the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Oct 2025
Evaluation of national Greenhouse Gas Removal potential under a changing climate using a process-based land surface model
Abstract. Global warming and climate change caused by greenhouse gas emissions (GHG) will have multiple impacts on forest ecosystems. As the UK’s currently planned contribution to global efforts to mitigating these impacts, the Climate Change Act has set a goal of net zero emissions of GHG by 2050. A core part of the strategy to meet this target is to use afforestation and forestry management to implement large-scale Greenhouse Gas Removal (GGR). These measures will need to be resilient to some level of climate change even if the international community successfully meets the goals of the Paris Agreement in limiting global warming. However, the effectiveness of afforestation as a GGR strategy is difficult to fully evaluate with standard empirical models due to a myriad of changing environmental conditions. Here we use the process-based land surface model, coupled to a model of large-scale forest demography (JULES-RED). We focus on a low climate change scenario, which would yield peak global warming close to 2oC. We project that widespread Sitka Forest afforestation could potentially sequester 15 MtCO2 annually by 2080 assuming a plantation rate of 30,000 ha year-1 from 2025 to 2050. If the world fails to meet the goals of the Paris Agreement, UK woodlands will need to be resilient to more severe regional climate changes and the plantation locations will need be selected more precisely.
- Preprint
(1327 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 06 Jan 2026)
- RC1: 'Comment on egusphere-2025-4536', Anonymous Referee #1, 16 Nov 2025 reply
-
RC2: 'Comment on egusphere-2025-4536', Anonymous Referee #2, 16 Dec 2025
reply
General comments
Chou et al. present a manuscript that evaluates the extent to which large scale re/afforestation in the UK will help the UK achieve net zero emissions. The authors explore carbon uptake under various future climate and CO2 scenarios by running the vegetation demography model JULES-RED at 300 sites across the UK.
This is an important topic, and the results of such an analysis will be of interest to policy makers. However, the methods section is lacking many important details making it difficult to evaluate the relevance of the results, for example how JULES-RED represents tree physiological stress, mortality and recruitment. The authors also use a parameterisation of a Sitka Spruce PFT, calibrated to a single highly productive site, at all 300 sites across the UK. As a result, they may overestimate carbon uptake potential in many regions.
The manuscript would be improved by some restructuring of the methods and results section to provide important details and justification. The introduction and discussion would benefit from more linkages to previous work on this topic.
Specific comments
The title could be updated to indicate that this is a GB specific study.
L36-39. This section needs references on how these environmental variables are predicted to change in the UK.
L41-47. This paragraph is lacking references to other land surface models and previous work using LSMs to quantify the impact of re/afforestation on biogeochemical and biogeophysical dynamics as well as potential risks.
L49. References?
L52. More information is needed regarding the probabilistic projects. What data are they based on?
L81. I think this section would be easier to follow if details of PFT calibration came after the model description.
L85. More information on the forest management implementation in JULES-RED would be relevant here. Which processes are represented?
L86. How are growth, recruitment and mortality represented in JULES-REDD? Are they sensitive to climate, if so, how? Many details are missing which are relevant here.
L94. So JULES-RED was calibrated to a highly productive site and then run across 300 new sites? Does this mean that carbon uptake is likely overestimated at many of these sites? More explanation is needed for how much dynamics are determined by parameterisation versus response to soil and climate forcings.
It is not clear which parts of section 2.2 are summarising new work, versus results from Argles et al. 2023.
L93-100. Given that simulations were run across GB it would be more appropriate to calibrate the Sitka Spruce PFT to data from across the whole study region for the historic period, and then project forward in time, rather than use a parameterisation from a single site which is characterised as being maximally productive.
L111. What is boundary mass?
L111. Recruitment rates and baseline mortality are likely to vary spatially depending on climatic and edaphic conditions. Are there additional sources of mortality in JULES-RED that will vary spatially? Is recruitment a function of mature tree biomass at all? It is difficult to understand the meaning of these parameters without further details about JULES-RED. What does alpha=0.005 mean? Is that number of plants per m2 per year? Or some amount of carbon?
L119-124. Please provide details of how the 300 sites were chosen. Was a clustering algorithm used? Did you consider the current land use of each site, and whether it had previously been forested?
L132-141. It might help the reader if these two ensemble members were given more informative names in the paper, e.g. EM15=low warming, EM06=drying.
L162. We would expect high CO2 to reduce sensitivity to water stress so it would have been interesting to quantify that effect by assessing the sensitivity of vegetation growth to meteorological drivers under all three CO2 concentrations.
L165. See note on Figure 5 also. I don’t understand the rational for classifying results in this way, rather than presenting the full distribution of vegetation carbon.
L170. Why are forests at 30 and 55 used to calculate carbon uptake at year 42.5. I don’t understand the need for equation 3. You could calculate total carbon taken up over by the total forest area given the age distribution in year 2080.
L176-189. I think this section along with Fig. 3 and Table 1 belongs in the methods after the paragraph from L132 – 141.
L196. I don’t understand what is meant by “at forest age of 55 years”.
L199. Or the parameterisation does not allow growth under those conditions.
