the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jan 2024
Evolution of river regime in the Mekong River basin over eight decades and role of dams in recent hydrologic extremes
Abstract. Flow regimes in major global river systems are undergoing rapid alterations due to unprecedented stress from climate change and human activities. The Mekong River Basin (MRB) was, until recently, among the last major global rivers relatively unaltered by humans, but this is changing alarmingly in the last decade due to booming dam construction. Numerous studies have examined the MRB’s flood pulse and its alterations in recent years; however, a mechanistic quantification at the basin scale attributing these changes to either climatic or human drivers is lacking. Here, we present the first results of the changes in natural hydrologic regimes in the MRB over the past eight decades and the impacts of dams in recent decades by examining 83 years (1940–2022) of river regime characteristics simulated by a river-floodplain hydrodynamic model that includes 126 major dams in the MRB. Results indicate that while the Mekong’s river flow has undergone substantial decadal trends and variabilities, the operation of dams in recent years is causing a fundamental shift in the seasonal volume and timing of river flow and extreme hydrological conditions. Even though the dam-induced impacts are small so far and most pronounced in areas directly downstream of major dams, dams are intensifying the natural variations in the Mekong’s mainstream wet season flow. Further, dams have exacerbated drought conditions by substantially delaying the MRB’s wet season onset, especially in years when the natural wet season durations are already shorter than in normal years (e.g., 2019 and 2020). Further, dams have shifted up to 20 % of the mainstream annual volume between dry and wet seasons in recent years; while this has minimal impact on the MRB’s annual flow volume, the flood occurrence in many major areas of the Tonle Sap Lake and Mekong Delta have been largely altered. This study provides critical insights on the long-term hydrologic variabilities and impacts of dams on the Mekong’s flow regimes, which can help improve water resources management in light of intensifying hydrologic extremes.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1936 KB)
-
Supplement
(2881 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1936 KB) - Metadata XML
-
Supplement
(2881 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-3158', Mauricio Arias, 28 Jan 2024
<strong>Publisher’s note: this comment is a copy of RC1 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/egusphere-2023-3158-CC1 -
AC2: 'Reply on CC1', Huy Dang, 28 Feb 2024
Thank you for your effort in reviewing our manuscript. Kindly find our response to your comments in the Reply to RC1.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC2
-
AC2: 'Reply on CC1', Huy Dang, 28 Feb 2024
-
RC1: 'Comment on egusphere-2023-3158', Mauricio Arias, 29 Jan 2024
hess-2023-3158 Dang
The manuscript presents an interesting and thorough examination of natural variation and dam impacts on the Mekong’s hydrological extremes over eight decades. The authors looked at a number of hydrological aspects such as flow trend, seasonal timings, flow volume, and extreme events and flooding patterns every decade. With the use of a hydrodynamic model, the author simulated flows with and without dams so as to quantify the dam impacts and look at Mekong’s hydrology under natural variability. Overall, the most significant contributions this study provided are (1) long-term trend of flow and (2) a full picture of natural flow and hydrological extremes, both drought and flood, compared to dam-altered flow and extremes in the whole Mekong Basin. In general, the manuscript in its current form is well written with insightful figures. That said, the manuscript can always use a bit of proof-reading and further synthesis.
While this study is certainly a step ahead of previous research in the Mekong looking at the effect of dams on river flows, the more comprehensive study results also means that the study ought to be controlled for two important factors which are directly linked to the conclusions of the study. First, while the study calibrated and validated the model in the conventional way, it is not clear to me that the authors are able to verify if the model captures the multidecadal variability and trends the study elaborates on. How can you ensure the reader that such term trends are real and not a model artifact? I suggest a stronger verification process, including how (or if) the study captures well-documented long-term variability and trends. See for instance the papers by Delgado (2010, 2012).
Another aspect I do not think the study controlled against well was water infrastructure in the floodplains and delta. As I hope the authors are aware, there has been widespread development of water control and irrigation infrastructure in both Cambodia and Vietnam, with effects on flooding well documented. Without taking a close look at what is happening with infrastructure within the floodplains, I would be extremely cautious about analysis and assertions made for those locations in Cambodia and Vietnam. The fact that the water level validation for these stations was not great may be a proof of this.
In addition to these general comments, I have several punctual comments and suggestions throughout the manuscript:
Abstract:
- Line 13: These are certainly not the first results of hydrological change in the Mekong. Consider deleting the word “first”.
- Lines 18-19: I suggest adding specific numbers associated with this claim that dams are intensifying variations in wet season flows.
Introduction:
- Line 29: I would be cautious with the use of the word "stable", as many can interpret this as a "flat hydrograph", which is the complete opposite from what the subsequent references allude to.
- Lines 49-50: Here is another newer reference related to dam impacts in biodiverse rivers:
Here is a good reference for this claim:
Winemiller, K.O., et al, 2016. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351, 128–129.
- Line 56: here you stay that the Mekong has “rather unpredictable seasonal variability”. Do you refer to interannual variability? I argue that the Mekong river flow intraannual/seasonal variability is highly predictable (wet season in May-Nov, driest months Feb-Apr). The fact that you have hydrological models with very high fit to observations supports the predictability of this system.
- Line 57: the statement here about seasonal timing sounds rather contradictory to the statement I talked about in the previous comment.
- Line 60: That the Mekong has a flood pulse dynamic has been documented for well over a decade, so consider updating this ref with one of the more historic ones:
Kummu, M., Sarkkula, J., 2008. Impact of the Mekong River flow alteration on the Tonle Sap flood pulse. Ambio 37, 185–192.
Västilä, K., Kummu, M., Sangmanee, C., Chinvanno, S., 2010. Modelling climate change impacts on the flood pulse in the Lower Mekong floodplains. Journal of Water and Climate Change 01, 67–86. https://doi.org/10.2166/wcc.2010.008
- line 60: a reference is needed for the sentence “During the remainder of the year, river flow gradually reduces to less than 10% (sometimes 5%) of its flood peak.”
- Line 71: I'm confused about the mentioning of the Chi-Mun basin here, as this is a basin upstream of what is normally considered the Mekong floodplains. Besides, this region has had water infrastructure for much longer than the rest of the MRB, so if there is anywhere in the basin where the flood pulse has been affected for a long time (since the 1970s), its in the Chi Mun.
- Line 82: Vastila et al (2010) looked primarily at inundation changes from climate change. A more relevant ref to the ecological changes would be Arias et al (2014b):
Arias, M.E., Cochrane, T.A., Kummu, M., Lauri, H., Koponen, J., Holtgrieve, G.W., Piman, T., 2014. Impacts of hydropower and climate change on drivers of ecological productivity of Southeast Asia’s most important wetland. Ecological Modelling 272, 252–263.
- Line 83: I suggest stating “hydropower development” instead of “booming hydropower dams”
Data and methodology:
- Data: Given the poor relationship we know exists between water level observations and derived discharged, I would be very cautious with analysis based from results for stations 10-13, for both levels and discharge. In addition, stations 12 and 13 are strongly influenced by water infrastructure in the Delta, which is not considered in this study.
- Lines 156: Though I am aware that this model has been documented multiple times before, it might be informative to mention the basic hydrodynamic assumptions the model use (e.g., kinematic wave?)
- lines 165-166: is it CFD or CMFD? I think there is a typo here. I would discourage you from using the acronym CFD, which is a very common acronym used in water science and engineering for Computational Fluid Dynamics.
- Line 177: Just to clarify: Your model switches on dams only after their corresponding year, is that correct? How did you consider reservoir infilling?
- Lines 198/Figs S2-3: Though I see that fig S2-5 present time series of ERA5 vs biascorrected, having 12 frames per figures, each with 20+ years of time series, makes it really difficult to truly see the difference between the simulations. I recommend a simpler graph illustrating the biascorrection process for a single station and shorter timeframe.
- I notice that statistical indicators for model performance were not mentioned in the Methodology section. They instead first appeared in the Results section. I suggest a mention of those indicators in Methodology.
Results:
- Model performance: What about testing for interannual variability and trends? Because the model predicts well seasonal variability, it does not mean that it also reproduces well how those patterns varied from year to year. Later on in the results you use the model results to make assertions on interannual variability and trends, but perhaps early on you have to demonstrate that your model (and input reanalyzed data) is in agreement with those trends that have been found from observations alone.
- line 221: Please correct the typo, Stung not Strung.
- Lines 221-223: I am a bit concerned that the water level simulations provide worse results than discharge. This probably has to do with the river bathymetry representation used in the model. What did you use by the way? Perhaps I missed that from the methods. I am sure that the authors realize that discharge measurements are seldom made, and that the vast majority of the discharge observations in the Mekong are derived from water level observations via regression equations. Thus, if there are inconsistencies in the water level measurements (which I do not think is the explanation here) there are certainly inconsistencies with the discharge data. The Mekong floodplain becomes really flat south of Kratie, so we should be very skeptical of any discharge data calculated (not observed) for the lower four stations.
- Lines 254-255: Here you state that the Lancang has experienced a decreasing trend, but do not really explain why. This is contrary to what has been reported for rivers starting in the Himalayas, where icemelt has increased flow in the upper parts of the Mekong and other rivers in the region:
Li, D. et al. High Mountain Asia hydropower systems threatened by climate-driven landscape instability. Nat. Geosci. 15, 520–530 (2022).
- lines 273-274: there’s a repetition of the word ‘increase’. Please fix.
