the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Sep 2025
Disrupted Flow Memory and Synchrony in the Mekong River under Dam Regulation and Climate Change: Implications for Tonle Sap Reverse Flow
Abstract. Dam construction and climate change have profoundly disrupted the hydrological dynamics of the Mekong River and its floodplain–lake system. This study provides an integrated, multi-scale assessment of flow regime alteration across the Mekong mainstream and its coupling with Tonle Sap Lake for three periods: pre-dam (1976–1991), transition (1992–2009), and post-dam (2010–2024). Using daily and sub-daily data from eight stations, we quantify changes in long-term flow memory, short-term variability, and flow amplitude. These analyses are complemented by a hydrodynamic response-time model that simulates how upstream regime shifts reshape the Tonle Sap–Mekong flow exchange.
We show that dam regulation and climate change have fragmented both spatial synchrony and temporal persistence, disrupting the Mekong’s behavior as a coherent hydrological continuum. Upstream stations, particularly Chiang Saen, exhibit a 33.3 % increase in minimum flows and a 40 % decrease in maximum flows during the post-dam period compared to the pre-dam period, alongside higher flow memory (+0.13), reflecting flow smoothing by reservoir operations. Downstream variability remains more pronounced due to the influence of monsoonal tributaries. At the most downstream station–Kratie, a key control point for Tonle Sap inflow–maximum flows decreased by 9.4 %, while minimum flows increased by 117 % in the post-dam period relative to the pre-dam baseline, also delaying the seasonal timing of peak discharge by two weeks.
Critically, these regime shifts–driven by the compounding impacts of dam regulations and climate change, together with observed riverbed lowering from sand mining, caused the discharge threshold for initiating reverse flow to rise significantly: the median onset flow increased from ~3,000 m³ s⁻¹ (pre-dam) to ~7,000 m³ s⁻¹ (post-dam), marking a >130 % increase. Collectively, these alterations have shortened the reverse flow period by 24 days compared to the historical baseline. Our findings demonstrate that cascading dam operations across the mainstream and tributaries, amplified by climate change and other anthropogenic stressors, have triggered multi-scalar hydrological fragmentation in one of the world's most ecologically productive river–lake systems. Preserving the Mekong's natural flood pulse and its interaction with Tonle Sap lake will require basin-wide hydrological monitoring and transboundary governance frameworks that account not only for volume but also for flow timing, variability, and ecological function.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Hydrology and Earth System Sciences.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2077 KB) - Metadata XML
-
Supplement
(856 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-3472', Anonymous Referee #1, 17 Jan 2026
-
AC1: 'Reply on RC1', Khosro Morovati, 19 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3472/egusphere-2025-3472-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Khosro Morovati, 19 Jan 2026
-
RC2: 'Comment on egusphere-2025-3472', Anonymous Referee #2, 06 Feb 2026
General comment:
The study addresses an important and timely problem by examining how hydropower development and climate change have altered the flow regime of the Mekong River. Using historical observations and a combination of hydrological and hydrodynamic modelling, the authors estimate statistical indicators of flow memory and synchrony. They show substantial post-dam changes at several mainstem gauges when compared to their pre-dam equivalents. This component of the analysis is methodologically sound, well supported by the results presented and broadly consistent with previous studies that have shown dampened wet-season flow peaks and enhanced dry-season flows under dam regulation. The attempt to link these changes in Mekong flow discharge to alterations in the reversal of the Tonle Sap River however is less convincing. The supporting results are limited relative to the strength of the claims and key relevant literature on the Tonle Sap-Mekong system is not adequately considered. As a result, the conclusions regarding reverse-flow dynamics would require more rigorous modelling and analysis to improve the manuscript. Please find my specific comments and technical corrections below:
Specific comments:
The terms ‘flow memory’ and ‘flow synchrony’ are used throughout the manuscript including the title, but they are not fully introduced and their significance is not properly explained. I suggest that a short description is added in the introduction.
Lines 21-23: The statement made here is misguided. The discharge threshold that must be exceeded in order for the Tonle Sap River to reverse its flow is governed by the geometrical characteristics of the Tonle Sap and Mekong Rivers and their confluence. These geometrical characteristics are indeed altered by sand-mining induced channel deepening. Mekong’s flow discharge, which is affected by dam regulation and climate change, may control whether and for how long this threshold is exceeded but does not affect the threshold itself as implied here.
Lines 40-50: The authors correctly argue that previous literature has already extensively studied the impacts of hydropower development on Mekong hydrology. Here also please specify the start and end of for the dry and wet seasons.
Lines 51-53: The authors here argue that the novelty of their study is on the incorporation of flow memory and synchrony but they do not explain this further or justify why this analysis is important.
