Dear Editor,

I have revised the manuscript entitled “Quantifying the impact of groundwater fauna and temperature on the ecosystem service of microbial carbon degradation” by Schmidt, Rütz and Marxsen. I found the work very interesting and timely. The study plausibly demonstrates the important ecosystem service provided by microbes and groundwater-dependent  fauna, both in terms of carbon degradation and in preventing brownification of groundwater used for human consumption. I recommend acceptance after the authors have considered the points listed below.

The introduction is well written and supported by robust and up-to-date literature. The rationale is clearly presented and the hypotheses are explicitly stated at the end of the section. Figure S1 is well constructed and informative.

Although I do not consider myself fully qualified to evaluate every aspect of the modelling framework, I carefully followed the methodological steps as presented. The calculations appear internally consistent and the logical structure is coherent. The authors also provide a transparent discussion of the main assumptions and limitations of the model in the final section of the manuscript, which I found very appropriate and commendable. However, it might be helpful to frame the model more explicitly, already in the Methods or early Discussion, as a strongly parameterised proof-of-principle approach. Given the number of assumptions and parameter choices, the results are best interpreted in qualitative and comparative terms among scenarios rather than as precise quantitative predictions. Making this perspective explicit earlier in the manuscript would further strengthen its clarity and robustness.

The results are overall convincing and internally coherent. The model reproduces the measured time series reasonably well (Fig. 1b–e), and the agreement between simulated and observed dynamics supports its use for exploring scenarios. Figure 2 clearly shows that differences among scenarios are driven mainly by the presence or absence of fauna rather than by moderate temperature increases, since the “reference” and “+1.5 °C” lines largely overlap across zones. The strong divergence of the “no fauna” scenarios in the “M” and “P” zones convincingly supports the conclusion that faunal grazing substantially influences BOC dynamics and therefore the ecosystem service of carbon degradation. Some statements, however, would benefit from slightly more cautious wording (see specific comments below).

I found the discussion particularly interesting where the authors provide a quantitative estimate of the loss of ecosystem services (in terms of BOC degradation) under scenarios combining fauna absence and increased temperature due to climate change. The estimates are convincing and I appreciate the effort, because although it is widely acknowledged that subterranean ecosystems provide important services (see, for instance, Mammola et al., 2026. Subterranean environments contribute to three-quarters of classified ecosystem services. Biol Rev Camb Philos Soc. 2026 Feb 10. doi: 10.1002/brv.70137), their quantitative evaluation is still limited.

In the final part of the manuscript, the authors openly discuss the weaknesses of the model and the study. This transparency is commendable and provides a solid basis for future refinement as new data become available. I found this attempt to model groundwater ecosystem services both courageous and valuable.

Specific comments

Abstract

No major comments.

Introduction

Line 24. I suggest expanding the references supporting the statement “Groundwater is no exception (Avramov et al., 2010)” by including the most recent work by Mammola et al. 2026 doi: 10.1002/brv.70137

Lines 42–44. The term “producers” may not be immediately clear. Consider revising to: “…producers (i.e. the basal microbial community including both autotrophs and heterotrophs sustained by imported organic matter).”

Line 44. Perhaps: “When even the producer biomass becomes substantial…”. In addition, regarding the term “grazers,” I would suggest using a more general term, since groundwater fauna can include both grazers and deposit feeders that ingest sediments and digest the microbial biofilm attached to them.

Lines 53–54. This sentence seems slightly disconnected from the surrounding text. It may benefit from relocation to improve the logical flow.

Line 111. Please consider adding: Vaccarelli et al. (2023, One Earth, 6, 1510–1522), which strongly supports your statement.

Line 115. You may consider citing: Iannella et al. (2020, Scientific Reports, 10, 19043) to support this point.

Line 128. Possibly: “in faunal composition (Di Lorenzo et al., 2025)?”

Results

Lines 236–238. The statement that event-based recharge “contributed little” seems plausible, but this is not directly demonstrable from the figure alone. It would be safer to write “appeared to contribute little under the model assumptions.”

Lines 255–257. The interpretation that summer temperatures became “lethal” at +1.5 °C is mechanistic and not directly shown in the plots. It would be safer to state that fauna declined seasonally under elevated temperature, without explicitly attributing lethality unless this is independently supported.

Discussion

Lines 353–357. Please clarify this interpretation. The text states that early sudden dips in microbial dry mass occurred in the “A” and “R” zones and might reflect feeding by higher-than-average fauna biomass. However, in Fig. 1d the most pronounced temporary reductions to near-zero microbial dry mass appear in the “M” and “P” zones, and these coincide with peaks in fauna dry mass in Fig. 1e (as mentioned in lines 369–370). Could this be a mislabelling of zones, or am I overlooking a specific pattern in “A” and “R”?