- Line 288: Regarding the comment about natural retention in the Tonle Sap, Stations 12 and 13 are deep into the Delta in Vietnam, quite a ways from the Tonle Sap, but in the middle of one of the largest rice production regions in the world. Thus, you need to show some strong evidence to support that the Tonle Sap retention has effects there greater and the massive and active water infrastructure network Vietnam has built in the Delta. This might be related with the calibration results for these stations being below optimal.
- line 317: “dams are generally delaying the wet season onset…”. The author should expand that part by discussing the wet season duration because the example inside the brackets is about duration, not onset.
- line 318: based on Figure 4e, I think longer duration is not just in the early 2000s but in the 2000s as seen in 2001, 2005, 2007, and 2009.
- Figure 5: why not include a figure of ‘Natural anomaly of dry season volume’? This should be consistent with how Figure 4 was depicted.
- Lines 367-370: This sentence here seems like a long way of describing what I think is an obvious logic (a decrease in peak flow decreases flow in the floodplains). Consider deleting or synthesizing.
Discussion:
- Lines 403: with regards to changes in the Lancang and 3S basins, it would probably be good to compare and contrast your results with those studies that have looked at dam development specifically for those regions. For example:
Räsänen, T.A., Koponen, J., Lauri, H., Kummu, M., 2012. Downstream Hydrological Impacts of Hydropower Development in the Upper Mekong Basin. Water Resources Management 26, 3495–3513. https://doi.org/10.1007/s11269-012-0087-0
Piman, T., Cochrane, T.A., Arias, M.E., 2016. Effect of Proposed Large Dams on Water Flows and Hydropower Production in the Sekong, Sesan and Srepok Rivers of the Mekong Basin. River Research and Applications 32, 2095–2108. https://doi.org/10.1002/rra.3045
- With regards to the flood pulse, there is a newer paper that is probably worth reviewing:
Morovati, K., Tian, F., Kummu, M., Shi, L., Tudaji, M., Nakhaei, P., Alberto Olivares, M., 2023. Contributions from climate variation and human activities to flow regime change of Tonle Sap Lake from 2001 to 2020. Journal of Hydrology 616, 128800. https://doi.org/10.1016/j.jhydrol.2022.128800
Conclusions:
- Lines 452-3: As mentioned earlier, this conclusion can only be drawn if you verify that your model effectively can replicate long-term observations.
- Line 459: Remove “or higher” to avoid confusion.
- Line 467: the use of static land cover is also a limitation, despite arguably not having substantial impacts on flow alterations compared to dams.
Citation: https://doi.org/10.5194/egusphere-2023-3158-RC1 -
AC1: 'Reply on RC1', Huy Dang, 28 Feb 2024
The manuscript presents an interesting and thorough examination of natural variation and dam impacts on the Mekong’s hydrological extremes over eight decades. The authors looked at a number of hydrological metrics such as flow trend, seasonal timings, flow volume, and extreme events and flooding patterns every decade. With the use of a hydrodynamic model, the authors simulated flows with and without dams so as to quantify the dam impacts and look at Mekong’s hydrology under natural variability. Overall, the most significant contributions this study provided are (1) long-term trend of flow and (2) a full picture of natural flow and hydrological extremes, both drought and flood, compared to dam-altered flow and extremes in the whole Mekong Basin. In general, the manuscript in its current form is well written with insightful figures. That said, the manuscript can always use a bit of proof-reading and further synthesis.
- Response: We thank Dr. Mauricio Arias for taking the time to read the manuscript and providing constructive comments that incisively helped improve the quality and credibility of the manuscript. We have revised the manuscript and correlated our results with relevant findings from suggested sources and addressed other comments. A detailed, point-to-point response is provided in the following.
While this study is certainly a step ahead of previous research in the Mekong looking at the effect of dams on river flows, the more comprehensive study results also means that the study ought to be controlled for two important factors which are directly linked to the conclusions of the study. First, while the study calibrated and validated the model in the conventional way, it is not clear to me that the authors are able to verify if the model captures the multidecadal variability and trends the study elaborates on. How can you ensure the reader that such term trends are real and not a model artifact? I suggest a stronger verification process, including how (or if) the study captures well-documented long-term variability and trends. See for instance the papers by Delgado (2010, 2012, the first already cited in the manuscript) and Rasanen (https://www.hydrol-earth-syst-sci.net/17/2069/2013/; 10.1016/j.jhydrol.2012.10.028).
- Response: This is a good point, thanks for raising it. To further prove the model’s ability to capture the multidecadal variability and trends, we will provide additional validation of these characteristics against the observed data at each selected stations in the supplementary and include additional analysis specifically on this in the model performance section of the revised manuscript.
Another aspect I do not think the study controlled against was water infrastructure in the floodplains and delta. As I hope the authors are aware, there has been widespread development of water control and irrigation infrastructure in both Cambodia and Vietnam, with effects on flooding well documented. Without taking a close look at what is happening with infrastructure within the floodplains, I would be extremely cautious about analysis and assertions made for those locations in Cambodia and Vietnam. The fact that the water level validation for these stations was not great may be a proof of this.
- Response: Thanks for raising this important point. Indeed, infrastructure development in the delta region is an important issue. While the massive system of channels transferring water across the Delta has been considered in our model as linkage between grid cells, its representation is partly limited due to its spatial resolution. Additionally, the current model version doesn’t include the capability to account for dikes or other infrastructures. We are aware of those and have been working to account for these infrastructures in the model, which will be presented in our forthcoming articles. We believe that our findings would not be drastically different if we had accounted for these missing factors, especially considering that the analysis in Figure 6 shows the possible difference of flood occurrence in decadal term instead of actual magnitude while the remaining analyses are generally based on discharge. In the revised manuscript, we will add caution regarding any potential impact in our results caused by these factors.
In addition to these general comments, I have several punctual comments and suggestions throughout the manuscript:
Abstract:
Line 13: These are certainly not the first results of hydrological change in the Mekong. Consider deleting the word “first”.
- Response: We appreciate your comments, and we note that there have been previous studies providing critical insights into the hydrological change in the Mekong, some include long-term (~80 years) analysis such as Delgado et al. 2010. However, to the best of our knowledge, previous studies mainly analyze observations at selected stations or basin scale in short time periods, and none have yet to provide an analysis at basin scale of this temporal coverage. To address this issue, we will modify the language in the revised manuscript to avoid confusion and misinterpretations.
Lines 18-19: I suggest adding specific numbers associated with this claim that dams are intensifying variations in wet season flows.
- Response: Thanks for the suggestion. We will consider adding this information to the revised manuscript.
Introduction:
Line 29: I would be cautious with the use of the word "stable", as many can interpret this as a "flat hydrograph", which is the complete opposite from what the subsequent references allude to.
- Response: Thanks for the suggestions. We will change the wording to “A consistent pattern of river regime…” in the revised manuscript.
Lines 49-50: Here is another newer reference related to dam impacts in biodiverse rivers:
Winemiller, K.O., et al, 2016. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351, 128–129.
- Response: Thanks for the suggestion. We will add this reference to the revised manuscript.
Line 56: here you stay that the Mekong has “rather unpredictable seasonal variability”. Do you refer to interannual variability? I argue that the Mekong river flow intraannual/seasonal variability is highly predictable (wet season in May-Nov, driest months Feb-Apr). The fact that you have hydrological models with very high fit to observations supports the predictability of this system.
- Response: Thanks for the comment. While there is a general understanding that the Mekong river’s wet season occurs during May-Nov, the actual onset and ending of it varies greatly (1-3 weeks) from year-to-year which is shown in Figure 4a. Additionally, the good fit from our model may suggest that it could replicate historical conditions based on reanalyzed data (ERA5), however, there would still be considerable uncertainties in predicting future seasons/conditions that have yet to happened. We will clarify this point in the revised manuscript.
Line 57: the statement here about seasonal timing sounds rather contradictory to the statement I talked about in the previous comment.
- Response: Thanks for the comment. We have addressed this in our response to comment #5 above.
Line 60: That the Mekong has a flood pulse dynamic has been documented for well over a decade, so consider updating this ref with one of the more historic ones:
Kummu, M., Sarkkula, J., 2008. Impact of the Mekong River flow alteration on the Tonle Sap flood pulse. Ambio 37, 185–192.
Västilä, K., Kummu, M., Sangmanee, C., Chinvanno, S., 2010. Modelling climate change impacts on the flood pulse in the Lower Mekong floodplains. Journal of Water and Climate Change 01, 67–86. https://doi.org/10.2166/wcc.2010.008
- Response: We will add the reference to the revised manuscript.
line 60: a reference is needed for the sentence “During the remainder of the year, river flow gradually reduces to less than 10% (sometimes 5%) of its flood peak.”
- Response: We will add Adamson et al., 2009 as the reference for this sentence in the revised manuscript.
Line 71: I'm confused about the mentioning of the Chi-Mun basin here, as this is a basin upstream of what is normally considered the Mekong floodplains. Besides, this region has had water infrastructure for much longer than the rest of the MRB, so if there is anywhere in the basin where the flood pulse has been affected for a long time (since the 1970s), its in the Chi Mun.
- Response: Thanks for the comment. We have decided to remove this mentioning in the revised manuscript because we found that it doesn’t add much value and could only cause confusion.
Line 82: Vastila et al (2010) looked primarily at inundation changes from climate change. A more relevant ref to the ecological changes would be Arias et al (2014b):
Arias, M.E., Cochrane, T.A., Kummu, M., Lauri, H., Koponen, J., Holtgrieve, G.W., Piman, T., 2014. Impacts of hydropower and climate change on drivers of ecological productivity of Southeast Asia’s most important wetland. Ecological Modelling 272, 252–263.
