Four decades of full-depth profiles reveal layer-resolved drivers of reservoir thermal regimes and event-scale hypolimnetic warming

Mi, Chenxi; Gai, Bo; Kong, Xiangzhen; Jiang, Yuzhe; Chan, Chun Ngai; Rinke, Karsten

doi:10.5194/egusphere-2025-6442

Preprints

https://doi.org/10.5194/egusphere-2025-6442

Preprints

20 Feb 2026

| 20 Feb 2026

Four decades of full-depth profiles reveal layer-resolved drivers of reservoir thermal regimes and event-scale hypolimnetic warming

Chenxi Mi, Bo Gai, Xiangzhen Kong, Yuzhe Jiang, Chun Ngai Chan, and Karsten Rinke

Abstract. Thermal structure shapes ecological dynamics in lakes and reservoirs. Yet full-profile temperature records over multi-decades remain scarce, constraining mechanistic understanding of depth-resolved thermal changes and subseasonal extremes (e.g., surface heat waves and late-season hypolimnetic warming). In this study, we focused on Rappbode Reservoir—Germany’s largest drinking-water reservoir—and compiled four decades of high-resolution, full-depth temperature profiles with concurrent hydro-meteorological records that are rarely available for stratified systems. Building on these data, we developed a novel two-step analytical framework that integrates long-term monitoring and process-based modelling to yield a high-resolution, internally consistent dataset of spatiotemporal temperature dynamics. We then applied interpretable machine learning to quantify dominant external controls on depth-specific stratification dynamics and determine causal mechanisms governing late-stratification hypolimnetic warming. Our results suggested that influence of external drivers on the thermodynamic structure varied markedly with depth and stratification phase: stratification-strength metrics governed by atmospheric heat fluxes (i.e., surface temperature, vertical temperature difference, Schmidt stability) were controlled mainly by 30-day antecedent shortwave radiation and air temperature. For hypolimnetic temperatures and mixed-layer depth, outflow discharge turned out to be the primary driver during late stratification. Further analysis indicated that episodic hypolimnetic warming up to 10 °C in four specific years was mainly triggered by intensified deep withdrawals that weakened the density gradient and shortened the compensatory-flow pathway. The dual-perspective framework developed here—integrating process-based and machine-learning approaches—is broadly transferable for analyzing ecological processes and supporting evidence-based management in stratified waters.

Received: 23 Dec 2025 – Discussion started: 20 Feb 2026

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.

Download & links

Preprint (PDF, 2792 KB)

Supplement (740 KB)

Download & links

Chenxi Mi, Bo Gai, Xiangzhen Kong, Yuzhe Jiang, Chun Ngai Chan, and Karsten Rinke

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-6442', Salim Heddam, 17 Mar 2026
The present study applies two types of modelling approaches namely the CE-QUAL-W2 numerical model and a machine learning model based on XGBoost to predict the thermal structure of a reservoir. The machine learning model is trained using data generated from the CE-QUAL-W2 simulations. The modelling framework is clearly structured, with the ML model used to predict temperature-related variables at several depths, including epilimnetic (5 m), metalimnetic (15 m), hypolimnetic (30 m), and bottom (50 m) temperatures, as well as key thermal structure indicators such as Schmidt stability, bottom-to-surface temperature difference, and mixed layer depth. Based on certain assumptions, the authors restricted the analysis to the May-October period, which corresponds to the stratified season of the reservoir. A set of meteorological variables was used as predictive features, and 30-day moving average predictors were introduced to represent the cumulative influence of atmospheric and hydraulic forcing on thermal conditions across different depths.

Although the topic of the manuscript is relevant and demonstrates a satisfactory degree of originality, the overall organization of the paper particularly the Results section does not yet meet the scientific standards expected for publication in a high-quality journal. Substantial improvements are required before the manuscript can be considered for publication. In its current form, the study requires major revision. The following issues should be addressed carefully by the authors:

Lack of baseline comparison. It is neither appropriate nor convincing to present a machine learning prediction framework without a solid baseline for comparison. The present study relies essentially on a single ML model, which makes the analysis incomplete and weak from a methodological standpoint. Additional models (e.g., classical regression models or alternative ML approaches) should be included to provide a meaningful benchmark.

