GENERAL COMMENTS

The manuscript titled “A Revised Temperature-Dependent Remineralization Scheme for the Community Earth System Model (v1.2.2)” presents a comprehensive evaluation of a new temperature-dependent remineralization scheme and its impacts on the global spatial patterns of POC transfer efficiency and nutrient distributions. The authors demonstrate that the new scheme improves the latitudinal contrasts of transfer efficiency and phosphate levels across ocean basins. By validating against in-situ observational profiles and comparing the previous and revised versions of CESM, the study successfully enhances model performance and provides valuable implications for future projections of ocean biogeochemistry. Furthermore, the authors suggest that this new scheme can be applied to future warming scenarios to yield more accurate projections that were previously uncertain.

While the manuscript provides a detailed description of the methodology and results of the temperature-dependent schemes, several key issues should be addressed before publication. In particular, the study should engage more thoroughly with previous research on remineralization schemes, clearly articulate its central research question, and emphasize the novel contributions and distinctions from prior studies to strengthen its scientific impact.

MAJOR COMMENT

1. First, the authors should discuss the previously developed remineralization schemes proposed by other ocean biogeochemical modeling groups. For instance, the IPSL, GFDL, and CSIRO groups implemented temperature-dependent remineralization schemes in their models more than a decade ago (Oke et al., 2013; Aumont et al., 2015; Laufkötter et al., 2017; Stock et al., 2020), as also noted in the Methods section of this manuscript. Therefore, the temperature-dependent remineralization scheme presented here cannot be considered entirely new. However, the manuscript lacks a proper introduction and discussion of these earlier schemes. The authors should describe these prior approaches, clearly state how their implementation differs from previous models, and highlight the specific novelty and advancement of this work relative to earlier studies. Implementation of this scheme within CESM alone would not be sufficient to justify publication in Geoscientific Model Development (GMD) without a more explicit demonstration of its scientific innovation and contribution.

Additionally, despite the detailed description of the new remineralization formula, it remains difficult to directly compare it with the previous remineralization formulation in BEC. It appears that the primary difference lies in the inclusion of k_temp,POC within the decay_POC term. I recommend that the authors clearly describe the previous remineralization scheme and specify what has been improved in the new formulation. For instance, providing a table or schematic that illustrates the previous scheme in black lines and the revised scheme in red lines would greatly help readers visualize and compare the changes. Including other temperature-dependent schemes from different models in the same comparison table would make this section even more informative.

2. The study by Rodgers et al. (2024, Nature) should be cited to highlight the importance of temperature-dependent remineralization schemes for projecting future changes in primary production under global warming. In that study, the largest contrast in projected production among models arose between the temperature-dependent IPSL model and the temperature-independent CESM model. Therefore, incorporating the authors’ new temperature-dependent scheme could potentially alter the direction of projected production trends. Although the CMIP6 simulations are based on CESM2 rather than CESM1, it is noteworthy that CESM1-BEC also lacked a temperature-dependent remineralization scheme. Hence, it would be valuable to examine how the inclusion of this scheme affects future projections, as mentioned in the manuscript. Beyond discussion, I strongly recommend that the authors include at least one future projection experiment-such as an SSP5-8.5 scenario or a simple 1pctCO2 scenario simulations-to compare with previous studies and to provide a more concrete implication of the temperature-dependent scheme.

3. The figures throughout the manuscript require further modification and improvement. For example, consistency should be ensured between filled and open circles representing the PI and Tdep experiments in Figures 9 and 11. In Figure 9, PI is shown with open markers and Tdep with filled ones, whereas in Figure 11, the labeling appears reversed—PI as open and Tdep as filled markers. Consequently, it is unclear whether the authors intended to show that the new scheme improves RMSE and R² for upper-ocean phosphate. However, Figure 9 appears to suggest the opposite trend, with R² decreasing and RMSE increasing from PI to Tdep experiments. If this discrepancy results from simulation errors or mislabeling, it should be clarified in the text or figure captions.