- Response: We will add the reference to the revised manuscript.
Line 83: I suggest stating “hydropower development” instead of “booming hydropower dams”
- Response: This is a good suggestion. We will change the term to “rapid hydropower development” as we want to emphasize the rate of change that is happening.
Data and methodology:
Data: Given the poor relationship we know exists between water level observations and derived discharged, I would be very cautious with analysis based from results for stations 10-13, for both levels and discharge. In addition, stations 12 and 13 are strongly influenced by water infrastructure in the Delta, which is not considered in this study.
- Response: Thanks for the comment. We agree that there could be many uncertainties arising from the poor relationship between observed water level and derived discharge. However, our aim is to highlight the long-term changes of water balance and separate dam impacts from natural drivers as well as other human factors. Indeed, analysis based on water level would have more uncertainties as the river morphology can change and has been changing drastically in recent years due to dam trapping and sand mining. Thus, our analyses focus mainly on long-term average discharge would eliminate these uncertainties. In the revised manuscript, we will add a note about this point along with the discussion on missing infrastructure in the Delta.
Lines 156: Though I am aware that this model has been documented multiple times before, it might be informative to mention the basic hydrodynamic assumptions the model use (e.g., kinematic wave?)
- Response: This information has been provided in lines 159-160 of the original manuscript and we will add reference to the model’s development paper (Yamazaki et al., 2013) in case readers seek further information.
lines 165-166: is it CFD or CMFD? I think there is a typo here. I would discourage you from using the acronym CFD, which is a very common acronym used in water science and engineering for Computational Fluid Dynamics.
- Response: We have changed all CFD to CMFD in the revised manuscript.
Line 177: Just to clarify: Your model switches on dams only after their corresponding year, is that correct? How did you consider reservoir infilling?
- Response: Thanks for this note. Considering that the actual date of reservoir operation/filling is generally not available or collected widely, we believe it is a good assumption that the dams start filling even before the official commissioned dates. Thus, our model switches on dams at the beginning of the reported commission year and starts operating from that point instead of the end of that year. We will add clarification in the revised manuscript.
Lines 198/Figs S2-3: Though I see that fig S2-5 present time series of ERA5 vs biascorrected, having 12 frames per figures, each with 20+ years of time series, makes it really difficult to truly see the difference between the simulations. I recommend a simpler graph illustrating the biascorrection process for a single station and shorter timeframe.
- Response: We will add a supplementary figure showing comparison of simulated results before and after bias correction, with and without dam for a shorter timeframe.
I notice that statistical indicators for model performance were not mentioned in the Methodology section. They instead first appeared in the Results section. I suggest a mention of those indicators in Methodology.
- Response: We will add additional sentences mentioning the statistical indicators in the data processing section of the revised manuscript.
Results:
Model performance: What about testing for interannual variability and trends? Because the model predicts well seasonal variability, it does not mean that it also reproduces well how those patterns varied from year to year. Later on in the results you use the model results to make assertions on interannual variability and trends, but perhaps early on you have to demonstrate that your model (and input reanalyzed data) is in agreement with those trends that have been found from observations alone.
- Response: Thanks for raising this point. To further prove the model’s ability to capture the multidecadal variability and trends, we will provide an additional table in the supplementary comparing these characteristics between observed and simulated data at each selected station during the period when the observed data is available in the revised version.
line 221: Please correct the typo, Stung not Strung.
- Response: Thanks for a very detailed reading. We will correct this in the revised manuscript.
Lines 221-223: I am a bit concerned that the water level simulations provide worse results than discharge. This probably has to do with the river bathymetry representation used in the model. What did you use by the way? Perhaps I missed that from the methods. I am sure that the authors realize that discharge measurements are seldom made, and that the vast majority of the discharge observations in the Mekong are derived from water level observations via regression equations. Thus, if there are inconsistencies in the water level measurements (which I do not think is the explanation here) there are certainly inconsistencies with the discharge data. The Mekong floodplain becomes really flat south of Kratie, so we should be very skeptical of any discharge data calculated (not observed) for the lower four stations.
- Response: The river-floodplain topography parameters used in our simulation are derived from the MERIT Hydro dataset (Yamazaki et al., 2017) and we have noted this in section 2.2, line 157. Indeed, the discrepancy between observed and simulated water level was due to river bathymetry representation which could be caused by the following reasons: (1) river-floodplain topography parameters are fixed over the entire simulation period, (2) uncertainty from the original satellite product that was used to derive the MERIT Hydro dataset. However, the model considers both flow in river and the floodplain, thus, while the representation of the river bathymetry might not be highly accurate during the validation period, we believe the water balance of the area is reasonably accurate for further analyses. We will consider noting these issues in the revised version.
Lines 254-255: Here you state that the Lancang has experienced a decreasing trend, but do not really explain why. This is contrary to what has been reported for rivers starting in the Himalayas, where icemelt has increased flow in the upper parts of the Mekong and other rivers in the region:
Li, D. et al. High Mountain Asia hydropower systems threatened by climate-driven landscape instability. Nat. Geosci. 15, 520–530 (2022).
- Response: We have carefully studied the referenced paper and plan to refer in our manuscript along with some discussion. That being said, we are struggling to find a substantial connection between the study and the current manuscript. In Li et al., 2022, ice melt has been noted as a major factor in altering slope stability, which further increases the risk of landslide/glacial lake outbursts. While these outburst events could substantially increase downstream flood magnitude, there have been no events recorded in the Lancang region since the 1950s as shown in Figure 1a (Li et al., 2022). Additionally, Li et al., 2022 noted that “The snow-water equivalent of mountain snowpacks has also declined in recent years and is projected to decline drastically …” while also presented in Figure 2 that Lancang only has a limited amount of glacierized area (<200 km2), and it is melting rapidly (~1 m/yr). These findings suggest that the overall water balance in this area is decreasing, similar to our findings, especially in our Figure 3a on the annual volume trend. Our analysis and discussion noted in lines 259-263 show that the annual volume in the Lancang area has been more influenced by rainfall than snowfall, however, we have yet to conduct a similar analysis on snow melt/ice melt. Thus, to the extent possible, we will consider providing additional analysis on similarities between long-term trends of river flow and snow melt to explain this phenomenon.
lines 273-274: there’s a repetition of the word ‘increase’. Please fix.
- Response: Thanks for your thorough reading. We have removed the duplicated word.
Line 288: Regarding the comment about natural retention in the Tonle Sap, Stations 12 and 13 are deep into the Delta in Vietnam, quite a ways from the Tonle Sap, but in the middle of one of the largest rice production regions in the world. Thus, you need to show some strong evidence to support that the Tonle Sap retention has effects there greater and the massive and active water infrastructure network Vietnam has built in the Delta. This might be related with the calibration results for these stations being below optimal.
- Response: Thank you for raising this point. We will provide additional water balance analysis and seasonal timing comparisons between station 11 (upstream of Phnom Penh), Prek Kdam (Tonle Sap River) and a location on the Mekong mainstream that is directly downstream from Phnom Penh but yet to enter the Mekong Delta to provide additional evidence on the impact of Tonle Sap retention effect in the revised version.
line 317: “dams are generally delaying the wet season onset…”. The author should expand that part by discussing the wet season duration because the example inside the brackets is about duration, not onset.
- Response: We will add additional discussion on the onset in this part.
line 318: based on Figure 4e, I think longer duration is not just in the early 2000s but in the 2000s as seen in 2001, 2005, 2007, and 2009.
- Response: Indeed, the longer duration happened in some of the upstream locations during the mentioned years. We will add a sentence mentioning this more specifically in the revised version.
Figure 5: why not include a figure of ‘Natural anomaly of dry season volume’? This should be consistent with how Figure 4 was depicted.
- Response: Thanks for raising this point. We will consider adding another panel on the “natural anomaly of dry season volume” to Figure 5 and provide additional discussion on this in the revised manuscript.
Lines 367-370: This sentence here seems like a long way of describing what I think is an obvious logic (a decrease in peak flow decreases flow in the floodplains). Consider deleting or synthesizing.
- Response: While it is generally understood that a decrease in peak flow lowers the flow in the floodplains, we wanted to highlight that this decrease is due to dam impact. Furthermore, this sentence was meant to contrast the findings in the following sentences which is that an increase of flood occurrence in the inner areas is also caused by dam operation. We realize that it warrants some clarification in the revised manuscript.
Discussion:
Lines 403: with regards to changes in the Lancang and 3S basins, it would probably be good to compare and contrast your results with those studies that have looked at dam development specifically for those regions. For example:
Räsänen, T.A., Koponen, J., Lauri, H., Kummu, M., 2012. Downstream Hydrological Impacts of Hydropower Development in the Upper Mekong Basin. Water Resources Management 26, 3495–3513. https://doi.org/10.1007/s11269-012-0087-0
Piman, T., Cochrane, T.A., Arias, M.E., 2016. Effect of Proposed Large Dams on Water Flows and Hydropower Production in the Sekong, Sesan and Srepok Rivers of the Mekong Basin. River Research and Applications 32, 2095–2108. https://doi.org/10.1002/rra.3045
- Response: Thank you for your suggestion. We will have additional discussion comparing our findings with the suggested papers in the revised version.
