Extreme Droughts Change in the Mekong River Basin: A Multidisciplinary Analysis Based on Satellite Data

Round 1

Reviewer 1 Report

The subject of the paper is very intersting and current but I think it needs a better and clearer explanation of the results in fact presents some elements of confusion.

For example a more detailed description of the climate of the Mekong River Basin to understand better the impact of a trend in extreme droughts could be useful. Figure 4 is not so clear, what "minimum precipitation" means? An histogram is usually used for the representation of the precipitation, so you can better understand the pattern both of time series that the seasonality.

To describe the difference between upstream and downstream you show only two zones (1 and 28) and two years (2001 and 2002). In 2002 wet season it seems that in 28 zone have fallen twice the millimeters fallen in the zone 1. The time series clustering related to preciptation time serie could be used for the characterization of MKB climate also emphasizing the differences in terms of millimeters inside wet and dry season between the most rainy and the least rainy cluster zone of MKB.

Please specify in the text the acronym, in line 80 you put for the first time MKB without explanation, and in line 81 you put MRB, what does that mean?

In Figure 1 describe the use of 3 SPI, 1, 3 and 6 months but in the results is not clear which SPI you used

In figure 7 you have to describe what means the round points in the Extreme drought interval both in the text and in the figure. The same in the figure 8 regarding the two graphs where a description of all the signs present could be better.

There are two times table 1 and table 2, in the text is clear the enumerations of the table but in the table there is a mistake.

The description of drought trend in MKB is shown with figure 9 and table 3 (pag. 17) but don't match. For example, zone 2 are orange (increasing) in table is no-trend, zone 5 are blu (no trend) in table increase.

The same happens between figure 11 and table 4 (pag.19), in table zone 1 belongs to cluster 2 but in the figure the colour seems cluster 4.

In the discussion in line 309 states that zone 4 belongs to Laos but in figure 2 has colour belonging to China (only a very little part is yellow), the same for zone 7 that in discussion put belongs to Thailand but in the figure is yellow (Laos).

In introduction there is a description of MKB that are repeated for same part in "2.1 study area......

In line 134 check the reference 22

 

Author Response

“Extreme Droughts Change in The Mekong River Basin: A Multidisciplinary Analysis Based on Satellite Data” by Vo et al, 2021

 

Response to Reviewer #1:

 

General comment: The subject of the paper is very interesting and current, but I think it needs a better and clearer explanation of the results in fact presents some elements of confusion.

 

Response: We would like to thank the reviewer for your constructive comments. We have responded all the comments one by one together with the revised manuscript. We highly appreciate the reviewer’s comments which has considerably improved our manuscript.

 

Specific comments:

Comment 1: For example, a more detailed description of the climate of the Mekong River Basin to understand better the impact of a trend in extreme droughts could be useful. Figure 4 is not so clear, what "minimum precipitation" means? A histogram is usually used for the representation of the precipitation, so you can better understand the pattern both of time series that the seasonality.

 

Response 1: More description of the climate of Mekong River Basin has been added in lines 55-59 of the revised manuscript.

“The climate of the Mekong River Basin (MRB) varies by topography, from a temperate seasonal climate in the upstream to a tropical monsoon climate in the downstream area. The rainy season usually lasts from June to November, while the dry season usually starts from December to May. The average annual rainfall of MKB ranges from 600 to 1700 mm [12]”.

 

In Figure 4, the “minimum precipitation” means the lowest value of monthly precipitation volume that computed in average for whole MKB, or the high of cumulative water column for whole basin in one month. To avoid reader’s confusion, we modified the caption as “The lowest average monthly precipitation” in Figure 4.

 

 

Comment 2. To describe the difference between upstream and downstream you show only two zones (1 and 28) and two years (2001 and 2002). In 2002 wet season it seems that in 28 zone have fallen twice the millimeters fallen in the zone 1. The time series clustering related to precipitation time series could be used for the characterization of MKB climate also emphasizing the differences in terms of millimeters inside wet and dry season between the rainiest and the least rainy cluster zone of MKB.