Figure 4. The figure does not match the caption. It looks as if figure 5 has is replicated.
Figure 5. This figure would be more informative if continuous colour scales were used, rather than four discrete bins for each variable. It is also best to avoid red-green colour schemes in general.
L245. I think these results would be easier to understand as a figure rather than a table.
Figure 7. What are the two blue lines in panel a? I see this explained in the text but a short description in the figure caption would be helpful.
L260. This whole paragraph belongs in the results section. These numbers haven’t been mentioned previously (except in the abstract) and there is no discussion of their significance here.
L269. I don’t understand this sentence. The previous paragraph states that vegetation carbon flux is higher in 2080 than in 2050 in both scenarios. But then this sentence states that vegetation carbon has considerably reduced.
The discussion summaries the results but does not really put them in the context of previous work. For example, there is a large literature on the potential for soil nutrients to limit CO2 fertilisation but this is not mentioned anywhere in the section discussing the increases in vegetation growth under future CO2. There is also no discussion of potential biogeophysical impacts of afforestation/reforestation, biodiversity impacts, or carbon permanence (e.g. risk of forest loss from fire or pests/pathogens) which should also be considered by policy makers alongside carbon uptake.
Technical corrections
Overall, the manuscript could use a close read for grammar. Tenses are frequently mixed up within a paragraph. The 2 in CO2 is often not subscript.
See comment above about figure 4.
Citation: https://doi.org/10.5194/egusphere-2025-4536-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 235 | 46 | 25 | 306 | 20 | 19 |
- HTML: 235
- PDF: 46
- XML: 25
- Total: 306
- BibTeX: 20
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of egusphere-2025-4536 (His-Kai et al.)
His-Kai et al. address the potential of afforestations for carbon sequestration in the UK under climate change. Afforestations and other forest-based carbon storage approaches are a highly important topic in the context of climate change mitigation. The study provides an example how a dynamic, climate-sensitive model could be used to estimate the carbon sequestration potential of afforestations under different future scenarios. The paper addresses certainly a scientifically significant topic that fits with the scope of BG.
However, the manuscript suffers from multiple and severe flaws regarding its scientific and presentation quality. The four major issues that are marring the manuscript are explained below ('General comments'). Additional specific comments can be found in the attached commented PDF version of the manuscript.
Based on the review criteria, I have come to the conclusion that the current version of the manuscript clearly does not meet the quality standards of Biogeosciences. There may be potential in the manuscript, but it would need to be essentially re-written from scratch based on partly rather different simulation studies. Therefore, I don’t think it can be pursued further in the present form for Biogeosciences.
General comments:
The study makes use of the coupled JULES-RED model to simulate carbon dynamics of afforestations along a wide environmental gradient in the UK, based on 300 site-specific simulations. The simulated vegetation dynamics in the model depends heavily on the parametrisation of Sitka spruce, which was based on a generic Plant Functional Type, whose parameters were heavily calibrated to only one Sitka spruce forest. This raises the question of over-calibration because (1) tree growth at this site is highly unlikely to be representative for all other 299 sites where simulation studies are conducted, and (2) one site is not representative for the environmental gradient covered by the study. Furthermore, no model validation against independent data was performed, thus not providing any assessment of model accuracy beyond the calibration site. This is scientifically highly questionable and raises doubts about the reliability of the study results overall.
JULES was developed to simulate carbon and other fluxes between the land surface and the atmosphere. Such models are usually applied at regional, country, or global scale. However, in this study the model is used at the stand scale, but the authors provide little evidence that the coupled JULES-RED model is suitable to represent forest dynamics at this spatial scale. Examples for this are the excessively simply mortality formulation (which is not sensitive to climate, as it appears) and the simple formulation of competition for light. As a matter of fact, the statements in the manuscript that refer to the model and its suitability are based on a limited number of studies, predominantly from the same author group (e.g. Argles et al., Clark et al.). Furthermore, the introduction focuses only on the model JULES but misses any broader review of forest modelling literature which would be important to put the model into context. It is everything but clear that JULES-RED is the most appropriate approach for studying stand-scale forest dynamics across 300 sites in the UK.
The text is often not logical and quite difficult to follow. The Introduction is lacking clear research questions or hypotheses, but it already contains descriptions of the methods. The Method section is incomplete and lacks transparency (for more details, cf. annotated PDF). The Result section describes the results only poorly and contains many additional explanations of the methods. The Discussion section is largely a re-cap of things from the Results section together with the presentation of new results, whereas this section actually lacks a proper discussion (i.e. putting the results achieved in this paper in the context of the broader literature). Lastly, the Conclusions section is far too long, unfocused, and contains surprising statements that are not warranted by the substance of the paper.
The paper appears to have been put together in great haste. The writing is often sloppy and there are many grammar errors and language glitches. The layout of the presented figures is often inconsistent (e.g. axis labels, placement of legend) and the captions are often insufficient to understand the figures on their own. Figures 4 and 5 are identical (i.e. we surmise that Figure 5 was pasted twice). The list of references is in a terrible state, and not all references cited in the text can be found in the list of references. Overall, the manuscript gives the impression of a quickly assembled paper using copy-paste from multiple other sources.