Changing Pattern of Water Level Trends in Eurasian Endorheic Lakes as a Response to the Recent Climate Variability

Round 1

Reviewer 1 Report

Review Remote Sensing

This is a good paper. I like the goal as stated in the discussion that “The essence of the above analysis consisted of evaluating the potential of terminal lakes as single-point indicators for multiple climate change stresses in large endorheic basins. In this way, the water level variations in endorheic lakes may provide valuable insight into both the regional hydrological regimes and global circulation changes”

I agree that terminal basins are “integrators” of the net effects of climate and can “amplify” the insight into regional hydrology changes.

Overall, the paper is well-written and the science is robust.

I have however, 2 recommendations for improvement.

1). Please go through the paper and try to clarify the results section. Most of the paper read really well, but there are just so many trends of lake and climate variables going up and down for different periods and so many descriptions, that if things could be simplified somewhat (and perhaps some material moved to a supplement to make less figures) that would be helpful.

2) Please provide literature review on previous studies on lake level from satellite and more details on the accuracy of the satellite lake level datasets.

The statement: Line 68 is too vague in my opinion: “high accuracy from centimeters to decimeters.”

Later, in section 2.2 the statement is provided: “The lake level records from this dataset were compared with 132 gauge data and showed acceptable height accuracy” – please elaborate on this accuracy. If it is decimeters, I don’t think this would be adequate to have a robust signal. Centimeters, yes, and even 0.5 m or so, probably.

The trends calculated in m/year in table 2 are pretty small for most of the lakes, so again knowing the +/- error bars would help know the signal versus noise. It is good that Pearson Correlation Coefficients were calculated. Perhaps in some figures like Figure 3 you could provide the upper and lower bounds of the accuracy of the calculations for the different altimetry databases…

I think that some background literature in the introduction on lake level calculations from satellites and trends from the different satellites needs to be discussed. What have previous studies found? There is no background or literature review in this area, and it is needed to place this work in the larger context.

Other minor points

Line 52: “Several previous studies [14-18] investigated water level dynamics in terminal lakes with a projection on climate change” – awkward sentence. I would rewrite as something like: “Several previous studies [14-18] have investigated the impacts of climate change projections on water level dynamics in terminal lakes.”

On page 2, why does the line spacing suddenly become less?

Line 64 “also few due” is awkward please rewrite with some thing like “The gauge stations in the Tibetan Plateau are also limited due to the harsh environment”

The title reads a bit awkward, so I would consider coming up with something a bit shorter, but this is a nit-picky point so ok also to leave as-is

Author Response

We thank the reviewer for careful review, positive evaluation of our study and for the valuable comments and suggestions helping to improve our manuscript. Our point-by-point replies are provided in the attachment together with references to the revised manuscript.

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Review of manuscript: Changing pattern of water level trends in Eurasian endorheic lakes as a response to the recent climate by Zhang et al. for a possible publication in the Journal of Remote Sensing

This manuscript aims to investigate the trend of water level change in recent 30 years, the precipitation/temperature impact to the observations, and the regime shifts at the three regions through a comparative analysis. The topic is interesting.

The major weaknesses would be as follows:

1. The hydrologic budget of an endorheic lake is not used or proposed in this manuscript as a fundamental theory to support the comparative analysis. Precipitation, evaporation, discharge from groundwater and storage change are the budget components. The lake water level change is not associated with evaporation, groundwater discharge and storage change well in this manuscript. Groundwater did not change in lake watersheds?
2. Temperature is related to evaporation, but relation of evaporation with temperature is not established well in this manuscript. How do the “precipitation” and “temperature” function on the water levels? In the paragraph of line 357 to 367, the authors concluded that the “precipitation change” played a major role. However, the two factors could be considered combined effected to the level change.

If look at the Fig. 7(k) and (I), according to the table above, one can explain its water level trend (gray line in Fig. 7(k)) after year 2008 (before 2008, the gray line shows positive correlation with the precipitation anomaly), then the impact of temperature anomaly started to impact the water level: less evaporation than usual compensated its loss in precipitation. A lot of figures can be explained with both factors. This problem induces my attraction because when I read till the discussion and conclusion, I feel there are many dividers (various regime shifts based on precipitation/temperature in three zones). It is a list rather than some summarized patterns. However, as a reader, I am wondering if the latitude, elevation, salinity, absolute temperature in different regions contribute to the water level change trend.

