the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Nov 2025
Implementing belowground controls on nutrient uptake in ELMv2-SPRUCE improves representation of a boreal peatland ecosystem
Abstract. Boreal peatlands store 13–32 % of the global soil carbon (C) stock, a service dependent on plant-mycorrhizal fungi associations. In these nutrient poor systems, ectomycorrhizal and ericoid mycorrhizal fungi supply up to >80 % of the nutrient requirements of their plant hosts, partly with mined nitrogen (N) and phosphorus (P) from soil organic matter that are otherwise inaccessible to plants. Despite the ecological significance, mycorrhizal associations are only represented in a few land surface or ecosystem models. We modify the peatland branch of version 2 of the Energy Exascale Earth System Land Model (ELMv2-SPRUCE) to replace the default photosynthesis-driven inorganic N and P (NP) uptake process with a more realistic representation of the process via three pathways: (1) direct inorganic NP uptake by uncolonized fine roots, (2) indirect inorganic NP acquisition and (3) indirect NP acquisition from organic sources by mycorrhizal roots. We systematically evaluated the performance of the default and modified models with field observations from a whole ecosystem warming and carbon dioxide fertilization experimental site: Spruce and Peatland Responses Under Changing Environment (SPRUCE), in northern Minnesota, USA. The modified model reduces the underestimation of the growth response of shrubs in the default model to warming from 40–80 % to 17–35 % and reduces the overall relative absolute error on C fluxes from 1.61 to 1.54. The improved growth response of shrubs to warming is accompanied by several-fold increase in direct inorganic NP uptake and decrease in fungal colonization rate. The modified model simulates a weaker transition of the ecosystem from C sink to C source under warming due to alleviation of plant nutrient limitation. Equifinality analysis shows the newly added parameters in the modified model can be constrained by the observed C fluxes. Sensitivity analysis shows the newly added parameters have stronger statistical interactions than the preexisting parameters in the default model. Overall, the modified model is an improvement over the default ELMv2-SPRUCE and will be a useful tool for understanding boreal peatland change.
- Preprint
(2606 KB) - Metadata XML
-
Supplement
(2729 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- AC1: 'Correction on Fig. 4', Yaoping Wang, 12 Dec 2025
-
RC1: 'Comment on egusphere-2025-5471', Anonymous Referee #1, 13 Jan 2026
General comments
The manuscript by Wang et al describes model development within a specific peatland branch (SPRUCE) of the land component of the E3SM model, ELM. The manuscript displays a substantial work with both model development, parameter optimization/calibration, sensitivity analysis and analysis of implications for model simulations in comparison to older versions of the model. In total there are four model setups are run, with the new and old model structures as well as with default and optimized parameters.
The topic is relevant and the modelling approach is reasonable from a theoretical perspective, although I am a bit concerned over the amount of new parameters introduced (37 in total). The paper do discuss the problem of equifinality, however, I would like a deeper discussion on constricting the model parameters or parameter ranges, either through observations or theoretical reasoning.
The analysis puts a lot of emphasis on the sensitivity of the different model setups, which is of course a sound approach, however, I feel that the analysis is sometimes overly complicated for a model setup such as this. The focus on sensitivity leads to a lot of relative figures where internal variables are compared to each other, e.g., N acquisition per GPP. While relative values can be informative, absolute values are also important for judging the performance of the model. A simple time-series or scatterplot comparing the different setups against observed data would help. That would also help the interpretation of other metrics such as the normalized values presented in figure 2. I would recommend to move some of the figures of sensitivity, e.g., Figure 7, to the supplementary materials and add a simple figure (time-series or scatterplot) showing absolute values in comparison to observational data.
In general, the paper is well written with an easy structure which is easy to follow. I have reviewed the text and some of the supplementary materials and have some comments that I would like to see resolved. After some revision I think this paper is suitable for publication within GMD.
Specific comments
Section 2.2. This section is generally well written however, I feel like the conceptual idea should be helped by a simple conceptual diagram. Also, I believe that the parameter values should be displayed in the main text and not “hidden” in the supplementary materials. At least the 18 parameter values selected for calibration should be shown along with their default values, calibration ranges, and justification of ranges (if available).
Section 2.5.1. The RAE is described here as a metric for calibrating the model and assess the optimal parameter combination. While I have not come across this metric before, it seems like a suitable metric to me. However, it is also used later as an evaluation metric (e.g., section 3.1.1), which seems like a breach of good calibration/validation practice. It is also unclear to me whether the same data was used for calibration and validation.
Technical comments
L305 “are not overemphasized” is written twice.
Figure 4. The unit description on the y-axis lacks a closing bracket.