Lines 389–397. The interpretation that higher microbial dry mass in the “no fauna” scenarios is associated with reduced BOC degradation and attributed to a lack of rejuvenating grazing could be further strengthened. An additional explanation might relate to microbial community structure and activity. For example, Di Lorenzo et al. (2025, Biogeosciences, 22, 1237–1256) reported a correlation between groundwater-obligate crustacean abundance and low-nucleic-acid (LNA) bacterial cells, suggesting selective feeding interactions. Since LNA cells are generally less metabolically active, the absence of fauna could allow the accumulation of microbial biomass that is not functionally active in terms of carbon uptake. In this perspective, higher microbial dry mass but lower BOC degradation in the “no fauna” scenarios may reflect a shift towards less active microbial fractions rather than more effective processing. Even if the current model does not distinguish active and inactive pools, mentioning this conceptual perspective could provide a biologically grounded explanation.

Line 533. Please compare the BOC values here with those reported in line 492 and line 18 for consistency (6.6 time vs. 660-fold).

Supplementary File

Line 78. Please clarify what “L” means in “1419.5 mm year⁻¹, i.e. L m⁻² year⁻¹.”

Line 108. Should this read “10^6 cells mL⁻¹”? Please check that “cells” is not missing and verify the exponent (−6). This last point is important because I think it is 10^6 cell/mL

Lines 109–111. It is not clear how dry mass per cell was derived (pg cell⁻¹? fg cell⁻¹?). This is not necessarily incorrect, but an important methodological step seems to be missing.

Line 111. The average prokaryotic biomass of ~200 μg L⁻¹ should be explicitly stated as dry mass to avoid confusion with carbon or COD.

Lines 112–115. Please check this sentence for clarity. A closing parenthesis may be missing.

Line 114. “Converted” may be more appropriate than “translated.”

Lines 120–122. The reaction appears correct, but I wonder whether COD (acetate) should be expressed as 2 mol O₂ per mol of acetate, rather than “2 mol L⁻¹” in absolute terms. Should this not be linked to acetate concentration?

Lines 166–169. The adopted temperature scaling can be justified as a pragmatic modelling choice. However, it should be stated more clearly that it relies on a linear regression derived from optimal growth rates of a limited number of bacterial species over a warm temperature range (14–44 °C), and that applying it to typical groundwater temperatures, often below this range, represents an extrapolation beyond the original domain of validity. Explicitly stating this assumption would improve transparency.

Line 253. I believe you are referring to S5 rather than S3.

The submitted manuscript aims at quantifying the change in ecosystem services, particularly carbon degradation, due to loss of groundwater fauna and increased groundwater temperature. While the topic is indeed very interesting and important, the presented study has multiple severe issues related to the soundness and robustness of the chosen approach, regarding amongst others the vast number of underlying assumptions, which are in my opinion not sufficiently justified (see comments below).

I therefore recommend to reject the current manuscript, although I think it could be resubmitted as a local case study on the relationship between groundwater fauna, microbiology and groundwater conditions under very specific hydrogeological conditions, with a more critical discussion of underlying modelling assumptions and limitations.

General comments:

Soundness of scientific approach:

I found that some of the information in the introduction is presented in an incomplete and biased manner. For example, line 100 states that Brielmann et al. (2009) found decreasing faunal diversity with increasing temperature. However, this study also found that faunal abundancy is not influenced by groundwater temperature. This finding is contradictive to the working hypothesis and modelling assumption, so please also comment on it.

Just as critical in my opinion is the statement on modelling in line 120 “Since it is hard to derive the relevant data on the required resolution, […], one way to approach answers […] are models.” Having representative, reliable and sufficient data is a hard prerequisite for any modelling task, otherwise one will end up with “garbage in, garbage out”.

The same goes for the underlying modelling assumptions and simplifications, of with there are in my opinion too many in the presented work. While I did not have the time to go all through all of them in detail, I found the one of linear mortality of groundwater fauna with increasing groundwater temperature insufficiently justified. Here again, findings from previous studies (e.g. Di Lorenzo et al., 2025) that indicate that fauna can exist at “higher” groundwater temperatures are actively dismissed. Instead the modelling is based on results from one single study (Brielmann et al., 2011) that was conducted under controlled conditions in a laboratory on just two groundwater species. It is neither clear how far this experiment reflects in-situ aquifer conditions, nor if those two species were actually found at the study site of the manuscript.

On the same note: there are several recent studies on invertebrate groundwater fauna, which either contradict or relativize the findings presented in the manuscript (e.g. Koch et al., 2021, 2024a, 2024b; Noethen et al., 2024; Becher et al., 2022), and are currently not cited in the manuscript (or the SI).