Absence of comprehensive result tables. The manuscript does not contain adequate tables summarizing the performance of the models. For a study focused on machine learning applications, detailed quantitative comparisons presented in well-structured tables are essential.

Insufficient description of the dataset. A much clearer presentation of the dataset used for model development is required. This should include:

Detailed description of input features and output variables

Descriptive statistical analysis

Explanation of the data splitting strategy (training and testing datasets)

Distributional comparisons between training and testing sets using Kolmogorov-Smirnov tests (empirical CDF comparison)

Jensen–Shannon distance comparisons between training and testing sets

Energy distance comparisons to further evaluate distributional consistency.

Hyperparameter specification. The hyperparameters used in the ML models should be clearly reported and justified to ensure reproducibility.

Clear separation of training and testing results. Model performance should be presented distinctly for both training and testing datasets, using appropriate evaluation metrics and visualizations.

Need for a methodological flowchart. A clear flowchart summarizing the overall framework of the study from data preparation to model development, validation, and interpretation should be provided.

Incomplete interpretability analysis. Since the authors devoted a significant portion of the manuscript to SHAP-based analysis (SHAP), both global and local interpretability analyses should be conducted. Currently, the local interpretability component is missing.

Additional explainability analyses. To improve the understanding of feature importance and interactions, the study should incorporate complementary explainability techniques, including Permutation Feature Importance (PFI) for global ranking, Partial Dependence Plots (PDP) for marginal effects, and Individual Conditional Expectation (ICE) plots to capture response heterogeneity. Including these analyses would significantly strengthen the scientific contribution of the work.

Statistical comparison of models. Once a robust baseline comparison is established among several models, the authors should perform appropriate statistical tests such as the Kruskal-Walli’s test and the Diebold-Mariano test to assess whether differences in predictive performance between models are statistically significant.

In summary, while the research idea is interesting and relevant, the manuscript requires substantial methodological, analytical, and presentation improvements before it can be considered suitable for publication. A thorough revision addressing the points outlined above is strongly recommended.
Citation: https://doi.org/10.5194/egusphere-2025-6442-RC1
- CC1: 'Reply on RC1', Chenxi Mi, 17 Mar 2026
  
  We highly appreciate the reviewer’s important comments regarding benchmarking, dataset description, result presentation, and reproducibility. We are currently preparing a comprehensive point-by-point response and corresponding manuscript revisions, and we will address these issues in detail in a structured reply together with the remaining comments.
  Chenxi
  
