General comments

In their aptly titled 2024 article “Biogeochemical versus biogeophysical temperature effects of historical land-use change in CMIP6,” authors Amali et al. quantify the biogeophysical (BGP) and biogeochemical (BGC) effect of historical land-use change (LUC) as rendered in 13 earth system models of the sixth Coupled Model Intercomparison Project’s (CMIP6) Land Use Model Intercomparison (LUMIP) activity. Specifically, the authors seek to analyze the effects of historical LUC for carbon emissions and near-surface air temperature. Although the relative contributions of BGC and BGP effects of historical LUC have been studied using CMIP5 and Land-Use and Climate, Identification of Robust Impacts (LUCID) data, CMIP6’s LUMIP activity,  prescribes a set of experiments to be carried out in common by modeling teams, using the latest generation of earth system models. The study is timely as the BGP impacts of LUC have often been overlooked. Where it has not been overlooked results have at times been difficult to interpret due to the variety of LUC schemes applied within CMIP5. This study avoids this particular challenge by using data from the latest generation of models and experiments where simulation protocols dictate greater consistency across models.

Two concentration-driven CMIP6 simulations are used by Amali et al. to analyze the effects of historical LUC. The historical simulation with LUC from 1850 to 2015 and hist-noLu where LUC is held constant from 1850. The difference between the two simulations is taken to determine the change in carbon storage and near-surface temperature. The authors use the TCRE to find the BGC temperature effect of LUC. To obtain gridcell depictions of this temperature effect, the authors use the regional-to-global ratio of temperature (or simple pattern scaling). These methods allow the authors to isolate the impact of historical LUC on the variables of interest and identify the contributions of BGC and BGP for each.

The study’s findings both align with and expand upon previous work. For example, the finding that near-surface temperature increase from BGC is greater than BGP for historical LUC aligns with the findings within the existing literature. However, the regional analysis in Amali et al. adds nuance to this story in that the regional effect of BGP on near-surface temperature can be significant depending on location. Also significant is the study’s contribution to our understanding of the BCG effect on near-surface temperature change at the gridcell level. Furthermore, the findings of this study demonstrates similar model spread and estimates to previous similar studies using LUCID or CMIP5 data, and identifies some reasons related to model architecture that contribute to this result. 

This study is ambitious in scope, well-referenced, and contributes significantly to our understanding of the relative temperature contributions of the BGC and BGP effects of LUC, using a novel RGRT approach to do so. Its conclusions are supported by the results, however, it’s possible that the conclusion that both the local and non-local effects of LUC ought to be considered in climate policy development should be qualified, noting that this is because combined local and non-local BGP effects of LUC found in this study are not insignificant. The article is recommended for publication pending consideration of the questions and comments that follow.

Specific comments

1. The abstract provides a complete and concise summary. 
2. The manuscript is also well-structured in that the sections and subsections allow the authors to present their methods, results, and discussion in a manner that is both logical and appealing from a reading flow perspective. One subsection that might benefit from being split in two is 2.3.2, where “Global temperature response” and “Local contributions to global temperature change” could each be their own sub-sections.
3. The figures do a good job of presenting the results and key points for discussion in a readable fashion. Related to the methods and the results presented in Figure 3 where ΔTbgc is presented, why is there no test of statistical significance as is the case for ΔTbgp?
4. The methods are clearly described, including useful model information presented in tabular form and details of the statistical analysis. Related to the latter, is it possible to include which type of interpolation method was used to bring all of the simulation data into a common grid? This would aid with reproducibility.
5. Related to Table 2, is it possible that the average ∆ cLand is -131.9 (±96) GtC rather than -122 (±96) GtC?
6. On page 14 where the methods for obtaining the grid cell temperature contribution and effect are discussed, is it possible to add a small amount of text to indicate the significance of or motivation for providing both quantities?
7. Page 25 line 511: “In magnitude, the warming pattern around Greenland can only be seen in the BGP contribution, which we attribute to mechanistic non-local LUC-induced effects on ocean currents and sea ice.” This result seems worthy of mention in the discussion and conclusion sections.
8. Page 28 line 550: Is it possible to include the direction in which AMOC may have been influenced?
9. For the subplots in Figures 2-4 that represent just the direction of change and not the magnitude, is it possible to remove the numbers from the colorbars?