With regards to the flood pulse, there is a newer paper that is probably worth reviewing:
Morovati, K., Tian, F., Kummu, M., Shi, L., Tudaji, M., Nakhaei, P., Alberto Olivares, M., 2023. Contributions from climate variation and human activities to flow regime change of Tonle Sap Lake from 2001 to 2020. Journal of Hydrology 616, 128800. https://doi.org/10.1016/j.jhydrol.2022.128800
- Response: We will review the suggested paper and add additional discussion on relevant findings.
Conclusions:
Lines 452-3: As mentioned earlier, this conclusion can only be drawn if you verify that your model effectively can replicate long-term observations.
- Response: We will address this issue as mentioned in previous comments.
Line 459: Remove “or higher” to avoid confusion.
- Response: We will remove it in the revised version.
Line 467: the use of static land cover is also a limitation, despite arguably not having substantial impacts on flow alterations compared to dams.
- Response: We will mention this in the revised version.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC1
-
RC2: 'Comment on egusphere-2023-3158', Anonymous Referee #2, 26 Feb 2024
The authors investigated the long-term trend and variability of hydrological extremes in the Mekong River basin under natural conditions as well as under the influence of growing dam development. The investigation is based on 83 years of hydrological data, simulated by a hydrodynamic model with and without representing the dams. Although quite a few previous studies investigated the Mekong’s hydrological regime, the use of relatively long hydrological data in this study has provided useful insights for the river’s water management that has multi-national and multi-sectoral importance. The paper is well-written with nice visualizations. I am overall in favor of the publication of this paper. That said, I agree with the comments by Reviewer-1, and add the following comments that could help improve the clarity and readability of the paper.
Major comments:
- Introduction: The first two paragraphs in the Introduction describe the general effects of human intervention on the globally important river basins. This is a bit too long. The authors could shorten this discussion and focus on the problem statement and research gap around the Mekong. In particular, given that Mekong's hydrologic regime has been studied by several other studies, I was not clear about the specific contribution of this study until reading the last couple of paragraphs. Therefore, revising and reshuffling the ideas in the Introduction could improve the readability of the paper.
- Methods (line 160): Why is the water released from multi-purpose dams set to optimize power generation? I suppose some large-storage dams in the Mekong prioritize water availability for irrigation rather than power generation.
Minor comments:
- Line 10: maybe ‘in the last few decades’ instead of ‘in the last decade’.
- Lind 15-16: maybe replace ‘has undergone’ with ‘shows’.
- Line 16 (and similar use thereafter): The use of ‘decadal trends and variabilities’ is not that clear. Are you referring to the changes/variability under large-scale climate drivers like ENSO?
- Line 21: Are 2019 and 2020 examples of normal years or years of shorter wet seasons?
- Figure on page 2: What is on the y-axis in the graphs? Also, you used the same color palette for different data on the maps and graphs which could be confusing. Maybe add a caption too.
- Line 36-37: Dams in general could benefit flood control but a majority of the Mekong's dams are for hydropower. Maybe it's worth mentioning that.
- Line 44: 'desiccation' seems a less commonly used word. Maybe change it.
- Line 50: The year for Zarfl et al. should be 2015.
- Line 67-68: Do you indicate the 'ecosystem services' by reliance on 'local communities' on flood pulse? Also, 'have developed to' probably does not fit to this sentence.
- Line 69-75: I suppose fish production is another important ecosystem service that could be mentioned here. Mekong is one of the key hotspots in the Global South where strong tension exists between freshwater resources and dam development for energy-economic growth. See the recent paper by Chowdhury et al. (2024): Hydropower expansion in eco-sensitive river basins under global energy-economic change. Nature Sustainability. DOI: https://doi.org/10.1038/s41893-023-01260-z
- Line 103: maybe replace ‘flood researches’ with ‘flood-related studies’.
- Line 154: What is the spatial resolution of the unit catchment? Could each unit catchment represent one or more reservoirs? Could some large reservoirs span over multiple unit catchments? A visual presentation (at least in SI) of the unit catchment could be helpful.
- Line 165: I agree with R1 about not using the "CFD" acronym here.
Citation: https://doi.org/10.5194/egusphere-2023-3158-RC2 -
AC3: 'Reply on RC2', Huy Dang, 14 Mar 2024
The authors investigated the long-term trend and variability of hydrological extremes in the Mekong River basin under natural conditions as well as under the influence of growing dam development. The investigation is based on 83 years of hydrological data, simulated by a hydrodynamic model with and without representing the dams. Although quite a few previous studies investigated the Mekong’s hydrological regime, the use of relatively long hydrological data in this study has provided useful insights for the river’s water management that has multi-national and multi-sectoral importance. The paper is well-written with nice visualizations. I am overall in favor of the publication of this paper. That said, I agree with the comments by Reviewer-1, and add the following comments that could help improve the clarity and readability of the paper.
- Response: We thank the referee for spending time reviewing our manuscript and providing constructive comments that helped improve the quality and clarity of the manuscript. We will revise the manuscript and provide additional information based on your thoughtful comments. A detailed, point-to-point response is provided in the following.
Major comments:
- Introduction: The first two paragraphs in the Introduction describe the general effects of human intervention on the globally important river basins. This is a bit too long. The authors could shorten this discussion and focus on the problem statement and research gap around the Mekong. In particular, given that Mekong's hydrologic regime has been studied by several other studies, I was not clear about the specific contribution of this study until reading the last couple of paragraphs. Therefore, revising and reshuffling the ideas in the Introduction could improve the readability of the paper.
- Response: Thanks for your recommendation. Indeed, we recognized that the general information at global scale in the first two paragraphs could be too long and it takes considerable reading effort to have a clear understanding of our studies contribution. Thus, to the best of our abilities, we will shorten the Introduction section and revise it as suggested.
- Methods (line 160): Why is the water released from multi-purpose dams set to optimize power generation? I suppose some large-storage dams in the Mekong prioritize water availability for irrigation rather than power generation.
- Response: Thank you for raising this, it is a good point. While we agree that multi-purpose dams can be prioritized for water availability for irrigation, water supply, or flood control instead of hydropower generation, this priority can be changed periodically based on local demand and government planning, which information is generally not available. In an ideal situation, these dams' operation should be controlled to meet different priorities in different season/flow situations, which we are working on to account for in our forthcoming studies. It should be noted that all 3 multi-purpose dams considered in our study are reported to have hydropower generation capacity. Additionally, while dams optimized for power generation might not meet the entire irrigation demands downstream, this demand should be partially satisfied from continuous dam release. Thus, we believe that our findings would not be drastically different if we had set the multi-purpose dams to operate as irrigation dams. In the revised manuscript, we will add additional clarification on this.
Minor comments:
- Line 10: maybe ‘in the last few decades’ instead of ‘in the last decade’.
- Response: Here, we specifically aim to highlight the 2010-2020 period considering that the basin-wide storage capacity has increased by almost 3 times compared to before that as shown in Figure 1.
- Lind 15-16: maybe replace ‘has undergone’ with ‘shows’.
- Response: Thank you; we will consider editing this in the revised manuscript.
- Line 16 (and similar use thereafter): The use of ‘decadal trends and variabilities’ is not that clear. Are you referring to the changes/variability under large-scale climate drivers like ENSO?
- Response: Yes, we were referring to the changes/variations under natural climate drivers such as El Nino, La Nina, etc.
- Line 21: Are 2019 and 2020 examples of normal years or years of shorter wet seasons?
- Response: Thank you for your comment. Those years are meant as examples of the years with shorter wet seasons. We will fix this in the revised manuscript.
- Figure on page 2: What is on the y-axis in the graphs? Also, you used the same color palette for different data on the maps and graphs which could be confusing. Maybe add a caption too.
- Response: The numbers on the y-axis of the top middle and right panels are the numbering order of the selected stations. Indeed, we agree that the current form of the figure can be confusing to readers without proper explanation. Thus, we will consider reorganizing this figure in the revised manuscript for more clarity.
- Line 36-37: Dams in general could benefit flood control but a majority of the Mekong's dams are for hydropower. Maybe it's worth mentioning that.
- Response: Thank you; we will add this information in the revised version.
- Line 44: 'desiccation' seems a less commonly used word. Maybe change it.
- Response: We will consider editing this in the revised version.
- Line 50: The year for Zarfl et al. should be 2015.
- Response: Thank you for your thorough review. We will fix this in the revised version.
- Line 67-68: Do you indicate the 'ecosystem services' by reliance on 'local communities' on flood pulse? Also, 'have developed to' probably does not fit to this sentence.
- Response: Here we refer to the long-term natural development of both ecosystem services (i.e., fish migration, wetlands growth, etc.) and local communities' culture and livelihood (fishery, rice cultivation) based on the timing of the Mekong’s wet season, which onset is signified by the flood pulse. We will edit this to provide more clarity in the revised version.
- Line 69-75: I suppose fish production is another important ecosystem service that could be mentioned here. Mekong is one of the key hotspots in the Global South where strong tension exists between freshwater resources and dam development for energy-economic growth. See the recent paper by Chowdhury et al. (2024): Hydropower expansion in eco-sensitive river basins under global energy-economic change. Nature Sustainability. DOI: https://doi.org/10.1038/s41893-023-01260-z
- Response: Thank you for your suggestions. We will add fish production and reference the suggested paper.
- Line 103: maybe replace ‘flood researches’ with ‘flood-related studies’.
- Response: We will replace this in the revised version.
- Line 154: What is the spatial resolution of the unit catchment? Could each unit catchment represent one or more reservoirs? Could some large reservoirs span over multiple unit catchments? A visual presentation (at least in SI) of the unit catchment could be helpful.