3. Lack of quantitative criterion in section 2.6 for determining regime shift years in Figs 11 to 13. It would be better to provide a table that lists the main results or accuracy of some key steps (like Table 2). Then, based on this big database that covers monthly/annual water level, precipitation, temperature, and their anomaly, the analysis could be made much solid.
4. Comparison of links of lake level change to precipitation and air temperature could stand out the factor weight. For instance, compare the precipitation/temperature of same month (such as in Apr., usually it should be stable, but once there is an obvious difference, it is worthy a big weight, and it is also better to analyze the intra reason of anomaly: whether more water comes from precipitation or snow melting?), while if only computing the annual average value, precipitation/temperature impacts are easily sheltered by the hottest/coldest and the driest/wettest seasons, which makes the weight vanish.
5. The selection of those famous lakes is very interesting, some of the lakes have different names, please keep consistent for the whole manuscript if one name is decided.
6. It would be better to add the figure number at the corresponding text, since there are many figures and different lakes are not intuitive to recognize. It could help improve the readability by adding necessary fig. number.

Specific comments.

Line 43- Better change “those three regions” to “CA, TP, and MP”.

Line 62- the phrase of “stations established in the Soviet Union” is easy to induce misunderstanding, since Soviet Union collapsed in 1991. It is not clear if the 5 lakes all establish before 1991.

Line 93- remove the right parenthesis of  “Kunlun Mountains)”

Line 94- extra space in “   [21,22].”, but it may be justified by aligning text to both left and right margins, never mind.

Line 139-  “(Birkett et al. 2017)”: should be cited by number in [].

Line 148- vertical shift?

Line 149- “,” could be deleted.

Line 169- “(Schneider et al., 2015)”:  please keep consistent in citation by number in [].

Line 193- prefer to use present tense rather than the past tense since the location of the three representative lake zone are facts. Better to reorganize this sentence to avoid repetition of “was” or “is”.

Line 202- How to calculate error variance?

Line 229- what does the subscript “m” in eq. (2) represent? Is the “n” number a constant value? such as 12 (1 year = 12 months). Without further example, as a reader I am not clear about how does the “n” impact the CUM value, is it a temporal or spatial sensitive value? Since there is a Σ sign in eq. (2), if the “n” value is different at different lakes, then the CUM values are different at different locations. The 1^st n in eq. 2 would be m which would be less than or equal to n. please check. How is ?_? from eq. (1) used in eq. (2)?

Line 230- section “2.7”?

Line 243 to 245- this paragraph was written in a way of emphasizing on the structure of “Result” section, instead, I suggest to add more related to the topics, such as some experiment results in the 3 zones, which would be be much more intuitive to readers.

Table 2-The “Aral Sea Large” and “Aral Sea small” are not consistent with the previous north/south Aral Sea, and also different from the expression in the figs.

Fig.3, Fig.4, Table 2, and Line 263- what is the unit of trend? 0.058cm/yr (line 263) or 0.058 m/y(Table2)? and the fig 3(f) is lack of unit, btw, although the south Aral Sea is an exceptional case, the decrease trend in this lake is astonished, almost 10 times than others. More information could be provided to explain the observation.

Line 295 to 310- still the unit problem, cm/yr or m/y?

Line 302- Ngangzco or Ngangco (line 111)? please check through and keep consistent.

Line 306- same lake name problem for “Siling”, Siling or Silingco?

Line 337- (Figures 7-9)?

Line 338- “The air temperature experienced a decreasing trend from 1990 to 1997, changing to an apparent increase afterwards”: I have a different view regarding the orange cumulative anomaly line. It depends on the definition of CUMm (eq. (2)), according to the limited information in pages 6 and 7, I am confused about the “mean/average” value- which is the base value to calculate anomaly amplitude.

Take the (b), (d), and (f) figures in Fig. 8 as an example, all the three orange lines are negative, which means negative anomaly line (lower temperature than average records). Thus, the year of 1997 is not the turning year from decreasing to increasing, it is the amplitude in change rather than direction. For the whole period, I think the lakes of (b), (d), and (f) figures in Fig. 8 experienced negative temperature anomaly, which means the air temperature are generally lower than average, but from 1990 to 1997, the negative temperature anomaly (you may use other expression, such as “air temperature decrease”, but I don’t think it is proper to use “air temperature” as a subject in line 338) is severer year after year, while after 1997, the degree of temperature anomaly becomes to be slower.

My understanding may not right, but I strongly suggest to add more details about section “2.6 Cumulative anomaly analysis” and “2.7 change point detection”.

Line 346- in this sentence, “For example, in Lake Issyk-Kul, the cumulative precipitation”, a specific lake has been mentioned, I suggest to add the figure number of this lake, like “in Lake Issyk-Kul (Fig. 7)”

Line 349- I suggest to add figure numbers “(Figs. 8 and 9)” at the end “In the TP region, the cumulative curves of precipitation had different patterns in the northern and southern parts.”  And another view from mine, I feel they share similar patterns (precipitation anomaly decreases first and then increases) with different turning years. I am eager to get more analysis of this topic, however, it seems there is an opening and then stops immediately.  

