. The Authors do not reply for the question: Water Quality for what? The Author state that the main direction of the SWR is “Water supply”, but of what kind: Drinking water supply? Domestic water supply? Irrigation? Or all these together? The WQ systems for all of these directions should be different.

The Authors did not establish functional correspondence between the values of CTSI and Water Quality indices (“rating values”. For instance, the CTSI values of 47.1 and 52.6 (Table 1): to what WQ rating values (“Good” or “Bad”, or Intermediate) are these CTSI values correspond? Is current state of SWR “good” or “bad”?

The reviewed MS does not provide response to these obvious questions. Therefore, I will recommend “Major revision” of this MS. One option: The Authors might change the title of the MS writing “Trophic state” instead of “Water quality”.

Minor comments

Abstract

Mean water depth seems to be one of the important key indexes (Mean depth or Water level?) for reservoir management in water quality since it reflects in water quality and can be easy-to-obtained. This study suggests that reservoir management administration can use the water level as a reference threshold for formulating water quality strategies. In proper hydrological conditions, administration should try to hold a higher water level in reservoir to maintain the quality for water supply What kind of supply?.

Lines 37-50 indicate lack of knowledge of literature. See for instance Hakanson and Peters “Predictive Limnology” and “Lake Kinneret: ecology and management” Zohary et al. (2014).

2.1. Study area. As fact, there is no description of the lake ecosystem

Lines 103-111 Indirectly assumption that the WQ is identical to the waterbody trophic status (not eutrophic as in line 103).

At this stage, it is unclear if the Authors use Water Level and/or Water mean depth. Water level does not give information about the reservoir morphometry, though, it is important information.

Fig.2. Electroconductivity and Ammonia are not parameters of water quality. Difficulty to read axis titles. It is difficult to estimate main temporal trends from this figure.

Figure 3. It is a description of the temporal dynamics of the monitoring data but not the WQ. It is very difficult to see the labels and to understand their meaning.

Lines 198-206. I think it would be more informative to analyze a set of the scatter plots of CTSI vs TP, Tu, Chl and SD. Generally saying, individual TSIs should be well intercorrelated, because the so called WQ parameters should be well intercorrelated.

Fig. 5. For me, these charts look very strange: formally, the represent a dependence of Y(water depth) from X(site number). What is meaning of the numbers inside of the columns.

Lines 238-239: algal abundance is not an index of water quality, but a monitoring parameter.

Lines 248-250  “Since water level is an important parameter for routine operation and management in reservoir, the accurate water level data can be easy-to-obtained at any time, while the sediments at the bottom of the reservoir vary slowly.” What does it mean?

Table 1. Modification of this Table had to be in the “study area” part. For some parameters (e.g., SS) standard deviation values exceed the mean values. Respectively, t-test is not applicable for statistics of these indices.

Conclusions: Lines 339-344 and  378-380 are not the results of this study.

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript has a major merit: it brings together a number of existing references on water level fluctuations in dams and creates a methodology that can be easily reproduced anywhere in the world, providing reservoir managers with a tool to define the water level threshold that allows water quality to be assured.  I therefore recommend that the manuscript be accepted after a minor revision. My suggestions and comments can be found in the manuscript file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

water-2551690 “Spatio-temporal variation of the Shihmen Reservoir water quality with water level fluctuation during 2011 to 2021” Wenwen Liao, Hsinan Chen, Meijeng Peng and Tawei Chang

This article presents a dataset collected over 11-years of water quality parameters in the Shihmen Reservoir in Taiwan. The title could be shortened by removing “with water level fluctuation” as this confuses the reader as to the importance of CTSI and water level. This article uses too many levels of data aggregation and is does not draw satisfactory conclusions.

The location shown in Figure 1 is difficult to view. Using a photographic view-from-space of the Shihmen Reservoir it is very much shades of green and darker green, making it hard to see boundaries and get a good impression of the surrounds. Perhaps a more map-like figure would be better, showing some of the upstream area of Sanmin Creek, Nanzigou Creek and Dahan Creek. The textual descriptions of the sampling sites should be moved to §2.1 Site Description.

Figure 2 is difficult to look at as the left y-axis has 4-and-a-bit cycles while the right y-axis always has 5 full cycles, thus the horizontal divisions do not match up. It would be better if the left y-axis for water level went from 200 to 250m so things matched up cleanly.

