the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Future global water scarcity partially alleviated by vegetation responses to atmospheric CO2 and climate change
Abstract. Accurate water scarcity projections are essential for effective adaptation strategies. Most existing studies rely on hydrology models that often neglect the effects of plant physiological responses to rising CO2 on the water cycle , such as reduced stomatal opening, which can decrease transpiration and enhance water availability over large scales. Using a land surface model driven by an Earth system model under a high-emission climate scenario, we evaluate how physiological and structural plant responses to rising CO2 and subsequent climate change affect the Water Scarcity Index (WSI). Our simulations suggest that the combined effects of these plant responses partially alleviate WSI for most regions, largely due to CO2-induced stomatal closure. However, CO2- and climate-induced vegetation changes do exacerbate water scarcity in some places, particularly arid regions. By 2076–2095, when incorporating all plant responses in our projections, global median WSI is approximately 12 % lower, and among 291 global river basins, median WSI is between 10 and 70 % lower in 138 basins, home to 80 % of the global population, and between 10 % and 60 % higher in 11 basins, home to 0.2 % of the population. These model results highlight the potential for plant responses to CO2 to somewhat alleviate water scarcity, noting water scarcity is still projected to worsen in many regions, including highly populated areas. There is an urgent need to gather empirical evidence on the strength of plant responses to CO2 at large scales to address modelling uncertainties.
- Preprint
(3094 KB) - Metadata XML
-
Supplement
(1400 KB) - BibTeX
- EndNote
Status: open (until 26 Apr 2025)
-
RC1: 'Comment on egusphere-2025-51', Anonymous Referee #1, 21 Mar 2025
reply
The manuscript by Stacey et al. examines how vegetation responses to rising atmospheric CO₂ and climate change affect global water scarcity projections using the JULES model. The authors assess how physiological (stomatal closure) and structural (leaf area and vegetation cover) changes influence future water supply. The results suggest CO₂-induced stomatal closure increases water availability, potentially alleviating scarcity, particularly in tropical regions, while vegetation expansion may worsen scarcity in some semi-arid and arid areas. Overall, this study makes a valuable contribution to the literature on vegetation-climate-hydrology interactions, and I appreciate the authors' analysis and findings.
Broader suggestions and concerns
Since this study relies on a single model, a more detailed description of key model processes would help readers better understand potential limitations. While a full breakdown of all JULES equations is unnecessary, some critical details are missing. In particular, a brief explanation of the coupled canopy conductance and photosynthesis model from Cox et al. (1998) would be helpful, as the results strongly depend on this component. (For instance, does this simplification overestimate the physiological impacts?) The discussion on limitations (L553-564) mentions issues with stomatal conductance parameterization; rather than only stating this as a limitation, it would be more helpful to discuss how it affects the key findings.
In some cases, such as in Section 3.1, the results come across as relatively intuitive. While I understand that this section serves to introduce the different simulations and their effects on vegetation and water cycle variables, I would argue that the main novelty of the study lies in the quantification and analysis of the Water Scarcity Index (WSI). For that reason, Section 3.1 could be streamlined or more tightly focused, allowing the reader’s attention to shift more quickly toward the WSI results and their broader implications.
The discussion section contains many well-acknowledged limitations, which is appreciated. However, rather than listing them in loosely connected paragraphs, they should be structured to explicitly discuss how each limitation affects the main results. Some parts could also be shortened and streamlined, for instance, the justification for using runoff as a proxy for water availability is reasonable in the context of its role as a water supply source. In my opinion this could be shortened to one sentence.
While this study is an idealized experiment rather than a model performance evaluation, I think it’s important to acknowledge that non-linear biases can emerge when driving JULES with biased GCM inputs. Even though the input fields are identical across simulations, their biases may interact differently with the model’s internal processes. This contradicts your statement in L569-571: „Given that our study focuses on the influence of plant responses – comparing the differences from two simulations with similar biases - HADGEM2-ES and this version of JULES are deemed suitable for our purposes, despite the runoff biases.“ A brief discussion on how these biases might influence the results would improve transparency.
Minor and Editorial Comments
L16: Simplifying and breaking this sentence into clearer segments would improve its readability and ensure that the statistical comparisons are more immediately accessible to the reader.
Section 2.3 (Data): This is more of a stylistic suggestion, but I found this section somewhat fragmented and lacking in flow. A more cohesive, narrative-style format could improve readability. Consider merging some of the sub-subsections to reduce fragmentation, and integrating the definitions of key terminology directly into the text rather than listing them separately in bullet points. This would help create a smoother and more engaging structure.
Figures: The manuscript contains many figures, and some could be streamlined. In my opinion you could merge Figure 1 and Figure 3 and move it to the supplementary materials or place it in the methods section, clarifying that it shows HadGEM2-ES input data to JULES. In Figure 2, adding simulation abbreviations (e.g., S2 - S1 CLIM: VEG) in the second legend for isolated factors could improve clarity.
Tables 1a and 1b: These were difficult to interpret at first glance. It might improve clarity to describe Table 1a in more detail in the text and move it to the supplementary materials. For me Table 1b was the more informative one in explaining the mechanisms, so I suggest keeping the focus there.
L263: Subject-verb agreement: "Climate-induced changes in vegetation distribution drive LAI decrease," not "drives."
L293: The statement: “The climate-induced changes in runoff from the present (2006-2025) to the future (2076-2095) largely align with shifts in precipitation (Fig. 3; right), varying greatly in both magnitude and direction (Fig. 4a,c).” seems contradictory. If they largely align, how can they vary greatly in direction?
L308 – 310: You state that the runoff decreases are driven by CO2 induced LAI increases, but this is hard to see on the map and that raises the question of how large and significant the impact really is.
General formatting: Check for missing spaces between words, unclosed brackets, and misplaced semicolons throughout the text (e.g., L293).
Citation: https://doi.org/10.5194/egusphere-2025-51-RC1
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
134 | 27 | 4 | 165 | 19 | 5 | 4 |
- HTML: 134
- PDF: 27
- XML: 4
- Total: 165
- Supplement: 19
- BibTeX: 5
- EndNote: 4
Viewed (geographical distribution)
Country | # | Views | % |
---|---|---|---|
United States of America | 1 | 71 | 43 |
China | 2 | 24 | 14 |
United Kingdom | 3 | 13 | 7 |
Germany | 4 | 11 | 6 |
South Africa | 5 | 6 | 3 |
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
- 71