Citation: https://doi.org/10.5194/egusphere-2025-5471-RC1 -
RC2: 'Comment on egusphere-2025-5471', Anonymous Referee #2, 01 Feb 2026
This manuscript presents new representations of plant–mycorrhizal nutrient acquisition in ELMv2-SPRUCE and evaluates them against multiple carbon and nutrient cycling observations from the SPRUCE experiment. Testing whether the inclusion of mycorrhizal fungal nutrient uptake improves model performance in predicting peatland responses to environmental change, leveraging the extensive SPRUCE experimental datasets, is timely and important work. I commend the authors for the significant amount of work involved in modifying and evaluating an already complex land surface model. I have one major comment, primarily related to clarifying key modeling assumptions in the main text (currently placed largely in the SI), as well as several minor comments. Overall, I recommend the manuscript for publication once these points are addressed.
Major comment
Methods are too concise relative to the importance of the new process representations. Right now, key assumptions and equations for the new modeling processes are mostly described in the Supplement (SI pp. 8–13), but they are central to interpreting the results. I think the main-text Methods should briefly summarize these model key assumptions and equations, so readers don’t need to go back and forth from main text to SI for key model structure clarification. Examples that seem important enough to mention in the main text include:
- Mycorrhizal colonization is modeled as a function of soil inorganic N availability, rather than plant nutrient limitation status (as a lot of existing ecosystem/land surface models do), and does not include sensitivity to soil inorganic P. This is an important model assumption that needs to be mentioned in the main text, and maybe discussed a bit the implication of this assumption in the discussion.
- Mycorrhizal organic N uptake is restricted to litter pools, excluding SOM pools with fixed stoichiometry. This likely underestimates the magnitude of organic nutrient uptake and may also underestimate the potential feedback whereby extraction of organic nutrients reduces nutrient availability for microbial SOM decomposition. This limitation should be stated and briefly discussed;
- In addition, I have several concerns regarding the formulation in Eqs. S19–S20 describing the effects of litter N pool size on mycorrhizal organic N uptake (i) litter pool size appears to impose only an upper bound on uptake rather than scaling uptake rates directly with pool size like first-order decomposition models do; (ii) the constant value (0.0001) in S19 is not justified; and (iii) organic N extraction rates do not differ among pools (e.g., lignin-associated N vs more labile pools)
- Carbon allocation to nutrient acquisition is capped at 50% of current NPP per timestep. Empirical studies suggest that belowground C allocation often peaks later in the growing season, supported by non-structural carbohydrate accumulation towards the end of the growing season. Thus imposing a cap tied to instantaneous NPP could therefore lead to misplaced mycorrhizal N uptake seasonality. It would be helpful to clarify this assumption in the model description in main text and may briefly discuss how this assumption may affect the results.
This is not a request for new simulations, but rather a request to move key information into the main text and to acknowledge the implications of omitting or simplifying these mechanisms.
Minor comments:
Line 57: consider adding an intro sentence on what plant-mycorrhizal associations do before listing mycorrhizal types, like “mycorrhizal fungi deliver nutrients to plants in exchange for carbon”
Line 94-95: regarding “excess flux mechanism is likely realistic at the microscopic level (Bunn et al., 2024)”, Bunn et al. (2024) do not provide support for surplus or “excess flux” mechanisms being locally regulated at the microscopic level, instead, they emphasize caution against reciprocal exchange interpretations. Surplus C concepts have been discussed primarily as ecosystem- and model-level frameworks (e.g., Prescott et al., 2020 Surplus Carbon Drives Allocation and Plant–Soil Interactions). Clarification or revision of this statement is needed.
Line 283–284: Do the ensemble simulations mentioned here correspond to the ~4000 ensemble members described later (Lines 290–291)? If so, it may improve clarity to introduce and describe these ensemble simulations before referencing them. Additionally, please clarify whether the parameter optimization refers to selecting the best-performing parameter set(s) from this ensemble.
Figure 2: Could the authors clarify whether these dots correspond to modeled values from different years, soil temperatures, and CO₂ treatments all plotted together? As currently presented, it is unclear what quantities are being compared in this figure.
Future directions: Fungal necromass needs to be considered too. As ERM necromass for example is known to have high melanin% and is propoased to be recalcitrant to decomposition and contributed to the large SOC accumulation.
Citation: https://doi.org/10.5194/egusphere-2025-5471-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 289 | 128 | 30 | 447 | 66 | 17 | 18 |
- HTML: 289
- PDF: 128
- XML: 30
- Total: 447
- Supplement: 66
- BibTeX: 17
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Dear reviewers, we apologize for a mistake made on Fig. 4 during the preparation of the draft. We pasted the wrong figure that shows the uptake-to-GPP ratio, but the figure should in fact show total N and P uptake. Please see attached for the correct Fig. 4. The descriptions and caption in the manuscript are correct.