 

Response 2: Many thanks to the reviewer for the interesting comment. Actually, we have interpreted the difference between the upstream and downstream area based on the findings of zonal characteristics (Zone 1 and 28 and the years of 2001 and 2002). Based on the comment, we added more text to the section of time series clustering to highlight more the seasonal differences of the zones in the MKB. Please find our text modified in Section 3.4. Clustering time series (Page 18, Line 312-319):

“The results show that the meteorological limit is relatively uniform in the upper Mekong region, with slight fluctuation. All basins stretching from latitude 32 degrees north to 22 degrees north (zone 2, zone 3, zone 4) share the same characteristics. In contrast, in the lower part of the Mekong, the meteorological drought is highly variable. Especially in zones 10 to 16, the characteristics of meteorological drought appear in this area. In addition, neighboring regions may have different meteorological drought characteristics (zone 1 and zone 2). In that climate, the locations are very far apart but still have the same meteorological drought characteristics (zone 1 and zone 24).”

 

Comment 3. Please specify in the text the acronym, in line 80 you put for the first time MKB without explanation, and in line 81 you put MRB, what does that mean?

 

Response 3: The acronym was described in line 84 of revised version. In line 56, we have first time introduced the acronym of MKB (Mekong River Basin), and have revised to use the only term of MKB for the whole manuscript.

 

Comment 4. In Figure 1 describe the use of 3 SPI, 1, 3 and 6 months but in the results is not clear which SPI you used?

 

Response 4: SPI 3 was chosen for analyses. The modification was added in the methodology lines 158-159.  

 

Comment 5. In figure 7 you have to describe what means the round points in the Extreme drought interval both in the text and in the figure. The same in the figure 8 regarding the two graphs where a description of all the signs present could be better.

 

Response 5: Fig 7 was revised as the reviewer’s suggestion.

 

Comment 6. There are two times table 1 and table 2, in the text is clear the enumerations of the table but in the table, there is a mistake. The description of the drought trend in MKB is shown in figure 9 and table 3 (page. 17) but doesn't match. For example, zone 2 is orange (increasing) in the table is no-trend, zone 5 is blue (no trend) in table increase.

 

Response 6:  It is confused readers because of the catalogue of drought following the descending arrangement. The lower the value SPI, the higher the severity drought trend was. The increased drought trend hereby is the SPI time series has a negative slope. The decrease drought trend is the SPI time series has a positive slope. To avoid a mismatch, we revise both Table 1 and Figure 9.

 

Comment 7. The same happens between figure 11 and table 4 (pag.19), in table zone 1 belongs to cluster 2 but in the figure, the color seems cluster 4.

 

Response 7:  The mismatch between figure 11 and table 4 was revised as the reviewer’s comments.

 

Comment 8. In introduction, there is a description of MKB that is repeated for the same part in "2.1 study area......

 

Response 8:  We agree with the reviewer, we have removed the repeated information in Section 2.1 Study area in the revised manuscript.

 

Comment 9. In line 134 check reference 22

 

Response 9: The correct reference 22 has been revised.

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

This work focuses on the evolution of precipitation over the Mekong basin divided into 28 zones. Rainfall data comes from satellite measurements. They show a clear downward trend in the upper part of the basin and an upward trend in the lower part.

The paper presents a strong point which is the analysis of precipitation trends using the well-established Mann-Kendall method, which requires an important work of data collection and organization given the size of the basin. Conclusions (Lines 317-323) are well supported by the methodological approach. However, many weaknesses mean that it must be seriously reworked.

Major recommendations:

• Very little information is drawn from the K-mean Clustering of time series. What are similar characters of zones within the six groups? Lines 325-327: What information from a hydrological point of view?
• There are many gaps in the methodological approach. In Figure 6: minimum, maximum, mean are calculated on which interval? In Figure 7 (interval duration of MKB) and in Figure 8, what is the meaning of the circles? The most extreme values? Line 301: the lowest precipitation occurred in November 2014. This information is not very important because hardly significant and does not deserve to dwell on it. Lines 308, 309 and Lines 312, 313: What observations are these conclusions based on?
• It would be interesting to characterize the Mekong basin in the world: it is characterized by a very high seasonality of precipitation and a very low interannual variability (http://climatorealist.neowordpress.fr/resonant-rainfall-oscillation/).

Minor recommendations:

English: trend of drought in the MKB (caption figure 9)

Line 112 : do not repeat

Line 206: xj are value of time series at time j among n observational values

Line 267: the lower part of the MKB? Line 268: the upper part? Should be reformulated

Replace Table 3 by a figure, colors depending on the slope in the regions.

