General comments

The article “Mapping the Future Afforestation Distribution of China Constrained by National Afforestation Plan and Climate Change” explored the distribution of future potential afforestation areas based on future high-resolution climate data from the WRF model and HLZ model. It is highlighted that the afforestation scenario is constrained by both the climatological suitability for tree and national afforestation plan. The climatology suitability for tree is decided by future climate conditions and determines the potentially available afforestation domain. The national afforestation plan determines the total afforestation area. The potential value is to provide the design framework for locations of future afforestation. Overall, the article is suitable for the scope of Biogeosciences, I recommend that the authors address the concerns below in a minor revision prior to publication.

Specific comments

Method: I'm confused about the spatial resolution of the article and please provide an explanation. Firstly, the authors emphasize the “high-resolution simulations” in this article. However, the spatial resolution is only 25 km. The other high-resolution climate dataset product (i.e., WorldClim data, https://www.worldclim.org/data/index.html.) is available at the ~1km spatial resolution. I'm confused if that description is appropriate, and please illustrate the advantages of WRF simulation in this study. L115: Why the spatial resolution of ERA5 reanalysis data is 1.0°×1.0°. In ECMWF, the highest resolution of the ERA5 product is 0.25°×0.25°, which is close to WRF simulation (25 km). In the HIS_ERA experiment, is downscaling 1.0° ERA5 data to 25 km necessary? L89: The spatial resolution of MCD12Q1 is 500m, which is different from the WRF simulation (25km). How do you match it well? Please give some detailed information.

L218: “Areas with high precipitation are allowed priority afforestation.” In this study, precipitation is treated as a key meteorological factor that restricts forest distribution. Indeed, precipitation is critical for forest growth. However, a single climate variable is slightly simple rather than representing climatology suitability for tree. Multivariate comprehensive indicators affecting forest growth are more appropriate. In this study, the essence of the HLZ model is the distance to the three bioclimatic variables. I recommend considering the distance as a comprehensive indicator to quantify the climatology suitability for tree.

L204: In the section on the approach of the newly afforestation allocation, I'm confused about the definition of forest. Please clarify it. For the national afforestation plan (NFMP), the total afforestation area is 73.78×10⁴ km². How to define the total afforestation area? I wonder whether the definition from the State Forestry Administration of China agrees with this study.

L113: The authors use the SSP2–4.5 scenario (the middle-of-the-road development) to represent the climate future projections. However, this study only used one model projections rather than multiple model ensemble mean. Following the methodology of CMIP6 climate projection, scenario-based climate projection may have large uncertainties. It is suggested the revision to address this issue. It is also worthy to discuss effects of single model projection uncertainties on the research result of this study.

L353: “Our findings indicated that future afforestation in China would mostly occur around and to the east of the Hu Line, consistent with Zhang et al. (2022).” The authors try to compare other similar studies on future potential afforestation distribution. More result differences should be discussed. I suggest to highlight the innovation and implications of the article by comparing with existing studies.

L180-186: Why the Holdridge life zone (HLZ) model is suitable for simulating the potential vegetation types in China. The author simply describes the extensive application of the HLZ model. I suggest validating the accuracy of the HLZ model. It is necessary to compare potential vegetation types with true vegetation types. Please add it to the Supplement Material.

L125: The authors have done substantial work on numerical experiments. For example, the authors correct the lateral boundary conditions rather than the raw GCM before dynamic downscaling. It is a very good solution to reduce the underlying bias. I suggest adding the comparison of raw GCM, bias-corrected GCM, and observation.

L351: This article emphasizes “The dataset would be valuable for studying the effects of future afforestation on carbon budget, ecosystem service, water resources, surface climate”. Would the data set be available to the public, especially in Figure 7?

L234: “The WRF simulation generally overestimates TP in most regions with a national-average bias of 92.883 mm”. According to Figure 3d-3f, the obvious overestimate is over the southeast Tibetan Plateau. It is suggested to explain the potential reasons of these bias in the revision.