- Response: The spatial resolution of each unit catchment in our simulation is 5km. Currently, the applied operation scheme only allows the representation of one reservoir. There was only one case that two dams (release gate) are in one same cell, in which, we’ve considered to keep the one with larger storage capacity for simulation. Indeed, some of the reservoirs span over multiple unit catchments in our simulation. Lastly, thank you for your suggestion, we will add a figure in the supplementary showcasing the reservoirs extent in the revised version.
- Line 165: I agree with R1 about not using the "CFD" acronym here.
- Response: Yes, we agree, and we have changed all “CFD” in the manuscript to “CMFD”.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC3
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-3158', Mauricio Arias, 28 Jan 2024
<strong>Publisher’s note: this comment is a copy of RC1 and its content was therefore removed.</strong>
Citation: https://doi.org/10.5194/egusphere-2023-3158-CC1 -
AC2: 'Reply on CC1', Huy Dang, 28 Feb 2024
Thank you for your effort in reviewing our manuscript. Kindly find our response to your comments in the Reply to RC1.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC2
-
AC2: 'Reply on CC1', Huy Dang, 28 Feb 2024
-
RC1: 'Comment on egusphere-2023-3158', Mauricio Arias, 29 Jan 2024
hess-2023-3158 Dang
The manuscript presents an interesting and thorough examination of natural variation and dam impacts on the Mekong’s hydrological extremes over eight decades. The authors looked at a number of hydrological aspects such as flow trend, seasonal timings, flow volume, and extreme events and flooding patterns every decade. With the use of a hydrodynamic model, the author simulated flows with and without dams so as to quantify the dam impacts and look at Mekong’s hydrology under natural variability. Overall, the most significant contributions this study provided are (1) long-term trend of flow and (2) a full picture of natural flow and hydrological extremes, both drought and flood, compared to dam-altered flow and extremes in the whole Mekong Basin. In general, the manuscript in its current form is well written with insightful figures. That said, the manuscript can always use a bit of proof-reading and further synthesis.
While this study is certainly a step ahead of previous research in the Mekong looking at the effect of dams on river flows, the more comprehensive study results also means that the study ought to be controlled for two important factors which are directly linked to the conclusions of the study. First, while the study calibrated and validated the model in the conventional way, it is not clear to me that the authors are able to verify if the model captures the multidecadal variability and trends the study elaborates on. How can you ensure the reader that such term trends are real and not a model artifact? I suggest a stronger verification process, including how (or if) the study captures well-documented long-term variability and trends. See for instance the papers by Delgado (2010, 2012).
Another aspect I do not think the study controlled against well was water infrastructure in the floodplains and delta. As I hope the authors are aware, there has been widespread development of water control and irrigation infrastructure in both Cambodia and Vietnam, with effects on flooding well documented. Without taking a close look at what is happening with infrastructure within the floodplains, I would be extremely cautious about analysis and assertions made for those locations in Cambodia and Vietnam. The fact that the water level validation for these stations was not great may be a proof of this.
In addition to these general comments, I have several punctual comments and suggestions throughout the manuscript:
Abstract:
- Line 13: These are certainly not the first results of hydrological change in the Mekong. Consider deleting the word “first”.
- Lines 18-19: I suggest adding specific numbers associated with this claim that dams are intensifying variations in wet season flows.
Introduction:
- Line 29: I would be cautious with the use of the word "stable", as many can interpret this as a "flat hydrograph", which is the complete opposite from what the subsequent references allude to.
- Lines 49-50: Here is another newer reference related to dam impacts in biodiverse rivers:
Here is a good reference for this claim:
Winemiller, K.O., et al, 2016. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351, 128–129.
- Line 56: here you stay that the Mekong has “rather unpredictable seasonal variability”. Do you refer to interannual variability? I argue that the Mekong river flow intraannual/seasonal variability is highly predictable (wet season in May-Nov, driest months Feb-Apr). The fact that you have hydrological models with very high fit to observations supports the predictability of this system.
- Line 57: the statement here about seasonal timing sounds rather contradictory to the statement I talked about in the previous comment.
- Line 60: That the Mekong has a flood pulse dynamic has been documented for well over a decade, so consider updating this ref with one of the more historic ones:
Kummu, M., Sarkkula, J., 2008. Impact of the Mekong River flow alteration on the Tonle Sap flood pulse. Ambio 37, 185–192.
Västilä, K., Kummu, M., Sangmanee, C., Chinvanno, S., 2010. Modelling climate change impacts on the flood pulse in the Lower Mekong floodplains. Journal of Water and Climate Change 01, 67–86. https://doi.org/10.2166/wcc.2010.008
- line 60: a reference is needed for the sentence “During the remainder of the year, river flow gradually reduces to less than 10% (sometimes 5%) of its flood peak.”
- Line 71: I'm confused about the mentioning of the Chi-Mun basin here, as this is a basin upstream of what is normally considered the Mekong floodplains. Besides, this region has had water infrastructure for much longer than the rest of the MRB, so if there is anywhere in the basin where the flood pulse has been affected for a long time (since the 1970s), its in the Chi Mun.
- Line 82: Vastila et al (2010) looked primarily at inundation changes from climate change. A more relevant ref to the ecological changes would be Arias et al (2014b):
Arias, M.E., Cochrane, T.A., Kummu, M., Lauri, H., Koponen, J., Holtgrieve, G.W., Piman, T., 2014. Impacts of hydropower and climate change on drivers of ecological productivity of Southeast Asia’s most important wetland. Ecological Modelling 272, 252–263.
- Line 83: I suggest stating “hydropower development” instead of “booming hydropower dams”
Data and methodology:
- Data: Given the poor relationship we know exists between water level observations and derived discharged, I would be very cautious with analysis based from results for stations 10-13, for both levels and discharge. In addition, stations 12 and 13 are strongly influenced by water infrastructure in the Delta, which is not considered in this study.
- Lines 156: Though I am aware that this model has been documented multiple times before, it might be informative to mention the basic hydrodynamic assumptions the model use (e.g., kinematic wave?)
- lines 165-166: is it CFD or CMFD? I think there is a typo here. I would discourage you from using the acronym CFD, which is a very common acronym used in water science and engineering for Computational Fluid Dynamics.
- Line 177: Just to clarify: Your model switches on dams only after their corresponding year, is that correct? How did you consider reservoir infilling?
- Lines 198/Figs S2-3: Though I see that fig S2-5 present time series of ERA5 vs biascorrected, having 12 frames per figures, each with 20+ years of time series, makes it really difficult to truly see the difference between the simulations. I recommend a simpler graph illustrating the biascorrection process for a single station and shorter timeframe.
- I notice that statistical indicators for model performance were not mentioned in the Methodology section. They instead first appeared in the Results section. I suggest a mention of those indicators in Methodology.
Results:
- Model performance: What about testing for interannual variability and trends? Because the model predicts well seasonal variability, it does not mean that it also reproduces well how those patterns varied from year to year. Later on in the results you use the model results to make assertions on interannual variability and trends, but perhaps early on you have to demonstrate that your model (and input reanalyzed data) is in agreement with those trends that have been found from observations alone.
- line 221: Please correct the typo, Stung not Strung.
- Lines 221-223: I am a bit concerned that the water level simulations provide worse results than discharge. This probably has to do with the river bathymetry representation used in the model. What did you use by the way? Perhaps I missed that from the methods. I am sure that the authors realize that discharge measurements are seldom made, and that the vast majority of the discharge observations in the Mekong are derived from water level observations via regression equations. Thus, if there are inconsistencies in the water level measurements (which I do not think is the explanation here) there are certainly inconsistencies with the discharge data. The Mekong floodplain becomes really flat south of Kratie, so we should be very skeptical of any discharge data calculated (not observed) for the lower four stations.
- Lines 254-255: Here you state that the Lancang has experienced a decreasing trend, but do not really explain why. This is contrary to what has been reported for rivers starting in the Himalayas, where icemelt has increased flow in the upper parts of the Mekong and other rivers in the region:
Li, D. et al. High Mountain Asia hydropower systems threatened by climate-driven landscape instability. Nat. Geosci. 15, 520–530 (2022).
- lines 273-274: there’s a repetition of the word ‘increase’. Please fix.
- Line 288: Regarding the comment about natural retention in the Tonle Sap, Stations 12 and 13 are deep into the Delta in Vietnam, quite a ways from the Tonle Sap, but in the middle of one of the largest rice production regions in the world. Thus, you need to show some strong evidence to support that the Tonle Sap retention has effects there greater and the massive and active water infrastructure network Vietnam has built in the Delta. This might be related with the calibration results for these stations being below optimal.
- line 317: “dams are generally delaying the wet season onset…”. The author should expand that part by discussing the wet season duration because the example inside the brackets is about duration, not onset.
- line 318: based on Figure 4e, I think longer duration is not just in the early 2000s but in the 2000s as seen in 2001, 2005, 2007, and 2009.
- Figure 5: why not include a figure of ‘Natural anomaly of dry season volume’? This should be consistent with how Figure 4 was depicted.
- Lines 367-370: This sentence here seems like a long way of describing what I think is an obvious logic (a decrease in peak flow decreases flow in the floodplains). Consider deleting or synthesizing.