Table 4: unnecessary

Lines 302, 303: it is simply the natural distribution of rainfall height. We do not quite understand the interest of such a deduction.

Author Response

Response to Reviewer #2:

 

General comment: This work focuses on the evolution of precipitation over the Mekong basin divided into 28 zones. Rainfall data comes from satellite measurements. They show a clear downward trend in the upper part of the basin and an upward trend in the lower part.

The paper presents a strong point which is the analysis of precipitation trends using the well-established Mann-Kendall method, which requires an important work of data collection and organization given the size of the basin. Conclusions (Lines 317-323) are well supported by the methodological approach. However, many weaknesses mean that it must be seriously reworked.

 

Response: Dear Reviewer, We appreciate the reviewer’s comments. During the revision, we thoroughly addressed each of our comments to the very best of our capacity. Our detailed point-to-point responses and changes that were undertaken in the revised version of the manuscript are also provided below, after each response. We firmly believe that the manuscript has greatly benefited from the comments and would like to express our appreciation for our work's time and effort.

 

Specific comments:

Comment 1.  Very little information is drawn from the K-mean Clustering of time series. What are similar characters of zones within the six groups? Lines 325-327: What information from a hydrological point of view?

 

Response 1: The K-mean clustering of time series has been further discussed in the revised manuscript, in line 312-319. These help to enhance the information from hydrological point of view.

We have modified text in the Section  3.4. Clustering time series (Page 18, Lines 312-319):

“The results show that the meteorological limit is relatively uniform in the upper Mekong region, with slight fluctuation. All basins stretching from latitude 32 degrees north to 22 degrees north (zone 2, zone 3, zone 4) share the same characteristics. In contrast, in the lower part of the Mekong, the meteorological drought is highly variable. Especially in zones 10 to 16, the characteristics of meteorological drought appear in this area. In addition, neighboring regions may have different meteorological drought characteristics (zone 1 and zone 2). In that climate, the locations are very far apart but still have the same meteorological drought characteristics (zone 1 and zone 24).”

Lines 351-355:

“However, our clustering results based on K-mean Clustering method represent that the drought characteristics vary across the MKB indicated by zone distributions; of those, only cluster 3 has all zones located in the lowest section at the same time, we realized that meteorological limit is relatively uniform in the upper Mekong region, with slight fluctuation.”

Comment 2: There are many gaps in the methodological approach. In Figure 6: minimum, maximum, mean are calculated on which interval? In Figure 7 (interval duration of MKB) and in Figure 8, what is the meaning of the circles? The most extreme values?

 

Response 2: The gaps were discussed to come up with our research objective. In Figure 6, minimum, maximum, mean is calculated on the interval of distance or grid-cell size based on zonal statistic models. It means comparison of all raster values inside one zone. To clarify this, we revised text to for clarification in lines 250-252: “Figure 6 shows the results of zonal statistical analysis in 2001-2002 with an example for Zone 1. Various raster pixel values of SPI in the same zone was compared by zonal statistical model.”

 

As the recommendation of the reviewer 1, we have combined Fig.7 and Fig.8 for comparison, therefore, text, with clarification of extreme values or the circle as outliers, has been revised as follows (Lines 284-289):

“The circles in Figure 7's graphs are the outliers that mainly occur when the drought duration in the region is more complex and unpredictable than the normal drought level. However, zone 24 has the highest average drought duration of 7 months, while severe extreme droughts occurred under two months. The interval drought events, or the time between two drought events indicate that the average interval of extreme drought is around ten months in the MKB. “

 

 

Comment 3: Line 301: the lowest precipitation occurred in November 2014. This information is not very important because hardly significant and does not deserve to dwell on it.

Response 3: We have excluded “2014” from the text.

 

Comment 4: Lines 308, 309 and Lines 312, 313: What observations are these conclusions based on?

Response 4: These observations are drawn based on our results in Section 3.1/Lines 229-235:

“ The average monthly precipitation fluctuates in a range of lower than 10 mm to over 350 mm in the period. Of those, the precipitation event in November 2014 has the lowest precipitation over the period, with the average monthly precipitation at 2.97 mm (Fig. 4). The results also show that rainfall increases from January to August and de-crease gradually from August to December. The most rainfall is in summer (JJA) and autumn (SON). Meanwhile, winter (DJF) and spring (MAM) rainfall are lower.”