Table 1: Why this parameterization scheme of the WRF model is appropriate in this study. Please give a specific reason or reference.

L300: What is the meaning of “The corresponding annual total precipitation is over 353.6 mm among

the selected grids”? How to obtain the value of 353.6 mm. Please clarify it.

L311: “It is generally common sense that afforestation is highly constrained by precipitation.” Please add specific explanations or references.

L275: To what does “total area” refer to? Is it the whole nation? Please clarify.

Figure 5b: The flow diagrams are not clear, and please give specific values.

Eq. (2): “ ,  , and ”. Please correct it.

Figure 3 and Figure 4: For Figure 3 and Figure 4 captions, suggest not to use the abbreviations “HLZ”, “AT”, “TP”, and “PE”.

L208: “national afforestation plan” is redundant. Please use the “NFMP”.

L98: “The total national afforestation area is about 73.78×10⁴ km² from 2020 to 2050”. Please give specific forest cover.

Figure 2: No citation for Figure 2 in the text.

Figure 6: Please do not use the abbreviations in the figure captions.

L338: “woody savannas” replaces “Woody savannas”.

General comments:

This manuscript attempted to map the future afforestation distribution in China. This future afforestation distribution plays an important role in land-atmosphere interactions and carbon cycle research, but it hardly been obtained so far. The authors provided a technological roadmap to deal with it. Compared to previous idealistic and hypothetical afforestation scenarios, this study designed a plausible afforestation scenario due to considering the national afforestation plan. The study also did a relatively good job at dynamical downscaling of GCM outputs in terms of future climate projection. Overall, the study adopted a novel perspective and robust technique for identifying future potential afforestation domains.

I find that this paper is very intriguing and important and lots of additional work behind this study is worth further exploring. The manuscript could be accepted as I believe. On the other hand, I also have several minor comments. I hope that these comments can improve the manuscript. My comments are given below.

Specific comments:

1. Why is the SSP2–4.5 scenario selected? There are several shared socioeconomic pathways (SSPs) for future climate projections in the CMIP6. The study results may be dependent on the selection of SSPs. Why is the SSP2–4.5 scenario suitable for your studies?
2. By comparing potential vegetation domain simulation with observation, some disagreement could be found. For example, in southern China, the observed subtropical forest expands northward up to 32°N. However, the simulation results reduce the extent. Given the bias in the WRF model simulation, why does this simulation still make sense?
3. This study only used an MPI–ESM1–2–HR model as the lateral boundary of WRF model. It may fail to obtain robust future climate projections. The NEX-GDDP-CMIP6 (NASA Earth eXchange Global Daily Downscaled Projections CMIP6 Data) datasets contain multiple GCMs and SSPs with a spatial resolution of 0.25° × 0.25°, which is approximate same with this study of 25- by 25-km. Why not use this dataset? The relevant reference is “Thrasher, B., Wang, W., Michaelis, A., Melton, F., Lee, T., & Nemani, R. (2022). NASA global daily downscaled projections, CMIP6. Scientific Data, 9(1), 262.”
4. From Figure 6c to Figure 7, you further constrained the afforestation area through the total precipitation. Precipitation is important but not the only determinant of afforestation allocation. More other factors may be needed to be considered.
5. It’s not clear that “This bias-corrected approach was applied to the variables such as air temperature, specific humidity, zonal wind, meridional wind, geopotential height, etc.” in Line 121. In addition to these five meteorological variables, were there other variables bias-corrected? More detail please.
6. The authors need to add more descriptions of the future potential afforestation distribution and shift types (Figure 7). It seems that this part of the manuscript is too short.
7. In line 209, here, it is stated that the cropland does not encroach on afforestation. However, Figure 7 shows that the shift types include croplands to MF and croplands to DBF. Is that a contradiction here? This should be commented.
8. In line 292, “We exclude some ineligible regions, including present forestland, cropland, urban, wetland, and water bodies based on the 2020 MCD12Q1 land cover data”. This sentence is repeated. The definition of “historical open space regions” has been clarified in section 2.2.3.
9. Line 114. The presentation on the ERA5 dataset is too short. Which meteorological variables are used in the study? What is the time scale and spatial extent?