The statement on page 4 line 135 is too strong for the details in Figure 2 that it cites. There is no regular annual cycle of any of the water quality variables, except EC. Given the data listed in Figure 3 it is easy to generate correlations between monthly averages, which yield EC (R²=0.93) as highly correlated to water level while CTSI (R²=0.14), TP (R²=0.06) and NH₄-N (R²=0.02) are not significant at the 5% level. With respect to correlation to CTSI, of the three parameters that are used to calculate it, both Chl-a (R²=0.72) and TP (R²=0.66) are highly positively correlated while SD (R²=0.12) is not. Other water quality parameters that have significant correlation to monthly CTSI are NH₄-N (R²=0.49), Turbidity (R²=0.47), SS (R²=0.41) and NO₃-N (R²=0.37). Algal abundance on a quarterly basis is both highly and positively correlated to CTSI, as expected.

The statistics used in the paper looking at difference from the mean only highlight the one period from April to May with a significant depression in reservoir water level, and seemingly ignores all the other temporal behaviour. While it is clear the least-squares linear correlations between monthly values do not explain the whole story either, the authors use the 11-year average at each site and Figure 5 to imply that water level is a main controlling factor for CTSI. I think they are mixing too many levels of aggregation to have a consistent conclusion. For reference, monthly water level (R²=0.31) is just significant at the 5% level for CTSI. The correlations shown in Figure 6 are lower, but with many more points. It is also of note that only the two most upstream locations Chang Xing and Amu Plain show the highest algae levels (>20,000) in the highest CTSI zone (>50) while all downstream locations have high measured algae levels in the two other CTSI zones.

It is at this point that a least-squares linear regression of seasonal CTSI and algal abundance (4 points) having R²=0.97 clearly contradicts the Abstract which states “… there is no statistically significant difference in mean algal abundance at this time compared with other seasons”.

Upon reaching Figures 6 and 7 it appears that the previous annual values analysis was of little use, as these new figures are much more demonstrative. The argument being made is one of discrete values and thresholds, rather than continuous variables, but that has to be done with all the data points. It is also illustrative in Table 1 of the variables used to estimate CTSI, only Transparency is significant at the 5% level while TP and Chl-a are not. The Conclusion admits this by listing simple physical water parameters, such as EC, SS and pH, as significant with respect to storage water level rather than nutrients and chlorophyll. Did the authors consider multi-variable clustering, for example, to determine if the datasets from each sampling location were distinctly (spatially) different, or the sets of conditions (spatial/temporal) that led to CTSI>50, and CTSI<40?

English expression is fine.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

The manuscript is well designed and written. Improving in some points can make it better. 

Introduction is clear and give sufficient backround for the importance of WLF

Line 33: do you mean in the resorvoir region or in the whole world* please be clear.

Materials and Methods

Please seperate Figure 1 as Table and Figure. 

Please use trophic status in state of eutrophic status in all text. Eutrophic shows the trophic status of reservoir. 

Line 121: Please add the magnificance information for phytoplankton counting.

Results are clearly presented and discussion is supported with results and relevant literature.

Conclusion

It may be good to add a paragraph How your method can be used globally.  You focused on your reservoir. Thus it seems like a local study not a global study. Please make suggestions for using water level in reservoir management and also how your work can be improved with future work?

Please see the attached pdf for  other comments.

The manuscript can be accepted after some minor revisions. 

 

 

Comments for author File: Comments.pdf

Please see the attached pdf for grammer and other comments for English. Moderate editing of English language required.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Unfortunately, in the revised MS, The Authors indirectly assume an identity between the Reservoir trophic state and its water quality. Despite their declarations, the Authors do not analyze changes of WATER QUALITY, but TROPHIC STATE only. Water Quality is correct scientific term allowing its correct quantification. As a minimum requirement, I would like to see in the MS something similar to such a Table:

Table  Correspondence between trophic status of the Reservoir and Its water quality

CTSI range	Water Quality
CTSI1 – CTSI2	Excellent
CTSI3 – CTSI4	Good
CTSI5 – CTSI6	Intermediate
CTSI7 – CTSI8	Bad
CTSI9 – CTSI10	Very bad

 Otherwise, the term "Water Quality" in the revised MS is contentless, suitable rather for slogans than for the scientific publication.

Authors demonstrated very insufficient knowledge in scientific literature concerning water quality.

Alternatively, I would suggest to the Authors remove term "water quality" from entire MS and to deal with the trophic state only. 

 

Author Response

Thank you very much for your incisive and in-depth suggestions. Indeed, in this study, the authors did not define and quantify the quality of water, and regarded higher nutritional status as poor water quality too subjectively. At your reminder, the manuscript has been added to the relevant literatures and revised again. Please refer to the attachment for the revised manuscript.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors have made significant changes (and improvements) to the text and figures, which has resulted in an acceptable revision.

Author Response

Thank you very much for your comments and suggestions, which make this manuscript more complete.