Regarding the Line 312-313:  Sorry for our mistake putting wrong information due to our old analysis results. We have revised the paragraph (Lines 334-345) as follows:

“The SPI calculations based on the Run theory have revealed 44 extreme drought events as the most frequent occurrence in zone 2 during the 19 analysis years. It means this zone (belongs to China) should consider the droughts in the coming years. This zone is located close to zone 1, 3, and 4 (also belongs to China), which were detected with increasing trends under the Man-Kendall test. Besides, the lowest extreme drought event (8 extreme events) was found in zone 25 (belongs to Cambodia) over the 19 years (Fig. 7). Next to this zone, however, zone 24 has the highest average drought duration of 7 months compared to the remaining zones. The decreasing drought trends were found in 5, 7, 9, 10-12, 14-16, 18-20, 23, 26, 27 and 28 in the lower part of the MKB (Myanmar, Thailand, Laos, Cambodia, Vietnam), which may indicate the increasing precipitations from 2001 to 2020 while the upper part (zones 1, 3, 4, 6, and 8 located mainly in China) has increasing drought trends.”

Comment 5: It would be interesting to characterize the Mekong basin in the world: it is characterized by a very high seasonality of precipitation and a very low interannual variability (http://climatorealist.neowordpress.fr/resonant-rainfall-oscillation/)

 

Response 5:  The comment is highly appreciated. We have added the information to the Introduction of the revised manuscript (Lines 55-59):

“The climate of the Mekong River Basin (MRB) varies by topography, from a temperate seasonal climate in the upstream to a tropical monsoon climate in the downstream area. The rainy season usually lasts from June to November, while the dry season usually starts from December to May. The average annual rainfall of MKB ranges from 600 to 1700 mm [12]”

 

Comment 6: To English: trend of drought in the MKB (caption figure 9).

 

Response 6:  We have rewritten the phrase.

 

Comment 7: Line 112: do not repeat?

 

Response 7:  Thanks the reviewer. The repeated has been removed.

 

Comment 8: Line 206: xj are value of time series at time j among n observational values.

 

Response 8:  It was rewritten (Line 212-213).

 

Comment 9: Line 267: the lower part of the MKB? Line 268: the upper part? Should be reformulated

 

Response 9:  Text was reformulated in lines 84-87:

“The Mann-Kendall test shows that the upper part of the MKB (zones 1, 3, 4, 6, and 8) has increasing drought trends (Fig. 8), while the lower part (zones 5, 7, 9, 10-12, 14-16, 18-20, 23, 26, 27 and 28) has decreasing drought trends. The remaining zones have no significant drought trends, including the Mekong Delta in Cambodia and Vietnam.”

 

Comment 10: Replace Table 3 by a figure, colors depending on the slope in the regions.

 

Response 10:  We have put Figure 10 to show colors depending on the slope in the regions; however, we still want to keep Table 3 due to its equations of zones’ slopes.

 

Comment 11: Table 4: unnecessary

 

Response 11:  Table 4 has been removed as suggestion.

 

Comment 12: Lines 302, 303: it is simply the natural distribution of rainfall height. We do not quite understand the interest of such a deduction.

 

Response 12:  Many thanks the reviewer for the comment. We agree with the reviewer that it is simply our description of the natural distribution of rainfall pattern in the study area, but we put the text to confirm our correct data analysis of satellite data sources used which could be incorrect to affect the tests’ results of the study.

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

From my point of view the manuscript has been
sufficiently improved to warrant publication in Water

Author Response

Dear Reviewer,

I am sincerely thank you very much for your time and your support!

Your valuable comments did improve our paper a lot. 

Wish you all the best!

Best regards,

Hoang.

Reviewer 2 Report

See the attached document

Comments for author File: Comments.pdf

Author Response

“Extreme Droughts Change in The Mekong River Basin: A Multidisciplinary Analysis Based on Satellite Data” by Vo et al, 2021

 

Reviewer #:

General comment: Most of the recommendations have been followed. Thus, the quality of the paper has been improved. However, one major recommendation was not fully understood, that relating to the interpretation of the clusters of the episodes of drought.