Discussion:
- Lines 403: with regards to changes in the Lancang and 3S basins, it would probably be good to compare and contrast your results with those studies that have looked at dam development specifically for those regions. For example:
Räsänen, T.A., Koponen, J., Lauri, H., Kummu, M., 2012. Downstream Hydrological Impacts of Hydropower Development in the Upper Mekong Basin. Water Resources Management 26, 3495–3513. https://doi.org/10.1007/s11269-012-0087-0
Piman, T., Cochrane, T.A., Arias, M.E., 2016. Effect of Proposed Large Dams on Water Flows and Hydropower Production in the Sekong, Sesan and Srepok Rivers of the Mekong Basin. River Research and Applications 32, 2095–2108. https://doi.org/10.1002/rra.3045
- With regards to the flood pulse, there is a newer paper that is probably worth reviewing:
Morovati, K., Tian, F., Kummu, M., Shi, L., Tudaji, M., Nakhaei, P., Alberto Olivares, M., 2023. Contributions from climate variation and human activities to flow regime change of Tonle Sap Lake from 2001 to 2020. Journal of Hydrology 616, 128800. https://doi.org/10.1016/j.jhydrol.2022.128800
Conclusions:
- Lines 452-3: As mentioned earlier, this conclusion can only be drawn if you verify that your model effectively can replicate long-term observations.
- Line 459: Remove “or higher” to avoid confusion.
- Line 467: the use of static land cover is also a limitation, despite arguably not having substantial impacts on flow alterations compared to dams.
Citation: https://doi.org/10.5194/egusphere-2023-3158-RC1 -
AC1: 'Reply on RC1', Huy Dang, 28 Feb 2024
The manuscript presents an interesting and thorough examination of natural variation and dam impacts on the Mekong’s hydrological extremes over eight decades. The authors looked at a number of hydrological metrics such as flow trend, seasonal timings, flow volume, and extreme events and flooding patterns every decade. With the use of a hydrodynamic model, the authors simulated flows with and without dams so as to quantify the dam impacts and look at Mekong’s hydrology under natural variability. Overall, the most significant contributions this study provided are (1) long-term trend of flow and (2) a full picture of natural flow and hydrological extremes, both drought and flood, compared to dam-altered flow and extremes in the whole Mekong Basin. In general, the manuscript in its current form is well written with insightful figures. That said, the manuscript can always use a bit of proof-reading and further synthesis.
- Response: We thank Dr. Mauricio Arias for taking the time to read the manuscript and providing constructive comments that incisively helped improve the quality and credibility of the manuscript. We have revised the manuscript and correlated our results with relevant findings from suggested sources and addressed other comments. A detailed, point-to-point response is provided in the following.
While this study is certainly a step ahead of previous research in the Mekong looking at the effect of dams on river flows, the more comprehensive study results also means that the study ought to be controlled for two important factors which are directly linked to the conclusions of the study. First, while the study calibrated and validated the model in the conventional way, it is not clear to me that the authors are able to verify if the model captures the multidecadal variability and trends the study elaborates on. How can you ensure the reader that such term trends are real and not a model artifact? I suggest a stronger verification process, including how (or if) the study captures well-documented long-term variability and trends. See for instance the papers by Delgado (2010, 2012, the first already cited in the manuscript) and Rasanen (https://www.hydrol-earth-syst-sci.net/17/2069/2013/; 10.1016/j.jhydrol.2012.10.028).
- Response: This is a good point, thanks for raising it. To further prove the model’s ability to capture the multidecadal variability and trends, we will provide additional validation of these characteristics against the observed data at each selected stations in the supplementary and include additional analysis specifically on this in the model performance section of the revised manuscript.
Another aspect I do not think the study controlled against was water infrastructure in the floodplains and delta. As I hope the authors are aware, there has been widespread development of water control and irrigation infrastructure in both Cambodia and Vietnam, with effects on flooding well documented. Without taking a close look at what is happening with infrastructure within the floodplains, I would be extremely cautious about analysis and assertions made for those locations in Cambodia and Vietnam. The fact that the water level validation for these stations was not great may be a proof of this.
- Response: Thanks for raising this important point. Indeed, infrastructure development in the delta region is an important issue. While the massive system of channels transferring water across the Delta has been considered in our model as linkage between grid cells, its representation is partly limited due to its spatial resolution. Additionally, the current model version doesn’t include the capability to account for dikes or other infrastructures. We are aware of those and have been working to account for these infrastructures in the model, which will be presented in our forthcoming articles. We believe that our findings would not be drastically different if we had accounted for these missing factors, especially considering that the analysis in Figure 6 shows the possible difference of flood occurrence in decadal term instead of actual magnitude while the remaining analyses are generally based on discharge. In the revised manuscript, we will add caution regarding any potential impact in our results caused by these factors.
In addition to these general comments, I have several punctual comments and suggestions throughout the manuscript:
Abstract:
Line 13: These are certainly not the first results of hydrological change in the Mekong. Consider deleting the word “first”.
- Response: We appreciate your comments, and we note that there have been previous studies providing critical insights into the hydrological change in the Mekong, some include long-term (~80 years) analysis such as Delgado et al. 2010. However, to the best of our knowledge, previous studies mainly analyze observations at selected stations or basin scale in short time periods, and none have yet to provide an analysis at basin scale of this temporal coverage. To address this issue, we will modify the language in the revised manuscript to avoid confusion and misinterpretations.
Lines 18-19: I suggest adding specific numbers associated with this claim that dams are intensifying variations in wet season flows.
- Response: Thanks for the suggestion. We will consider adding this information to the revised manuscript.
Introduction:
Line 29: I would be cautious with the use of the word "stable", as many can interpret this as a "flat hydrograph", which is the complete opposite from what the subsequent references allude to.
- Response: Thanks for the suggestions. We will change the wording to “A consistent pattern of river regime…” in the revised manuscript.
Lines 49-50: Here is another newer reference related to dam impacts in biodiverse rivers:
Winemiller, K.O., et al, 2016. Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science 351, 128–129.
- Response: Thanks for the suggestion. We will add this reference to the revised manuscript.
Line 56: here you stay that the Mekong has “rather unpredictable seasonal variability”. Do you refer to interannual variability? I argue that the Mekong river flow intraannual/seasonal variability is highly predictable (wet season in May-Nov, driest months Feb-Apr). The fact that you have hydrological models with very high fit to observations supports the predictability of this system.
- Response: Thanks for the comment. While there is a general understanding that the Mekong river’s wet season occurs during May-Nov, the actual onset and ending of it varies greatly (1-3 weeks) from year-to-year which is shown in Figure 4a. Additionally, the good fit from our model may suggest that it could replicate historical conditions based on reanalyzed data (ERA5), however, there would still be considerable uncertainties in predicting future seasons/conditions that have yet to happened. We will clarify this point in the revised manuscript.
Line 57: the statement here about seasonal timing sounds rather contradictory to the statement I talked about in the previous comment.
- Response: Thanks for the comment. We have addressed this in our response to comment #5 above.
Line 60: That the Mekong has a flood pulse dynamic has been documented for well over a decade, so consider updating this ref with one of the more historic ones:
Kummu, M., Sarkkula, J., 2008. Impact of the Mekong River flow alteration on the Tonle Sap flood pulse. Ambio 37, 185–192.
Västilä, K., Kummu, M., Sangmanee, C., Chinvanno, S., 2010. Modelling climate change impacts on the flood pulse in the Lower Mekong floodplains. Journal of Water and Climate Change 01, 67–86. https://doi.org/10.2166/wcc.2010.008
- Response: We will add the reference to the revised manuscript.
line 60: a reference is needed for the sentence “During the remainder of the year, river flow gradually reduces to less than 10% (sometimes 5%) of its flood peak.”
- Response: We will add Adamson et al., 2009 as the reference for this sentence in the revised manuscript.
Line 71: I'm confused about the mentioning of the Chi-Mun basin here, as this is a basin upstream of what is normally considered the Mekong floodplains. Besides, this region has had water infrastructure for much longer than the rest of the MRB, so if there is anywhere in the basin where the flood pulse has been affected for a long time (since the 1970s), its in the Chi Mun.
- Response: Thanks for the comment. We have decided to remove this mentioning in the revised manuscript because we found that it doesn’t add much value and could only cause confusion.
Line 82: Vastila et al (2010) looked primarily at inundation changes from climate change. A more relevant ref to the ecological changes would be Arias et al (2014b):
Arias, M.E., Cochrane, T.A., Kummu, M., Lauri, H., Koponen, J., Holtgrieve, G.W., Piman, T., 2014. Impacts of hydropower and climate change on drivers of ecological productivity of Southeast Asia’s most important wetland. Ecological Modelling 272, 252–263.
- Response: We will add the reference to the revised manuscript.
Line 83: I suggest stating “hydropower development” instead of “booming hydropower dams”
- Response: This is a good suggestion. We will change the term to “rapid hydropower development” as we want to emphasize the rate of change that is happening.
Data and methodology:
Data: Given the poor relationship we know exists between water level observations and derived discharged, I would be very cautious with analysis based from results for stations 10-13, for both levels and discharge. In addition, stations 12 and 13 are strongly influenced by water infrastructure in the Delta, which is not considered in this study.
- Response: Thanks for the comment. We agree that there could be many uncertainties arising from the poor relationship between observed water level and derived discharge. However, our aim is to highlight the long-term changes of water balance and separate dam impacts from natural drivers as well as other human factors. Indeed, analysis based on water level would have more uncertainties as the river morphology can change and has been changing drastically in recent years due to dam trapping and sand mining. Thus, our analyses focus mainly on long-term average discharge would eliminate these uncertainties. In the revised manuscript, we will add a note about this point along with the discussion on missing infrastructure in the Delta.