Technical corrections

1. Page 5 line 119: Please delete the “s” at the end of “backdrop.”
2. Page 13 line 313: The text reads “for a period ranging between 150 to 165.” Please include the unit for “150 to 165” if units apply.
3. Page 17 line 403: In “- a trend” [...] “∆cSoil -,” please replace the hyphens with em-dashes.
4. Page 24: The acronyms given in the caption for Figure 5 are not consistent with those given in the figure and in some places in the main text. Please adjust.
5. Page 27 line 543: In the sense that is likely intended, “widespread” ought to be written as “wide spread.”
6. Page 29: For Figure 7, is it possible to replace the dashed line separating the temperature (panels a and b) and carbon stock (panel c) with a solid line? This might further emphasize that data on two effects are presented in this figure. 
7. Page 30 line 609: Should “BGC and BGC” read “BGC and BGP”? If so, please change this.
8. Page 36 line 791: Is it possible that temperature is being referred to here rather than “climate”?
9. The supplementary information is very helpful for understanding the results model-by-model. It’s possible that colorbars in figures S13, S14, S15, and S16 are a bit high compared to previous multi-model plots in the supplement. 

This paper updates estimates of the biogeochemical and biogeophysical effects of land use change on surface temperature using CMIP6 output. They find large differences is estimated temperature effects across models, and generally similar magnitudes to prior work. 

Overall I find the main conclusions of the paper are somewhat buried. The text is very long and the main take home points are not very clear within the paper as a whole. There is extensive comparison with prior similar estimates and recapping of prior literature, but I didn't come away with an understanding for what was learned through this work specifically that wasn't already in previous work. This was especially the case in the discussion section. I understand that this paper uses different simulations than prior work, but what other insights the authors present that were not covered by previous literature I am unsure. I say this as someone slightly outside of the land use change sphere, where the detailed findings in the field are less known to me. I encourage the authors to make their specific contributions more clear in a revised version.

There are a lot of acronyms in this paper! (LUC, BGP, BGC, ESM, SNR, A/R, GHG, RGRT, TCRE, DGVM, all of the model model names, etc). There is a lot to keep track of, especially for anyone not well embedded in this discipline. I suggest picking as few as you can that used extensively and spelling the rest out. For example A/R, SNR, RGRT don't seem necessary as acronyms to me. Remove any you can reasonably remove to help out your readers!

Specific comments:

line 70: "local and non-local temperature changes" There is additional literature on this that is relevant, for example Laguë et al. 2019.

line 309 - not clear how "RGRT" relates to the equations shown.

line 331, equation7: I find this equation confusing to interpret. The authors acknowledge elsewhere in the manuscript that there are non-local effects of land cover change and that that this metric can't capture them because the temperature change in a single gridcell may be comprised already of both local and non-local impacts from land use change. Why then call it Tlocal? I'd suggest calling it someting else that more accurately describes what it represents. 

line 358: "form a distinct cluster" - I do not see anything visually obvious like this in Figure 1. Needs further illustration or description. 

line 360: "likely leads to similar trajectories" why? Please provide evidence and a hypothesis.

line 411: "across dynamic global vegetation models" do the authors mean TRENDY models? This is a confusing change of language, and additionally I don't think all TRENDY models are DGVMS - do the authors mean a subset of TRENDY?

line 450-451: see additional papers on how plant feedbacks with changing CO2 can amplify warming in high latitudes - Park et al. 2020 and  Park et al., 2021. I think this literature is highly relevant to understanding the BGP effects of land use change.

Figures 2-4: I suggest that the authors make the signal to noise ratio and inter-model agreement plots on a different color bar from the quantities being plotted (T, carbon) and the same colorbar for signal to noise ratio and inter-model agreement on all three plots. It is confusing to have it plotted in the same colorbar but showing a different unit.

line 496-497: "obviously driven foremost by the LUC in that gridcell" Can the authors say why this is obvious?

line 501: "local BGP effects appear to dominate" How do the authors know that? I don't see evidence for this statement since they just stated that they can't calculate how much is local vs. non-local. Please provide further evidence.

Figure 7. I think it would be helpful to label each of the three sections directly on the figure (biogeophysical, biogeochemical, change in carbon stock).

line 686: "additional insights" - What about BCC and CESM2 provides insights? It isn't clear.

line 731-734: This explanation isn't clear to me. Also what are the authors finding here about CESM2? Many of these paragraphs emphasize prior work but it isn't clear what is a new insight from this work.

line 745: "this underscores" I'm not sure what this is referring to? Is this a finding of this paper or of prior work?  

References:

M. M. Laguë, G. B. Bonan, and A. L. S. Swann. Separating the impact of individual land surface properties on the terrestrial surface energy budget in both the coupled and uncoupled land–atmosphere system. Journal of Climate, 32(18):5725–5744, 2019. https://doi.org/10.1175/JCLI-D-18-0812.1

S.-W. Park, J.-S. Kim, and J.-S. Kug. The intensification of arctic warming as a result of co2 physiological forcing. Nature Communications, 11(1):2098, 2020. https://doi.org/10.1038/s41467-020-15924-3

S.-W. Park, J.-S. Kug, S.-Y. Jun, S.-J. Jeong, and J.-S. Kim. Role of cloud feedback in continental warming response to co2 physiological forcing. Journal of Climate, 34(22):8813–8828, 2021. https://doi.org/10.1175/JCLI-D-21-0025.1