I insist on the need to take advantage of the clustering of the drought episodes throughout the MKB. The figure 11 is extremely difficult to interpret due to 1) the colors in the caption are different from those in the Figure 2) there is no logical gradation of the color pallet throughout the classes.

 

Response: We would like to thank the reviewer for your valuable comments. We have addressed the comments tracked in the revised manuscript. We highly appreciate the reviewer’s comments which have helped improving our manuscript’s quality. Based on the comments, The colors in legends of Figure 2 were revised and chosen as the same colors palette for classes as Figure 11.

 

Specific comments:

Comment 1: In Figure 11 the colors must be chosen judiciously to highlight a continuity in the number of clusters of the episodes of drought. For example, choose the color palette varying from blue to red within 6 classes. In notation RGB= (0,0,255), (102,0,255), (204,0,255), (255,0,204), (255,0,102), (255,0,0) or in notation html=#0000ff, #6600ff, #cc00ff, #ff00cc, #ff0066, #ff0000.

 Discontinuities will clearly appear between contiguous regions. The question that needs to be asked is: Do the discrepancies have statistical significance?

 

Response 1: Thank you for the clear, detailed comment of choosing judicious color palette. In Figure 11, the colors have been revised with a continuity color palette from bule to red within 6 classes (#0000ff, #6600ff, #cc00ff, #ff00cc, #ff0066, #ff0000) as the reviewer’s comment.

 

K-mean clustering depends on the k number of predetermined centroid points. Therefore, it is a difference statistical significance corresponding to k of clustering. The other algorithm such as G-mean can determine the k with statistical significance level a(Hamerly and Elkan, 2004). The G-means algorithm is based on a statistical test for the hypothesis that a subset of data follows a Gaussian distribution. G-means runs k-means with increasing k in a hierarchical fashion until the test accepts the hypothesis that the data assigned to each k-means center are Gaussian. G-means only requires one intuitive parameter, statistical significance level a. In this study, we used other methods to determine the number k of clustering. This method computed Sum of squares of distances of every data point from its corresponding cluster centroid which is called WCSS (Within-Cluster Sums of Squares). We performed k means clustering on different value from 1 to 28. For each k, the WCSS was computed. Plot the values for WCSS and number of cluster k. The k corresponding to the location of a bend (knee) in plot was chosen as optimal k number.

 

Comment 2. Interpret the figure by regionalizing the number of clusters of drought episodes. The number of clusters seems to decrease downward in the MKB (except region 1 where the Mekong begins, and regions 16, 20, 28 at the edges of the lower basin). Are there socio-economic impacts?

 

Response 2: The socio-economic activities can impact to the environments and climates. In our study, the clustering was analysed based only the meteorology factors, but we believe it also effected indirectly by socio-economic aspects. We hope the future studies could explicitly clarify the combined factors.

 

Comment 3. From a phenomenological point of view, the rainfall pattern of the Mekong basin is not subject to the rainfall oscillation, which is an essential characteristic of the MKB (http://climatorealist.neowordpress.fr/resonant-rainfall-oscillation/). You need to update the references appropriately. Consequently, the episodes of drought do not reveal any proven periodicity. This may explain the apparent low spatial coherence of the number of clusters, which may reflect local climatic conditions.

 

Response 3: The useful explanation about rainfall pattern of Mekong basin and rainfall oscillation has been added to discussion (Page 20, lines 356-359):

“Precipitation of Mekong basin is subject to the precipitation oscillation explained by Räsänen and Kummu, (2013) [39]. However, the episodes of drought do not reveal any proven periodicity. This may explain the apparent low spatial coherence of the number of clusters, which may reflect local climatic conditions.”

 

Minor recommendation:

Caption Figure 10: Slope of drought trend in the MKB

Response: Many thanks the reviewer. The caption of Figure 10 has been revised.

 

Reference:

Hamerly, G., Elkan, C., 2004. Learning the k in k-means. Adv. Neural Inf. Process. Syst. 16, 281–288.

Räsänen, T., Kummu, M., 2013. Spatiotemporal influences of ENSO on precipitation and flood pulse in the Mekong River Basin. J. Hydrol. 476, 154–168. https://doi.org/10.1016/j.jhydrol.2012.10.028

 

 Best regards,

Hoang

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

My recommendations have been incorporated into the manuscript. I recommend publishing the paper as is.