  Citation: https://doi.org/10.5194/egusphere-2025-6442-CC1
RC2:
'Comment on egusphere-2025-6442', Anonymous Referee #2, 18 Mar 2026
Mi and colleagues used a long data record of thermal profiles from Rappbode reservoir to fit a two-dimensional hydrodynamic model, and then used a machine learning to link patterns in temperature and stratification dynamics to potential causes. The study is well-written, clear, and the methodology is reasonably novel and could be applied outside the presented study case. The study suits the audience well and could eventually be published, but I have two major issues that need to be addressed first. I list them below, including some additional minor comments further down.
My first main comment relates to the aim of coupling the machine learning (ML) model XGBoost to the process-based (PB) model CE-QUAL-W2. The ML model is trained on the PB model and is employed specifically to investigate external drivers and causal mechanisms (e.g. L. 32-33). Using data-driven approaches to investigate causality can absolutely be done, but in the present paper, there is no evidence presented on the accuracy of this, although the authors do try to link the outputs from the ML model to processes in the Discussion. Moreover, this could be tested even better than in usual cases, because the PB model that the ML model tries to emulate, has the processes included by design. The authors rightfully argue that the ML-approach primarily gains upon PB scenario runs in terms of time (L. 465-466), but because the novelty of the approach, I argue that the conclusions from the ML model should be validated using PB scenarios first. Some of the causalities highlighted by the ML model (e.g. increasing importance of shortwave radiation with depth; importance of withdrawal volume for end-of-stratification deep-water temperature increase) can be explored with the PB model as well. This should be done in order to build trust in the accuracy of the presented method. If the conclusions between the two methods deviate, the underlying reason should be discussed. The current method has a strong risk of mixing correlation and causality (for instance in telling apart the influence of air temperature and shortwave radiation), and this risk should be mitigated by validation using the PB model. For example, the increasing importance of shortwave radiation with depth is surprising, and this validation could shed light on the reliability. This validation exercise could be presented in the supplementary material and referred to in the main text.
My second main comment relates to the lack of information on the calibration and validation, primarily of CE-QUAL-W2. What periods were used for training and validation, what parameters were adjusted, what were the calibration targets, what method was used? The terms “site-validated” (L. 124) and “site-calibrated CE-QUAL-W2 model” (L. 457) suggest that this was done before, but it is not clearly indicated where this information can be found. If a model from another publication is used, this should be clearly stated in the data availability statement.
Minor comments:
Title: “event-scale hypolimnetic warming” is not clear. Suggest to change to “episodic hypolimnetic warming” as you did in the abstract (or another term that you find more suitable).

L. 37-38: change to “by the 30-day antecedent moving average of shortwave radiation and air temperature” (or similar)

L. 40: change to ”episodic hypolimnetic warming by up to 10 °C…”

L. 42: use of “compensatory-flow pathway” not clear in this context

L. 56: I could not find information on the effect of vertical temperature structure on temperature-sensitive organisms in Carr et al., although some of the references in there might have looked at this. Please assess if this citation is appropriate or cite the original source if possible.

L. 72-73: The GLTC initiative is mentioned but not used further in the manuscript. Suggest to remove this (but keep the Sharma et al. reference and rest of the sentence).

L. 97-100: I think that “temperate” is a too broad term here to say that hypolimnetic temperatures are around 4-6 degrees. Boehrer & Schultze also do not support the generality of this rule for the temperate climate zone. Suggest to change to lakes that cool to the temperature of maximum density.

L. 108-110: Bouffard et al. (2013) does not seem to provide support for the general statement that biological and chemical reaction rates roughly double per 10 °C. Please find a more appropriate reference or change the statement to be more representative of the findings in that paper.

L. 112 (and elsewhere, such as L. 375): General editing comment that there should be spaces around the dash

L. 147-150: What data did you use for the depth-varying outflow discharge? Was this based on data from the reservoir operator?

L. 185: “an monitoring buoy” -> “a monitoring buoy”

L. 262-266 (and links to comment on L. 147-150): In this part you show an impressive R2 value for water level (0.99), but without knowing more about the data sources (see previous comment on L. 147-150) and optional calibration/validation (e.g. added or scaled inflows/outflows to better fit the water level) (see 2^nd main comment), it is difficult to judge the extent to which this builds trust in the model.

L. 276-279: Add something like “…at increasing depth” at the end of the sentence

L. 292: It is very important here that XGBoost was compared to W2, NOT to observations. The term “error” might be a confusing concept here, so I would underline this by explicitly saying “the RMSE between W2 and XGBoost”

L. 301: I wouldn’t describe the mixed depth as a “tight clustering along the 1:1 line”. It is better to quantify this by showing the R2 value (both in the figure and the text).

L. 311: I think “decreases”, “reduces” or “diminishes” are clearer terms than “attenuates” in this context.

L. 367: “combinng” -> “combining”

L. 371: “bridges an inter-decadal gap” could suggest a gap-filling exercise. I suggest to change the wording.

L. 376-377: the part “and the pan-European synthesis of temperate lakes” in not integrated with the rest of the sentence; please rewrite

L. 383: “enhances” -> “enhance”

L. 400: According to the equation and the provided units, shouldn’t the unit of thermal inertia be time per Kelvin, instead of only time? Currently the units do not check out (though I guess a ∆T of 1 K is assumed). I did a quick scan of Imberger & Patterson but could not find this formula; please reply with the equation number in Imberger & Patterson or alternatively how this equation and the units were derived from the reference.