Additionally, more detailed information should be provided in the Methods section. The manuscript states that the pre-industrial control simulation was used for validation against observations and for comparison between the two remineralization schemes, but details are missing. Specifically, the authors should indicate the total simulation length, the integration period used in the figures. If fully coupled simulations were used, equilibrium may not have been reached, leading to year-to-year variability in climatological fields. In such cases, uncertainty ranges—such as those shown in Figure 6—should be included. The authors report variations in transfer efficiency and latitude-dependent changes, but it remains unclear whether these differences reflect genuine mechanistic responses to the new scheme or simply internal climate variability. Therefore, an estimation of uncertainty, for example using interannual standard deviations, is strongly recommended to clarify these features.

References

Oke, P. R., Griffin, D. A., Schiller, A., Matear, R. J., Fiedler, R., Mansbridge, J., Lenton, A., Cahill, M., Chamberlain, M. A., and Ridgway, K.: Evaluation of a near-global eddy-resolving ocean model, Geosci. Model Dev., 6, 591–615 (2013)

Aumont, O., Ethe, C., Tagliabue, A., Bopp, L. & Gehlen, M. PISCES-v2: an ocean biogeochemical model for carbon and ecosystem studies. Geosci. Model Dev. 8, 2465–2513 (2015).

Laufkötter , C., John ,J.G., Stock, C.A., & Dunne, J.P. Temperature and oxygen dependence of the remineralization of organic matter. Global Biogeochemical Cycles,31(7),1038–1050. (2017)

Stock, C. A., Dunne, J. P., Fan, S., Ginoux, P., John, J., Krasting, J. P., et al. Ocean biogeochemistry in GFDL's Earth System Model 4.1 and its response to increasing atmospheric CO2. Journal of Advances in Modeling Earth Systems, 12(10), e2019MS002043  (2020)

Rodgers, K.B., Aumont, O., Toyama, K. et al. Low-latitude mesopelagic nutrient recycling controls productivity and export. Nature 632, 802–807 (2024)

The manuscript by Brabson et al. describes the implementation of a temperature-dependence of the remineralisation of particulate organic carbon in the CESM1 model. The authors conclude that the new temperature dependence results in an improvement of the transfer efficiency in different temperature regimes of the ocean. The study is interesting and overall clearly presented, with some details missing (see comments below).  I recommend it for publication after the following points are addressed:

• I recommend to add a description of the actual values of newly introduced parameters, e.g. the tuning result, for the Tdep parameters (e.g. in a table), and ideally also show a plot of the temperature influence on the remineralisation rate.
• Methods: More information on the simulations performed, including information on the tuning process, would be informative in this section.
• l 179ff: it would be good to list the number of data points used for comparison somewhere, along with an exact computation of the error metrics used.
• l. 5: The figure size should be enlarged to enhance readability of the figure inlet.
• l. 250: “negative biases of 0 to 1.0” – please provide the unit.
• l. 255: Why does the Indian and Arctic Ocean perform almost identical with respect to the phosphate concentration, when they have the highest differences in the basin-scale efficiency (Fig. 6)? A sentence addressing this would be helpful.
• Also l 255.: I don’t agree with the statement that Fig. 9 shows a decrease in cRMSE and an increase in R2 for most of the simulations. If the filled symbols indicate indeed the PI simulations, they are all higher for R2 and lower for cRMSE, except for the Indian Ocean and the Arctic. Please doublecheck the symbol description in Fig. 9 (also in comparison to Fig. 11, see comment below), and if it is correct, adjust the statement in l.255 accordingly. (Same for supplementary figures). For now, I assume this is a plotting mistake, otherwise, some of the conclusions need to be revised.
• Fig. 8: The model performance with respect to the WOA2023 phosphate concentration looks slightly worse in the Tdep simulations in the low latitudes when comparing the PI simulations. This should be addressed in the discussion – why is that the case? Could other processes or controlling factors be missing? Have other parameter values overcompensated the temperature effect before, and require adjusting now that the temperature dependence is included?
• Figs. 11 and 9: For consistency, I would advise to use filled symbols for either the PI or the Tdep simulation, and not change between the meaning in different figures. At the moment, Fig. 11 has filled symbols for Tdep, and Fig. 9 has filled symbols for PI.
• I could not access the data sets provided in the accompanying link for the review process, perhaps due to an embargo before publication. I can therefore not make any statement on the supplementary datasets.