Major comments:

This perspective manuscript highlighted the microbial datasets based on meta-omics technologies, and explored the integration of such datasets into Earth System Models to examine the microbial responses to environmental changes. Some key points and future efforts are discussed in the current version, and this would raise some concerns of ecologists to considering the important effects of microbiomes. The manuscript was written well to enjoy the reading, I have only some open concerns for specific issues.

1. The inclusion of meta-omics based microbial datasets in models was not well discussed. Because some of the biogeochemical features such as pH or organic matters are easy to measure with physical equipment, why it is important to model these parameters with microbial datasets? Actually, sequencing microbiomes will also consume additional times instead of reflecting simultaneous dynamics. Directly observatory may provide more accurate results.
2. Lack of summary of current well-established framework or models by integrating meta-omics microbial datasets into biogeochemical changes. Currently, the manuscript only provided one case study by using G2E framework to show this idea, but there was limited information on how to linking the microbes with the biogeochemical features. Such as the Microbial-Enzyme Decomposition (MEND) model, which can indicate the soil functions based on soil enzyme and microbial GeoChip-based functional genes, already provides a basis for testing hypotheses about microbially mediated biogeochemical processes in response to environmental changes (Wang et al. 2022 https://doi.org/10.1111/gcb.16036 ; Gao et al., 2020 https://doi.org/10.1073/pnas.2002780117 ).
3. Besides, the manuscript discussed some potential work to link biodiversity with ecosystem functions, but lacking detailed methods for this and details microbial traits or paths. For example, machine-learning or AI-based models were used to linking bacterial communities with physico-chemical variables and soil quality (Hermans et al., 2020 https://doi.org/10.1186/s40168-020-00858-1), further another study expanded the models to predict future grassland soil pH changes in response to global warming (Feng et al., 2024 https://doi.org/10.1016/j.oneear.2024.06.002). Also another pioneering work explored the bacterial pH preferences by linking microbial genomic traits to ecosystem functions, i.e., soil pH (Josep et al., 2023 https://doi.org/10.1126/sciadv.adf8998). More discussion about the these possible methods or approaches would help readers to follow in their researches.

The integration of genetic/omics data into Earth System Models (ESMs) is a promising direction to improve predictions of ecosystem evolution, particularly for element cycling and biodiversity–function relationships. This manuscript addresses a timely topic and provides a valuable forward-looking perspective. I recommend publication after minor-to-moderate revision, mainly to improve clarity and usefulness to readers.A key improvement would be to strengthen figure/box captions and add a few summary visuals/tables. Several sections contain important ideas (especially the model–omics linkage), but the presentation is sometimes too high-level for non-specialists to follow efficiently.

General / presentation

1. Figures and boxes: Please provide more detailed captions for each figure/box so that they are informative as stand-alone items (what the panels show, what variables/axes represent, what the workflow inputs/outputs are, what assumptions/limitations apply).

2. Add summary visuals: The manuscript proposes several frameworks/steps for future research; a small number of schematic summaries (one workflow diagram + one database table + one worked example box) would substantially improve readability.

Specific comments (by line / element)

L119: Typo/formatting: please correct 10 ^ 30 to 10^30 (or “10³⁰”). The sentence currently reads “est. 1030 cells on Earth”, which should be 10^30. 

L128: Please mention the latest microbial database/resource published in Nature (2024) that is relevant to large-scale ocean microbiome catalogues. For example, the 2024 Nature study that constructed a unified global ocean microbiome genome catalogue (GOMC) from large-scale marine metagenomes would fit well here and is directly aligned with the manuscript’s message about global genomic catalogues. 

L177: Please add quantitative detail (exact numbers/ranges) to support the statement about trait variation across taxonomic/biological levels and its consequences. For instance, in the Prochlorococcus example (ecotypes/strain clusters), it would help to report at least one numeric range (e.g., contribution estimate, temperature niche range, or projected shift magnitude) rather than only qualitative wording. 

L215: I suggest revising the subsection title to “Example from an ocean system” and expanding the discussion with open-ocean examples. Many foundational concepts (e.g., mixotrophy, microbial loop) were developed in open-ocean systems and are now known to be important for carbon sequestration and broader element cycling. Adding 1–2 open-ocean examples would make this section more representative and historically accurate.

L220: Please explicitly mention the expansion and intensification of OMZs (oxygen minimum zones) under climate change (spatial expansion/shallower boundaries) and clarify how this affects redox-sensitive element cycling and microbial functional shifts. 

Box 2: Omics studies also highlight the roles and risks of viruses in polar regions (e.g., viral impacts on microbial mortality, carbon shunting, lysogeny/reactivation risks in warming/ice melt contexts). Please add a short paragraph (and key citations) reflecting this.

~L255 (pathway annotation paragraph): This part is important for future research. I strongly recommend adding a small schematic chart/flow diagram summarizing the proposed strategy (marker genes → remote homology methods → phylogenetic validation → pathway completeness checks → operon/context evidence). 

~L290: Please include a table listing recommended databases (genomes/MAG catalogues, functional annotation resources, metadata standards, etc.). Suggested columns: database name, data type (metagenome/MAG/metatranscriptome, viruses/euks), ecosystem coverage, key strengths/limitations, access/metadata standard, primary reference.

~L320 (core section): This is the heart of the review. I suggest adding a boxed worked example that walks the reader through one complete pipeline from omics to ESM parameters (e.g., genome → inferred traits → kinetic parameters → model output), including what the inputs are, what assumptions are made, and where uncertainty enters. 

~L350: The text mentions the key complexity (typo in my notes: “kexy”) that may confuse many readers, including researchers outside the subfield. Please add a short, concrete summary of how complexities are handled/discriminated in practice, e.g., a bullet list of strategy + toolkits (QC, assembly/binning, dereplication, contamination checks, trait inference tools, uncertainty quantification, validation pathways). (If you discuss ancient DNA here, also include standard authenticity/discrimination criteria.)

Fig. 4: Consider revising/expanding the figure to connect more directly to ocean microbiome projects (e.g., Tara Oceans), and to show at least one example where omics data have been used (or could be used) to predict carbon dynamics (carbon export, remineralization, plankton network effects). The current text already frames Tara Oceans as the archetypal global survey and notes that Fig. 4 is an example of projecting end-century biodiversity anomalies; strengthening the omics→carbon-dynamics linkage would better match the manuscript’s thesis. 

L449: I agree that metatranscriptomics can provide essential information on ecosystem responses to climate change. Please consider explicitly integrating metatranscriptome data into your proposed data-to-model framework (where it fits, what it resolves beyond metagenomes, and how it helps constrain “potential vs realized” function).