L. 424-427: Can you please clarify how increased vulnerability to mixing events relate to the importance of deep-layer withdrawals?

L. 427: “event” -> “events”

L. 428-435: I missed the argument that a significantly lowered hypolimnetic volume if the mixed layer deepens. Water withdrawn from a smaller hypolimnion will more quickly lower the thermocline. This could be quantified using hypsographic information.

Figure 8: the x-axes are not aligned
Citation: https://doi.org/10.5194/egusphere-2025-6442-RC2
- CC2: 'Reply on RC2', Chenxi Mi, 19 Mar 2026
  
  We sincerely thank the referee for the careful and constructive assessment of our manuscript. We greatly appreciate the positive evaluation of the manuscript’s clarity, methodological novelty, and potential relevance to the HESS readership. We are also grateful for the two major comments, which highlight important aspects that require further clarification and strengthening.
  In particular, we acknowledge the reviewer’s concern regarding the interpretation of the XGBoost results in relation to external drivers and underlying mechanisms. We agree that additional analysis is needed to better demonstrate the extent to which the machine-learning-based interpretations are supported by the process-based model framework. We also appreciate the reviewer’s request for a more complete and transparent description of the calibration and validation of the CE-QUAL-W2 model, including the calibration/validation periods, target variables, adjusted parameters, and evaluation approach.
  We are carefully considering these points and will address them thoroughly in a revised version of the manuscript, together with the reviewer’s many helpful minor comments on terminology, wording, references, and figure presentation. A detailed point-by-point response will be provided during revision.
  We thank the referee again for the thoughtful and helpful review.
  
  Citation: https://doi.org/10.5194/egusphere-2025-6442-CC2
RC3:
'Comment on egusphere-2025-6442', Anonymous Referee #3, 19 Mar 2026
Greetings. I have revised the manuscript entitled ‘Four decades of full-depth profiles reveal layer-resolved drivers of reservoir thermal regimes and event-scale hypolimnetic warming ’. The paper deals with Germany’s largest drinking-water reservoir. By comparing two modelling approaches: the CE-QUAL-W2 numerical model and a machine learning XGBoost-based model, the goal is to predict the thermal structure of a reservoir, assessing the performance of each method. The paper is suitable for publication in the HESS journal, whereas it presents some lacks to be fixed before going further down the publication way. These can be listed as follows:
The paper only presents one type of Machine Learning approach and one kind of model. There is a need for at least a small paragraph that frames, in a general way, machine learning approaches. Moreover, there is no presence of reference to Genetic Algorithms and Metaheuristics ones, that have been proved to outperform other ML methods (see Rajwar et al., 2023; Schiavo and Pedretti, 2026). This does not mean that these approaches should be implemented, nor that the paper should compare these results with the offered ones, but at least citing their existence and weight in the literature framework, and clearly explaining why the XGBoost method has been preferred to Genetic algorithms.

Some recent literature works assessed how, for some earth sciences applications, ‘usual’ geostatistical methods (e.g. kriging or co-kriging-based ones) still perform better than Neural Network algorithms or XGBoost methods (Brckovic et al., 2025), or at least they are good enough at predicting the target variable that they should be involved in ML algorithms (Grey et al., 2025) to achieve reliable predictions. I suggest discussing this point further.

I am not sure that a strong enough system conceptualization is given. Thus, I am a bit reluctant about the goodness of the results without grounding them to physically-based description, meshing, and imposition of boundary conditions. Maybe the Authors can clarify this point, underlining the (i) assumptions they have made, (ii) giving more info about the geological structure, and (iii) providing hydrogeological or geological sections to support their claims. Indeed, a max depth of 80 m (Figure 2) requires strong geological proof of concepts.

