The article is devoted to the urgent challenges of microbially-driven SOC models. The authors created numeric mirobially-driven and depth explicite model of SOC dynamics using R code. They used different scenarios based on isotopic constraints to optimize model parameters, assessed parameter equifinality and checked model sensitivity related to uncertainties of input parameters. The authors revealed that numerical simulations of depth profiles of δ13C are prone to uncertainties connected with data availability, wide range of δ13C values of C3 vegetation, challenges in estimation of δ13C value of root exudates. One more important findings is that despite the including data on the δ13C and/or Δ14C values of SOC to constrain parameter values during calibration did not substantially reduce equifinality of  the most parameters, Δ14C data needs to be incorporated to the model calibration for correct simulation of the turnover time of SOC (models without this data substantially overestimated SOC turnover rate).  To include additional data during the parameter calibration process is one way forward to improve microbially-driven SOC models. However, they advise avoiding overparametrisation which lead to behavioural models.  Defining the optimal structure of soil biogeochemical models and finding a balance between model complexity and available data is an important prerequisite to increase confidence in global projections of the soil carbon - climate feedback. Also, the authors suggests to create and use global databases with data related to SOC cycling to better constrain model parameters.

This article spotlights the biogeochemical modeling challenges and can help to improve simulation of SOC dynamics. But I have several questions concerning modeling methodology and I would appreciate if the authors explain  some of the details. I have also added suggestions to improve the quality of the illustrations in the article and in the Supplementary information file. See attached files with comments.

• Manuscript Title

A microbially-driven and depth-explicit soil organic carbon model constrained by carbon isotopes to reduce equifinality

• Recommendation

This paper presents a novel SOC model (SOIL carb), designed to mitigate equifinality by integrating δ¹³C and ∆¹⁴C values of soil organic carbon (SOC). Calibration solely based on SOC stock data results in imprecise estimations of subsoil organic carbon (OC) residence times. The inclusion of δ¹³C has a minimal effect, whereas the incorporation of ∆¹⁴C  accurately captures the SOC turnover rate but only partially alleviates equifinality for certain parameters. Given that all parameters are susceptible to equifinality, additional data is required to establish reliable constraints. Achieving an optimal balance between model complexity and data availability is crucial for accurately predicting soil carbon-climate feedback mechanisms.

The article's topic selection is significant and demonstrates robust logic and academic rigor. Nevertheless, certain sections necessitate further refinement. The subsequent revision recommendations are outlined below

Major revisions

Introduction

1. Further explanation can be provided on why accurately predicting the reserves and dynamics of Soil Organic Carbon (SOC) is crucial for combating climate change. Additionally, pointing out the problems caused by inaccurate SOC models, it can enhance readers' comprehension of the urgency and significance of this research endeavor.
2. It is suggested that a brief discussion be included at the end of the introduction regarding the potential impacts of this study on soil carbon cycling, climate change prediction, and land management practices, enhancing the practicality and relevance of the research.

Materials and Methods

1. In Sensitivity Analysis: The authors should provide more details on the parameter sensitivity analysis, particularly for those parameters that have the greatest impact on the model's output.

Results

1. At the beginning of each results section paragraph, the key findings of this study can be highlighted using concise and clear language, enabling readers to quickly grasp the main outcomes of the research.
2. The results should present a sensitivity analysis of the parameters that significantly affect the model output, which aids in understanding which parameters are most critical to the model's outcomes.
3. If there are limitations to the results, such as the representativeness of the data or the applicable conditions of the model, they should be clearly stated in the Results section.

Discussion

1. The discussion should be expanded to address the generalizability of the model results, specifically whether the model is applicable to other soil types or environmental conditions, with further explanation provided in the discussion.
2. The relative importance of different mechanisms at different soil depths can be further explored regarding its causes and potential influencing factors.
3. Although the article mentions the importance of accurately simulating the turnover time of SOC for predicting changes in the global carbon cycle, it can further expand the discussion on the specific significance of the research results in practical applications, such as the potential impacts and inspirations on soil management, climate change response strategies, and other aspects.
4. Although the author mentioned that the model does not include the effects of temperature and soil moisture, it is suggested to further discuss the specific impacts of these limitations on the model's predictive ability.

Minor revisions

1. In Figure 3, the "calibrated for C" section is in italic format; please consistent formatting.
2. There are several sentences in the article that are rather cumbersome, such as the following ones:

• The model first calculates fluxes of 12C between pools and subsequently uses the ratio of 12C leaving every pool to the total amount of 12C of the respective pools to calculate how much 13C and 14C leave every pool, based on the respective 13C/12C and 14C/12C values of the pools. The model parameters are thus defined based on the 12C content of every pool.( Line 135 )
• The parameter sets to calculate the conditional and unconditional CDFs were obtained using the Matlab® version of SAFE toolbox (Pianosi et al., 2015), which was also used to post-process the results and calculate the sensitivity of the tested parameters using the Kolmogorov - Smirnov (KS) statistic. ( Line 282 )
• Similarly for the soil, after parameters are optimised using measurements of depth profiles of OC, δ13C and ∆14C of the POC and MAOC pools, simulated depth profiles of these fractions closely reproduce measurements. (Line 307).

The authors present a new mechanistic model (SOILcrab) to assess the potential to decrease parameter equifinality by including isotope data of tow soil fractions as constraints on parameter values during model optimization. They found that adding Δ¹⁴C data as a calibration constraint, can correct simulation of the turnover rate of SOC and only substantially reducing equifinality for the parameter regulating desorption rate of OC from minerals. However, adding δ¹³C data had little effect to improve simulations of the turnover rate of SOC or limit parameter equifinality. These findings are interesting and can improve the predictions of soil carbon dynamics under environmental change scenario.

Major concern:

This model was only applied in a deciduous forest site with different soil profile, it is unclear what’s the performance of this model when it is applied to a large spatial scale.