Robustness of the approach:

Throughout the entire manuscript a critical reflection or discussion on the mentioned underlying modelling assumptions is missing. In my opinion, it is not sufficient to simply state that models reflect the field data “satisfactorily”, “well” or “appropriately” (e.g. lines 13, 232, 329, 387). Error metrics are shown in Fig. 2, but not discussed in the main text. Likewise, the number of data points (“N”) is shown, but also not critically reflected upon. In particular, for groundwater fauna, the number of data points is very low (between 1 and 4). This raises severe concerns about the robustness of the approach.

The extrapolation to global scale (section 4.8) is absolutely absurd from a hydrogeological perspective. You cannot extrapolate findings from a few wells in a local floodplain aquifer to the entirety of global groundwater, whilst ignoring basic fundamental needs of invertebrates for e.g. pore space and dissolved oxygen. Even within Germany, a large portion of shallow groundwater does not contain enough dissolved oxygen to sustain invertebrate life (see e.g. Kunkel et al. 2004), correspondingly many sampled wells were simply not inhabited by groundwater fauna (e.g. Koch et al., 2024a).

Also, the mentioned implications for drinking water supply and treatment are too general and vague in my opinion, and again they are mostly not referenced. Not all groundwater is drinking water, and vice-versa. Before highlighting such aspects in the conclusions, robust background information on this have to be shown and referenced in the manuscript.

Clarity:

The manuscript is built around just two figures: one showing the fit of the hybrid model to the measured data, another showing the scenario analysis. While these figures are presented in a clear way, there is almost no information about the actual site and the used data in the main manuscript.

The introduction lacks a clear structure leading to the hypotheses, and contains several aspects such as drinking water pipes, piscivorous fish and bark beetles, which seem unrelated to the actual topic.

The results and discussion contain many references to the different zones R, M, P, and A in the study area, which are neither explained, nor justified in the main manuscript. Also, I found it impossible to follow the description and discussion of the results without cross-reading the lengthy (29 pages) SI on a second screen.

References:

• Becher, C. Englisch, C. Griebler, P. Bayer, Groundwater fauna downtown – Drivers, impacts and implications for subsurface ecosystems in urban areas, J. Contam. Hydrol. 248 (2022) 104021.
• Brielmann, C. Griebler, S.I. Schmidt, R. Michel, T. Lueders, Effects of thermal energy discharge on shallow groundwater ecosystems, FEMS Microbiology Ecology 68(3) (2009) 242 - 254.
• Brielmann, T. Lueders, K. Schreglmann, F. Ferraro, M. Avramov, V. Hammerl, P. Blum, P. Bayer, C. Griebler, Oberflächennahe Geothermie und ihre potenziellen Auswirkungen auf Grundwasserökosysteme, Grundwasser 16 (2011) 77-91.
• Di Lorenzo, A. Tabilio Di Camillo, S. Iepure, D.M.P. Galassi, N. Mori, T. Simčič, Oxygen Consumption and Carbon Budget in Groundwater-Obligate and Surface-Dwelling Diacyclops Species (Crustacea Copepoda Cyclopoida) Under Temperature Variability, Environments 12(1) (2025) 32.
• Koch, K. Menberg, S. Schweikert, C. Spengler, H.J. Hahn, P. Blum, Groundwater fauna in an urban area – natural or affected?, Hydrol. Earth Syst. Sci. 25(6) (2021) 3053-3070.
• Koch, P. Blum, K. Korbel, K. Menberg, Global overview on groundwater fauna, Ecohydrology 17(1) (2024a) e2607.
• Koch, P. Blum, H. Stein, A. Fuchs, H.J. Hahn, K. Menberg, Temporal shift in groundwater fauna in southwestern Germany, Hydrol. Earth Syst. Sci. 28(22) (2024b) 4927-4946.
• Kunkel, R., Voigt, H.-J., Wendland, F., and Hannappel, S., Die natürliche, ubiquitär überprägte Grundwasserbeschaffenheit in Deutschland, Schriften des Forschungszentrums Jülich Reihe Umwelt 47, Forschungszentrum Jülich GmbH Programmgruppe Systemforschung und Technologische Entwicklung (2004).
• Noethen, J. Becher, K. Menberg, P. Blum, S. Schüppler, E. Metzler, G. Rasch, C. Griebler, P. Bayer, Environmental impact of an anthropogenic groundwater temperature hotspot, Sci. Tot. Environ. 955 (2024) 177153.
• Spengler, Die Auswirkungen von anthropogenen Temperaturerhöhungen auf die Crustaceagemeinschaften im Grundwasser-Versuch einer Prognose zur Klimaerwärmung und lokalen Wärmeeinträgen, Universität Koblenz-Landau, 2017.
• Spengler, C., & Hahn, J., Thermostress: Ökologisch begründete, thermische Schwellenwerte und Bewertungsansätze für das Grundwasser. KW Korrespondenz Wasserwirtschaft, 11(9), 521–525 (2018).