Line 385- in this sentence, “Here, two methods were jointly applied to estimate the change points in the annual time series of precipitation and air temperature”, are the two methods mean the “Vertical and Horizontal lines” in the following figures? And what is the criterion of dividing the regime shift years?

Line 388- Fig. 11 uses the same color and line type as previous figs, suggest to change line type or color to differentiate from the precipitation and temperature cumulative anomaly.

Line 407 to 412- this paragraph is about the pattern in MP, however, in the caption of Fig. 14 about the MP pattern, there are different shift years, please check.

Line 415- “over three  the lakes in the northern part of the Tibetan Plateau”, same problem in line 422, 436.

Line 532 to 535- why the sentence “The lake level patterns in the TP and CA had different interannual variabilities: the increasing trends were significant until 2008 in CA, while the increasing trends were more noticeable since approximately 1997 in the southern TP and since approximately 2005 in the northern TP” has different expression about lake level in CA (line 449 “All lakes in the CA region experienced a dramatic increase in lake water levels since 1997”). it seems compared with South TP, it also increase from 1997 in CA.

Line 731- Change indentation for the link.

Comments for author File: Comments.pdf

Author Response

We thank the reviewer for careful review, positive evaluation of our study and for the numerous valuable suggestions. We incorporated most of the Reviewer’s suggestions to the revised manuscript. We provide the point-by-point replies together with references to the revised manuscript in the attachments.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors word very hard to dramatically improve the paper over a short period. In my opinion, it is ready to publish, except perhaps a few more "caveats" or limitations of this approach in comparison to the full endorheic basin budget discussed by the other reviewer. There are also  a few awkward sentences in terms of writing, but a review of the english should catch those. 

Author Response

Response to the Reviewer#1:

We are thankful for the Reviewer’s work on the manuscript and for valuable remarks and suggestions made during the two rounds of reviews. We have additionally revised the style and grammar in the revised version of the manuscript.

Reviewer 2 Report

The cover letter list answers to reviewer’s concern item by item, and the supplementary material provides the water level, precipitation, and temperature variation in details. Both documents are carefully written to serve the main manuscript. They answer some questions about the filter methods’ accuracy level, consider more influence factors such as snow melting process. The new manuscript has improved a lot in organization with better plots to provide information directly, and some charts display proofs about glacial area ratio, mean precipitation and temperature values in different regimes.

While the general content/mission of the manuscript is appropriate, the manuscript requires fairly significant revisions before it could be considered for publication.

1. What assumption could be established to support “the relationship of the water level signal to air temperature and precipitation as the most robust available climate variables” in this manuscript based on the water budget of the endorheic areas? 

2. The newly added Figure 4 and Table 4 are good plot and chart, which directly display the relationships of water level, precipitation, and temperature. Based on the plot, I have two questions:

As the caption of Fig. 4 indicates “Asterisks mark significant correlations”, it seems there are way more asterisks in relationship with temperature than with precipitation. If the monthly statistic data demonstrate this strong correlation, how could we conclude that the precipitation is the major contributor?

3. The authors might try to answer the above question in Page 16: “The precipitation was considered the main climatic driver of the distinct spatial patterns of lake water levels in the three neighboring regions, as it was found that the various spatial characteristics of precipitation, but similar patterns of air temperature were present.” Thus, comes to my second question: Is it necessary to include Lake 1,2, and 3, especially the area of Aral Sea is very large? Same amount water level change requires way more precipitation when comparing with small lakes (see attached excel spread sheet). If taking a look at Fig.1, you will find the Lake 1,2, and 3 (Aral Sea, and sarykamysh) are located especially west.

While precipitation and temperature are two special sensitive/non-sensitive characters, they are different if in a large range but keep consistent if in a relative small area. Thus, if excluding Lake 1,2, and 3, the water level change trend becomes clear- both CA and TP are positive, and MP is negative. Another good reason to exclude Aral Sea (both north and south) can be found in the discussion at page 15: Aral Sea south’s main tributary  “is strongly regulated and intensively used for irrigation”. If the anthropogenic influence is unclear, how could the “-37.8 cm/y” of Aral Sea South water level variation in Table-2 reflects the truth?.

4. After excluding Lake 1,2, and 3, the three regions can be analyzed within its own specialty: TP has high elevation, MP has high latitude, and CA has a moderate latitude with a mild elevation. They are important features that can impact temperature variation. For specific unusual lake, its elevation, area can be analyzed case by case (such as Issyk-Kul in CA, Hyargas in MP).

Author Response

We are thankful for the Reviewer’s work on the manuscript and for valuable remarks and suggestions. Below are the point-to-point replies to the Reviewer’s comments (please, see the attachment).

Author Response File: Author Response.docx