Lines 53-66: Here the focus shifts towards the Tonle Sap River flow reversal but key literature that has studied this topic is omitted. The authors should review previous work focusing on the effects of climate change and flow modulation by dams (see for example: Wang et al., Environ. Res. Lett. 15, 0940a1 (2020); Frappart, F. et al Sci. Total Environ. 636, 1520–1533 (2018); Kummu and Sarkkula, Ambio 37, 185–192 (2008)) and sand mining (see for example: Quan L.Q. et al., Nat Sustain 8, 1455–1466 (2025)). When relevant literature is considered, the statement made in lines 63-66 is not supported.
Line 131: Please specify the quality checks that were applied to the data, what proportion of the data did not pass the quality control and how were these values replaced.
Line 171: Figure 1 does not show the Delft 3D model domain, I suggest making a separate figure for this.
Lines 195-196: Please specify the distance between neighbouring cross-sections. This could improve confidence in the bathymetric interpolation that was applied.
Lines 200-203: Could you show how well does the derived DEM approximate natural river and lake morphology. Can the error be quantified? It should be noted here that the supplement provides only a single cross-section as an example of good fit (Figure S1).
Lines 234-238: Please provide more detailed information on the data that were used in the hydrological model, including sources.
Lines 276-277: The manuscript here refers to the supplement for detailed validation of the models used in the study. Figure S5 of the supplement shows that on some occasions the model underpredicts peak discharge values for Kompong Luong while simultaneously overpredicting for Prek Kdam (the opposite also occurs), what are the implications of these discrepancies for the simulated hydraulic head and the reversal of the Tonle Sap River? Can you clarify what is an ‘acceptable limit’ for RMSE values, mentioned in the supplement right before Figure S5?
Lines 279-282: A sentence should be added here to explain that the no-dam scenario data presented are outputs from the THREW model.
Lines 292-300: The RBI patterns described in the text do not reflect what is shown in Figure 2 panels b and e. In addition to the discrepancies between text and Figure 2, post-dam RBI (panel b) and measured RBI (panel e) should be identical, but this is not the case here.
Lines 413-415: It is unclear what these lines refer to. Earlier it has been demonstrated (Lines 355-377) that the median value of the annual maximum discharge is reduced in the post-dam period by 9%. RBI flashiness is also reduced at Kratie in the post-dam period based on Figure 2. These results do not support the claim of intensification of the hydropeaks.
Lines 421-425: The argument here is constructed in a confusing way and it is not clear why cessation of the reversal of the Tonle Sap River requires high discharge values at Kratie.
Lines 427-428: The authors omit the study published by Quan et al., Nat Sustain 8, 1455–1466 (2025) which demonstrates the impacts of sand mining on the flow reversal of the Tonle Sap River.
Lines 435-436: The effect of channel deepening which has drastically changed the hydraulic head required to reverse the flow of the Tonle Sap River should be included here
Lines 439-444: The argument here is speculative. The figure shows very well the modulation of Mekong’s water flux in the post-dam era and the shortening of the duration of the TSR reversal but do not show the drivers for these changes.
Lines 484-485 The statement here is misguided. The reduction of Mekong water discharge cannot affect the threshold required to initiate flow reversal in the Tonle Sap River. This threshold is governed by channel geometry. The magnitude of the hydropeak affects when and for how long this threshold is exceeded and reversal occurs.
Figure 1: The labels of the panels need attention as two panels are labelled as ‘(a)’. Then later in the main text (line 192) ‘Figure 1 panel c’ is mentioned. Also, you should provide the sources for the data presented (for example on dam locations).
Figure 2: I am not convinced that the use of rose diagrams is appropriate here. For example, a connection between the furthest upstream station (Chiang Saen) and most downstream (Kratie) is implied. I suggest using line plots with stations placed in order along the x-axis, possibly with the in-between distances scaled according to the station km point along the Mekong mainstem. Also, please explain the abbreviations used for the names of the stations in the figure caption.
Figure 3: Presenting data as monthly averages using all years (right panels) and annual averages (left panes) suppresses information that would be helpful to understand spatio-temporal changes of the metrics. For example, the monthly data show an uptick for Nakhon Phanom in amplitude and flashiness for January, is this primarily driven by the huge spike in 2022?
Also on Figure 3: Extreme values in 2018 for Chiang Khan (amplitude and flashiness) and Pakse (all metrics) and 2022 Nakhon Phanom (all metrics) should be discussed in the text.
Technical corrections:
Line10: add ‘the’ before Tonle Sap Lake
Line 99 and throughout the manuscript MCM is not a universal abbreviation. I suggest to use M m3
Line 126: replace ‘part (a)’ with ‘panel (a)’
Line 170: delete ‘of the Sea’
Line 180: the hydrological model is described in Sect. 2.5.3
Line 408: the correct figure is 5a (not 6a)
Line 421: Replace ‘Panel 5b’ with ‘Figure 5b’
Throughout the manuscript there is an overuse of em dashes (–). In most cases these should be replaced with commas or with hyphens when used for ranges (e.g. 1976-1991, not 1976–1991) or connections (e.g. Tonle Sap-Mekong , not Tonle Sap–Mekong).