Lines 156: Though I am aware that this model has been documented multiple times before, it might be informative to mention the basic hydrodynamic assumptions the model use (e.g., kinematic wave?)
- Response: This information has been provided in lines 159-160 of the original manuscript and we will add reference to the model’s development paper (Yamazaki et al., 2013) in case readers seek further information.
lines 165-166: is it CFD or CMFD? I think there is a typo here. I would discourage you from using the acronym CFD, which is a very common acronym used in water science and engineering for Computational Fluid Dynamics.
- Response: We have changed all CFD to CMFD in the revised manuscript.
Line 177: Just to clarify: Your model switches on dams only after their corresponding year, is that correct? How did you consider reservoir infilling?
- Response: Thanks for this note. Considering that the actual date of reservoir operation/filling is generally not available or collected widely, we believe it is a good assumption that the dams start filling even before the official commissioned dates. Thus, our model switches on dams at the beginning of the reported commission year and starts operating from that point instead of the end of that year. We will add clarification in the revised manuscript.
Lines 198/Figs S2-3: Though I see that fig S2-5 present time series of ERA5 vs biascorrected, having 12 frames per figures, each with 20+ years of time series, makes it really difficult to truly see the difference between the simulations. I recommend a simpler graph illustrating the biascorrection process for a single station and shorter timeframe.
- Response: We will add a supplementary figure showing comparison of simulated results before and after bias correction, with and without dam for a shorter timeframe.
I notice that statistical indicators for model performance were not mentioned in the Methodology section. They instead first appeared in the Results section. I suggest a mention of those indicators in Methodology.
- Response: We will add additional sentences mentioning the statistical indicators in the data processing section of the revised manuscript.
Results:
Model performance: What about testing for interannual variability and trends? Because the model predicts well seasonal variability, it does not mean that it also reproduces well how those patterns varied from year to year. Later on in the results you use the model results to make assertions on interannual variability and trends, but perhaps early on you have to demonstrate that your model (and input reanalyzed data) is in agreement with those trends that have been found from observations alone.
- Response: Thanks for raising this point. To further prove the model’s ability to capture the multidecadal variability and trends, we will provide an additional table in the supplementary comparing these characteristics between observed and simulated data at each selected station during the period when the observed data is available in the revised version.
line 221: Please correct the typo, Stung not Strung.
- Response: Thanks for a very detailed reading. We will correct this in the revised manuscript.
Lines 221-223: I am a bit concerned that the water level simulations provide worse results than discharge. This probably has to do with the river bathymetry representation used in the model. What did you use by the way? Perhaps I missed that from the methods. I am sure that the authors realize that discharge measurements are seldom made, and that the vast majority of the discharge observations in the Mekong are derived from water level observations via regression equations. Thus, if there are inconsistencies in the water level measurements (which I do not think is the explanation here) there are certainly inconsistencies with the discharge data. The Mekong floodplain becomes really flat south of Kratie, so we should be very skeptical of any discharge data calculated (not observed) for the lower four stations.
- Response: The river-floodplain topography parameters used in our simulation are derived from the MERIT Hydro dataset (Yamazaki et al., 2017) and we have noted this in section 2.2, line 157. Indeed, the discrepancy between observed and simulated water level was due to river bathymetry representation which could be caused by the following reasons: (1) river-floodplain topography parameters are fixed over the entire simulation period, (2) uncertainty from the original satellite product that was used to derive the MERIT Hydro dataset. However, the model considers both flow in river and the floodplain, thus, while the representation of the river bathymetry might not be highly accurate during the validation period, we believe the water balance of the area is reasonably accurate for further analyses. We will consider noting these issues in the revised version.
Lines 254-255: Here you state that the Lancang has experienced a decreasing trend, but do not really explain why. This is contrary to what has been reported for rivers starting in the Himalayas, where icemelt has increased flow in the upper parts of the Mekong and other rivers in the region:
Li, D. et al. High Mountain Asia hydropower systems threatened by climate-driven landscape instability. Nat. Geosci. 15, 520–530 (2022).
- Response: We have carefully studied the referenced paper and plan to refer in our manuscript along with some discussion. That being said, we are struggling to find a substantial connection between the study and the current manuscript. In Li et al., 2022, ice melt has been noted as a major factor in altering slope stability, which further increases the risk of landslide/glacial lake outbursts. While these outburst events could substantially increase downstream flood magnitude, there have been no events recorded in the Lancang region since the 1950s as shown in Figure 1a (Li et al., 2022). Additionally, Li et al., 2022 noted that “The snow-water equivalent of mountain snowpacks has also declined in recent years and is projected to decline drastically …” while also presented in Figure 2 that Lancang only has a limited amount of glacierized area (<200 km2), and it is melting rapidly (~1 m/yr). These findings suggest that the overall water balance in this area is decreasing, similar to our findings, especially in our Figure 3a on the annual volume trend. Our analysis and discussion noted in lines 259-263 show that the annual volume in the Lancang area has been more influenced by rainfall than snowfall, however, we have yet to conduct a similar analysis on snow melt/ice melt. Thus, to the extent possible, we will consider providing additional analysis on similarities between long-term trends of river flow and snow melt to explain this phenomenon.
lines 273-274: there’s a repetition of the word ‘increase’. Please fix.
- Response: Thanks for your thorough reading. We have removed the duplicated word.
Line 288: Regarding the comment about natural retention in the Tonle Sap, Stations 12 and 13 are deep into the Delta in Vietnam, quite a ways from the Tonle Sap, but in the middle of one of the largest rice production regions in the world. Thus, you need to show some strong evidence to support that the Tonle Sap retention has effects there greater and the massive and active water infrastructure network Vietnam has built in the Delta. This might be related with the calibration results for these stations being below optimal.
- Response: Thank you for raising this point. We will provide additional water balance analysis and seasonal timing comparisons between station 11 (upstream of Phnom Penh), Prek Kdam (Tonle Sap River) and a location on the Mekong mainstream that is directly downstream from Phnom Penh but yet to enter the Mekong Delta to provide additional evidence on the impact of Tonle Sap retention effect in the revised version.
line 317: “dams are generally delaying the wet season onset…”. The author should expand that part by discussing the wet season duration because the example inside the brackets is about duration, not onset.
- Response: We will add additional discussion on the onset in this part.
line 318: based on Figure 4e, I think longer duration is not just in the early 2000s but in the 2000s as seen in 2001, 2005, 2007, and 2009.
- Response: Indeed, the longer duration happened in some of the upstream locations during the mentioned years. We will add a sentence mentioning this more specifically in the revised version.
Figure 5: why not include a figure of ‘Natural anomaly of dry season volume’? This should be consistent with how Figure 4 was depicted.
- Response: Thanks for raising this point. We will consider adding another panel on the “natural anomaly of dry season volume” to Figure 5 and provide additional discussion on this in the revised manuscript.
Lines 367-370: This sentence here seems like a long way of describing what I think is an obvious logic (a decrease in peak flow decreases flow in the floodplains). Consider deleting or synthesizing.
- Response: While it is generally understood that a decrease in peak flow lowers the flow in the floodplains, we wanted to highlight that this decrease is due to dam impact. Furthermore, this sentence was meant to contrast the findings in the following sentences which is that an increase of flood occurrence in the inner areas is also caused by dam operation. We realize that it warrants some clarification in the revised manuscript.
Discussion:
Lines 403: with regards to changes in the Lancang and 3S basins, it would probably be good to compare and contrast your results with those studies that have looked at dam development specifically for those regions. For example:
Räsänen, T.A., Koponen, J., Lauri, H., Kummu, M., 2012. Downstream Hydrological Impacts of Hydropower Development in the Upper Mekong Basin. Water Resources Management 26, 3495–3513. https://doi.org/10.1007/s11269-012-0087-0
Piman, T., Cochrane, T.A., Arias, M.E., 2016. Effect of Proposed Large Dams on Water Flows and Hydropower Production in the Sekong, Sesan and Srepok Rivers of the Mekong Basin. River Research and Applications 32, 2095–2108. https://doi.org/10.1002/rra.3045
- Response: Thank you for your suggestion. We will have additional discussion comparing our findings with the suggested papers in the revised version.
With regards to the flood pulse, there is a newer paper that is probably worth reviewing:
Morovati, K., Tian, F., Kummu, M., Shi, L., Tudaji, M., Nakhaei, P., Alberto Olivares, M., 2023. Contributions from climate variation and human activities to flow regime change of Tonle Sap Lake from 2001 to 2020. Journal of Hydrology 616, 128800. https://doi.org/10.1016/j.jhydrol.2022.128800
- Response: We will review the suggested paper and add additional discussion on relevant findings.
Conclusions:
Lines 452-3: As mentioned earlier, this conclusion can only be drawn if you verify that your model effectively can replicate long-term observations.
- Response: We will address this issue as mentioned in previous comments.
Line 459: Remove “or higher” to avoid confusion.
- Response: We will remove it in the revised version.
Line 467: the use of static land cover is also a limitation, despite arguably not having substantial impacts on flow alterations compared to dams.