Technical corrections:

Line 100 – “Climate Modelling” replace “Climate modelling”.

Table 1 – Give specific model top pressure.

Line 125 – Check the Equation (2).

Line 393– “Woody savannas” -> “woody savannas”

Line 76 – “The fourth section will be the discussion.”

Table 1 – This should be “Initial and lateral boundary conditions”

Line 14 – 7.  In the abstract section, the abbreviation (WRF) should be the full name. Make sure the reader understandings.

Table 1 – “ERA5 analysis” -> “ERA5 reanalysis”

Line 102– “CMIP6”. Add full name.

Line 17– “SSP”. Add full name.

Line 20– “occur” -> “be located”

Line 54– “employ” -> “employed”

Line 74– “the total area afforestation” -> “the total afforestation area”

Line 83– “from 1995–2014” -> “from 1995 to 2014”

Line 89– “features” -> “featured”

Line 378– “historical periods” -> “historical period”

In this manuscript, Song, Zhang, and Yan mapped the future afforestation distribution of China under political guidance and climate change. It is a good example to serve the society using numerical techniques. Overall, this manuscript is clear-written, easy to understand, and seems to be methodologically sound. I like the most of plots in this manuscript. However, I still have several comments that should be addressed below.

Specific comments:

Line 20: Please explain the Hu Line. Readers outside are not familiar with this geographical division.

Line 24: Replace “surface climate” with “surface hydroclimate regime”.

Line 29: Afforestation not only influences the land surface energy and mass budgets, but also affects the water cycle. Water cycle should be mentioned, since in the main text PRCP and ET are analyzed.

Lines 27-32: The authors listed several papers describing the benefits of afforestation. More details would be helpful for readers to understand the impacts of afforestation from the process level.

Line 33: “Aggressively” is not a positive word.

Line 36: Add the time constraint for global greening.

Line 44: Please refine this sentence: “trigger consequent effects on climate change, hydrological processes, carbon budget, ecosystem services”.

Line 45-46: Please provide the details for “sensitive to wetland reduction caused by afforestation” and “properties and intensities of these effects are highly dependent on the afforestation location and area. ” I left confused about how the authors conclude.

Line 49-55: The authors listed several papers and did not explain their findings on climate impact; in addition, please identify the deficiency of “employ idealistic and hypothetical afforestation scenarios”.

Lines 63-71: Besides the dynamic downscaling, it would be beneficial to discuss the statistical downscaling. Moreover, dynamic vegetation studies for future projections in China are relevant to this topic, and the related studies should be mentioned in the literature review.

Line 67: Please talk about the uncertainties for GCMs.

Lines 79-80: please add sequence numbers for three categories.

Figure 2: The red text on a dark blue background is hard to read.

Lines 162-163: Did the authors test whether the model has reached the equilibrium state with only one year of spinning up?

Line 164: Delete the space between FUT_  and MPI.

Line 212: Please change the unit mu into a standard international unit.

Figure 3: in addition to the difference in FigS2. A pattern correlation and RMSE for AT, TP, and PE in Fig.3 would be beneficial.

Figure 4: Maybe I missed something, but adding texts to identify the difference among a, b, and c would be helpful. More info in the caption also would be beneficial for reader to understand this figure.

One interesting finding from the figure is that the model tends to underestimate the TP in the high value (>1600 mm) category and overestimate the PE in the high value (>3; unit?) category.

Figure 5: Please add some values for change in the Fig. 5b.

Figure 6: Some text overlaps with the map.