Citation: https://doi.org/10.5194/egusphere-2025-3472-RC2 -
AC2: 'Reply on RC2', Khosro Morovati, 07 Feb 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-3472/egusphere-2025-3472-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Khosro Morovati, 07 Feb 2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 2,307 | 275 | 24 | 2,606 | 65 | 30 | 30 |
- HTML: 2,307
- PDF: 275
- XML: 24
- Total: 2,606
- Supplement: 65
- BibTeX: 30
- EndNote: 30
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
The study provides a comprehensive and timely analysis of the alterations in flow regime in the Mekong mainstream and their intricate, yet crucial linkage with the Tonle Sap Lake system, both of which have entered a critical phase of change under dam regulation and climate change. This study offers a new perspective on the changes in river flow regime and river-lake connectivity through alternative hydrological metrics, flow extremes, and the response time, as well as reverse flow periods. While the methodology is robust and the conclusions are generally well-supported, several key concerns require clarification and improvement for better readability and scientific rigor. Please find my specific and minor comments as follows.
Specific comments:
L109-L113: Since the study importantly points to the Mekong River-Tonle Sap Lake dynamics, more numerical details on the hydrology/hydrodynamics of the lake are necessary. For instance, Tonle Sap Lake is also contributed by the Tonle Sap tributaries to an extent that is, however, less than the reverse flow contribution.
Section 2.2: The section did not list Phnom Penh Port station, while Figure 1 depicts it. What is the role of the station in the study? Even if the station’s location is used, not the hydrological data, how it was used should be clarified.
Section 2.3: Weather data for the THREW model were not introduced. Please provide the details and sources of all input data for the hydrological and hydrodynamic models.
Section 2.4: This section should appear before Data Sources and Preprocessing, as it gives precedent information on changing morphology in the segregated periods.
L187: Is a one-year warm-up period good enough to initialize the model, given the complex system? Should there be any potential limitations pertaining to this setup?
Sections 2.5.1, 2.5.2, and 2.5.3: I believe the order of the three sections should be reorganized, as the development of the hydrological model is crucial as the boundary condition for the hydrodynamic model, and then the response model was embedded in the hydrodynamic framework. This is also in line with the order of the analyses.
Section 2.5.2: The response time model is for the stretch between Kratie and the Mekong-Tonle Sap confluence, which is in Phnom Penh, specifically at the Phnom Penh Port location. However, it was stated in Section 2.2 that Prek Kdam and Kompong Loung were listed for hydrodynamic validation and also discharge lag adjustment, implying that either station was considered for adjusting the discharge lags. Please clarify both sections.
L288: Higher monsoonal rainfall predominantly in the lower Mekong can also be highlighted in addition to tributary inflows.
Section 3.3: Some of the methods, like amplitude and peak count, should be briefly explained prior in the Method section.
Figure 3: The figure and its caption mention 2018-2024, but the caption also states 2017-2024. Could you elaborate on the difference?
Section 4.1: I find that this discussion section provides new major results along with their discussion. Those results are directly relevant to the topic and objectives. It makes more sense to transition those findings to the Results section; hence, this leaves more room for their implications to be expanded in the Discussion. Equally important is that previous studies should be cited in the section to enhance the interpretation of results and discussion, as the Mekong-Tonle Sap Lake connectivity has been increasingly explored with a wide range of implications beyond hydrodynamics.
Minor comments:
L24-25: It is better to mention “one of the world’s most ecologically productive river–lake systems” early in the Abstract to highlight the significance of the study area.
L45 and L47: Please clearly indicate the months in the wet and dry seasons.
L46: The relative locations of Chiang Saen and Kratie stations should be addressed for the first time (i.e., the most upstream and downstream stations).
L80: For the last paragraph, I noticed that sub-daily variability was not indicated, although it was analysed with its own subsection in the Results. It should be listed here in the Introduction for a full picture of the objectives.
L126: The Delft3D-Flow hydrodynamic model first appeared in Section 2.2. It would be better to mention it first in the Introduction.
L133: (see Figure 2d–f) should be moved to the end of the sentence.
L140: Provide the full form of PMFM.
L180: Should (see Sect. 2.5.2) actually refer to Sect. 2.5.3? Please also verify all other cross-references.
L192: Figure 1c should be Figure 1b. Please correct.
L239: Please mention the period of calibration and validation for the model in the main text, although this appears in the supplementary file.
L279: Change (a-f) to (Figure 2a-f) or (Figure 2)
Figure 4: The cross-reference of Figure 4d is apparently missing in the text.
L408: Figure 6a should be 5a.
L421 and L431: Panel 5b and Panel 5c should be addressed as Figure 5b and Figure 5c.
Figure 5a: It is hard to identify the monthly average discharge during the post-dam period. Please improve the figure.