- Response: We will mention this in the revised version.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC1
-
RC2: 'Comment on egusphere-2023-3158', Anonymous Referee #2, 26 Feb 2024
The authors investigated the long-term trend and variability of hydrological extremes in the Mekong River basin under natural conditions as well as under the influence of growing dam development. The investigation is based on 83 years of hydrological data, simulated by a hydrodynamic model with and without representing the dams. Although quite a few previous studies investigated the Mekong’s hydrological regime, the use of relatively long hydrological data in this study has provided useful insights for the river’s water management that has multi-national and multi-sectoral importance. The paper is well-written with nice visualizations. I am overall in favor of the publication of this paper. That said, I agree with the comments by Reviewer-1, and add the following comments that could help improve the clarity and readability of the paper.
Major comments:
- Introduction: The first two paragraphs in the Introduction describe the general effects of human intervention on the globally important river basins. This is a bit too long. The authors could shorten this discussion and focus on the problem statement and research gap around the Mekong. In particular, given that Mekong's hydrologic regime has been studied by several other studies, I was not clear about the specific contribution of this study until reading the last couple of paragraphs. Therefore, revising and reshuffling the ideas in the Introduction could improve the readability of the paper.
- Methods (line 160): Why is the water released from multi-purpose dams set to optimize power generation? I suppose some large-storage dams in the Mekong prioritize water availability for irrigation rather than power generation.
Minor comments:
- Line 10: maybe ‘in the last few decades’ instead of ‘in the last decade’.
- Lind 15-16: maybe replace ‘has undergone’ with ‘shows’.
- Line 16 (and similar use thereafter): The use of ‘decadal trends and variabilities’ is not that clear. Are you referring to the changes/variability under large-scale climate drivers like ENSO?
- Line 21: Are 2019 and 2020 examples of normal years or years of shorter wet seasons?
- Figure on page 2: What is on the y-axis in the graphs? Also, you used the same color palette for different data on the maps and graphs which could be confusing. Maybe add a caption too.
- Line 36-37: Dams in general could benefit flood control but a majority of the Mekong's dams are for hydropower. Maybe it's worth mentioning that.
- Line 44: 'desiccation' seems a less commonly used word. Maybe change it.
- Line 50: The year for Zarfl et al. should be 2015.
- Line 67-68: Do you indicate the 'ecosystem services' by reliance on 'local communities' on flood pulse? Also, 'have developed to' probably does not fit to this sentence.
- Line 69-75: I suppose fish production is another important ecosystem service that could be mentioned here. Mekong is one of the key hotspots in the Global South where strong tension exists between freshwater resources and dam development for energy-economic growth. See the recent paper by Chowdhury et al. (2024): Hydropower expansion in eco-sensitive river basins under global energy-economic change. Nature Sustainability. DOI: https://doi.org/10.1038/s41893-023-01260-z
- Line 103: maybe replace ‘flood researches’ with ‘flood-related studies’.
- Line 154: What is the spatial resolution of the unit catchment? Could each unit catchment represent one or more reservoirs? Could some large reservoirs span over multiple unit catchments? A visual presentation (at least in SI) of the unit catchment could be helpful.
- Line 165: I agree with R1 about not using the "CFD" acronym here.
Citation: https://doi.org/10.5194/egusphere-2023-3158-RC2 -
AC3: 'Reply on RC2', Huy Dang, 14 Mar 2024
The authors investigated the long-term trend and variability of hydrological extremes in the Mekong River basin under natural conditions as well as under the influence of growing dam development. The investigation is based on 83 years of hydrological data, simulated by a hydrodynamic model with and without representing the dams. Although quite a few previous studies investigated the Mekong’s hydrological regime, the use of relatively long hydrological data in this study has provided useful insights for the river’s water management that has multi-national and multi-sectoral importance. The paper is well-written with nice visualizations. I am overall in favor of the publication of this paper. That said, I agree with the comments by Reviewer-1, and add the following comments that could help improve the clarity and readability of the paper.
- Response: We thank the referee for spending time reviewing our manuscript and providing constructive comments that helped improve the quality and clarity of the manuscript. We will revise the manuscript and provide additional information based on your thoughtful comments. A detailed, point-to-point response is provided in the following.
Major comments:
- Introduction: The first two paragraphs in the Introduction describe the general effects of human intervention on the globally important river basins. This is a bit too long. The authors could shorten this discussion and focus on the problem statement and research gap around the Mekong. In particular, given that Mekong's hydrologic regime has been studied by several other studies, I was not clear about the specific contribution of this study until reading the last couple of paragraphs. Therefore, revising and reshuffling the ideas in the Introduction could improve the readability of the paper.
- Response: Thanks for your recommendation. Indeed, we recognized that the general information at global scale in the first two paragraphs could be too long and it takes considerable reading effort to have a clear understanding of our studies contribution. Thus, to the best of our abilities, we will shorten the Introduction section and revise it as suggested.
- Methods (line 160): Why is the water released from multi-purpose dams set to optimize power generation? I suppose some large-storage dams in the Mekong prioritize water availability for irrigation rather than power generation.
- Response: Thank you for raising this, it is a good point. While we agree that multi-purpose dams can be prioritized for water availability for irrigation, water supply, or flood control instead of hydropower generation, this priority can be changed periodically based on local demand and government planning, which information is generally not available. In an ideal situation, these dams' operation should be controlled to meet different priorities in different season/flow situations, which we are working on to account for in our forthcoming studies. It should be noted that all 3 multi-purpose dams considered in our study are reported to have hydropower generation capacity. Additionally, while dams optimized for power generation might not meet the entire irrigation demands downstream, this demand should be partially satisfied from continuous dam release. Thus, we believe that our findings would not be drastically different if we had set the multi-purpose dams to operate as irrigation dams. In the revised manuscript, we will add additional clarification on this.
Minor comments:
- Line 10: maybe ‘in the last few decades’ instead of ‘in the last decade’.
- Response: Here, we specifically aim to highlight the 2010-2020 period considering that the basin-wide storage capacity has increased by almost 3 times compared to before that as shown in Figure 1.
- Lind 15-16: maybe replace ‘has undergone’ with ‘shows’.
- Response: Thank you; we will consider editing this in the revised manuscript.
- Line 16 (and similar use thereafter): The use of ‘decadal trends and variabilities’ is not that clear. Are you referring to the changes/variability under large-scale climate drivers like ENSO?
- Response: Yes, we were referring to the changes/variations under natural climate drivers such as El Nino, La Nina, etc.
- Line 21: Are 2019 and 2020 examples of normal years or years of shorter wet seasons?
- Response: Thank you for your comment. Those years are meant as examples of the years with shorter wet seasons. We will fix this in the revised manuscript.
- Figure on page 2: What is on the y-axis in the graphs? Also, you used the same color palette for different data on the maps and graphs which could be confusing. Maybe add a caption too.
- Response: The numbers on the y-axis of the top middle and right panels are the numbering order of the selected stations. Indeed, we agree that the current form of the figure can be confusing to readers without proper explanation. Thus, we will consider reorganizing this figure in the revised manuscript for more clarity.
- Line 36-37: Dams in general could benefit flood control but a majority of the Mekong's dams are for hydropower. Maybe it's worth mentioning that.
- Response: Thank you; we will add this information in the revised version.
- Line 44: 'desiccation' seems a less commonly used word. Maybe change it.
- Response: We will consider editing this in the revised version.
- Line 50: The year for Zarfl et al. should be 2015.
- Response: Thank you for your thorough review. We will fix this in the revised version.
- Line 67-68: Do you indicate the 'ecosystem services' by reliance on 'local communities' on flood pulse? Also, 'have developed to' probably does not fit to this sentence.
- Response: Here we refer to the long-term natural development of both ecosystem services (i.e., fish migration, wetlands growth, etc.) and local communities' culture and livelihood (fishery, rice cultivation) based on the timing of the Mekong’s wet season, which onset is signified by the flood pulse. We will edit this to provide more clarity in the revised version.
- Line 69-75: I suppose fish production is another important ecosystem service that could be mentioned here. Mekong is one of the key hotspots in the Global South where strong tension exists between freshwater resources and dam development for energy-economic growth. See the recent paper by Chowdhury et al. (2024): Hydropower expansion in eco-sensitive river basins under global energy-economic change. Nature Sustainability. DOI: https://doi.org/10.1038/s41893-023-01260-z
- Response: Thank you for your suggestions. We will add fish production and reference the suggested paper.
- Line 103: maybe replace ‘flood researches’ with ‘flood-related studies’.
- Response: We will replace this in the revised version.
- Line 154: What is the spatial resolution of the unit catchment? Could each unit catchment represent one or more reservoirs? Could some large reservoirs span over multiple unit catchments? A visual presentation (at least in SI) of the unit catchment could be helpful.
- Response: The spatial resolution of each unit catchment in our simulation is 5km. Currently, the applied operation scheme only allows the representation of one reservoir. There was only one case that two dams (release gate) are in one same cell, in which, we’ve considered to keep the one with larger storage capacity for simulation. Indeed, some of the reservoirs span over multiple unit catchments in our simulation. Lastly, thank you for your suggestion, we will add a figure in the supplementary showcasing the reservoirs extent in the revised version.
- Line 165: I agree with R1 about not using the "CFD" acronym here.
- Response: Yes, we agree, and we have changed all “CFD” in the manuscript to “CMFD”.
Citation: https://doi.org/10.5194/egusphere-2023-3158-AC3
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 506 | 225 | 27 | 758 | 53 | 18 | 21 |
- HTML: 506
- PDF: 225
- XML: 27
- Total: 758
- Supplement: 53
- BibTeX: 18
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1936 KB) - Metadata XML
-
Supplement
(2881 KB) - BibTeX
- EndNote
- Final revised paper