Temperature fluctuations seem extremely noisy and subject to historical trends. Why have these not been performed and offered in the results?

I think that the revision could proceed further only once these observation have been clarified.
Best regards.
Suggested references:
Brcković, A., Malvić, T., Orešković, J., & Kapuralić, J. (2025). Comparison of Neural Network, Ordinary Kriging, and Inverse Distance Weighting Algorithms for Seismic and Well-Derived Depth Data: A Case Study in the Bjelovar Subdepression, Croatia. Geosciences, 15(6), 206. https://doi.org/10.3390/geosciences15060206
Schiavo, M., & Pedretti, D. (2026). Genetic and Iterative Metaheuristics-Informed Algorithms for Precision Shallow Groundwater Modeling and Drought Inference. Journal of Geophysical Research: Machine Learning and Computation, 3(1), e2025JH000854. https://doi.org/10.1029/2025JH000854
Rajwar, K., Deep, K., & Das, S. (2023). An exhaustive review of the metaheuristic algorithms for search and optimization: Taxonomy, applications, and open challenges. Artificial Intelligence Review, 56(11), 13187–13257. https://doi.org/10.1007/s10462-023-10470-y
Grey, V., Fletcher, T. D., Smith-Miles, K., Hatt, B. E., & Coleman, R. A. (2025). Harnessing the strengths of machine learning and geostatistics to improve streamflow prediction in ungauged basins; the best of both worlds. Journal of Hydrology, 662, 133936. https://doi.org/10.1016/j.jhydrol.2025.133936
Citation: https://doi.org/10.5194/egusphere-2025-6442-RC3
- CC3: 'Reply on RC3', Chenxi Mi, 19 Mar 2026
  
  We highly appreciated the reviewer for the constructive and detailed comments and for considering the manuscript suitable for publication after revision! We will prepare a point-by-point response and revise the manuscript accordingly. In particular, we will (i) add a brief overview positioning our XGBoost–SHAP approach within the broader landscape of machine-learning and optimization methods, including metaheuristics, and clarify our method choice; (ii) expand the discussion on alternative approaches where relevant and delineate the scope of our attribution objective; (iii) strengthen the system description of the CE-QUAL-W2 setup, including bathymetry, grid design, boundary conditions, and withdrawal representation; and (iv) further quantify and present long-term trends versus event-scale anomalies in the temperature record. We appreciate the reviewer’s suggestions and believe these revisions will substantially improve the clarity and robustness of the study.
  Best Cheers,
  Chenxi
  
  Citation: https://doi.org/10.5194/egusphere-2025-6442-CC3

Chenxi Mi, Bo Gai, Xiangzhen Kong, Yuzhe Jiang, Chun Ngai Chan, and Karsten Rinke

Supplement

https://doi.org/10.5194/egusphere-2025-6442-supplement

Chenxi Mi, Bo Gai, Xiangzhen Kong, Yuzhe Jiang, Chun Ngai Chan, and Karsten Rinke

Viewed

Total article views: 266 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
153	81	32	266	34	13	19

HTML: 153
PDF: 81
XML: 32
Total: 266
Supplement: 34
BibTeX: 13
EndNote: 19

Views and downloads (calculated since 20 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	56	34	9	99
Mar 2026	92	43	23	158
Apr 2026	5	4	0	9

Cumulative views and downloads (calculated since 20 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	56	34	9	99
Mar 2026	92	43	23	158
Apr 2026	5	4	0	9

Viewed (geographical distribution)

Total article views: 298 (including HTML, PDF, and XML) Thereof 298 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 05 Apr 2026

Short summary

This study examines temperature changes in Rappbode Reservoir, Germany’s largest drinking-water reservoir, over four decades. By analyzing detailed temperature data and weather records, the research shows how factors like sunlight and air temperature impact water temperature at different depths. It also finds that late-season warming in deep waters is mainly caused by water withdrawals. These findings help improve understanding of lake ecosystems and can guide better water resource management.


Total:	0
HTML:	0
PDF:	0
XML:	0