Weakening Correlation and Delaying Response Time of Ecosystem Water Use Efficiency to Drought

Wang, Zijun; Wu, Rong; Liu, Yangyang; Zhang, Zhaoying; Wen, Zhongming; Wang, Zhenqian; Sitch, Stephen; Yuan, Wenping

doi:10.5194/egusphere-2025-6195

Preprints

https://doi.org/10.5194/egusphere-2025-6195

Preprints

14 Jan 2026

| 14 Jan 2026

Status: this preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).

Weakening Correlation and Delaying Response Time of Ecosystem Water Use Efficiency to Drought

Zijun Wang, Rong Wu, Yangyang Liu, Zhaoying Zhang, Zhongming Wen, Zhenqian Wang, Stephen Sitch, and Wenping Yuan

Abstract. Ecosystem water use efficiency (WUE), defined as the ratio of carbon gain to water loss, is significantly affected by drought. Elucidating the coupling relationship between WUE and drought is essential for understanding the carbon–water trade-off strategies of vegetation under drought stress. Most existing studies mainly evaluated coupling relationship using correlation coefficients or regression slopes. However, the optimal drought timescale governing WUE responses to drought has not yet been systematically investigated. To fill these gaps, this study investigated the spatiotemporal patterns of the WUE–drought coupling relationship (characterized by the maximum correlation coefficient, R_max, and optimal lag time, T_opt) across global terrestrial ecosystems from 1982 to 2018, and further explored the potential causal mechanisms. The results revealed a delaying of the drought-response timescale of WUE, accompanied by a weakening in the WUE–drought correlation at the optimal timescale, as evidenced by a decrease in R_max at a rate of -0.0003/year and an increase in T_opt at a rate of 0.0155 months/year, indicating a globally weakened coupling relationship. Moreover, pronounced heterogeneity in the changes of coupling relationships changes was observed across different drought gradients and vegetation types. Attribution analysis indicated that CO₂ fertilization was the primary factor contributing to the weakening of the coupling relationship. Surface soil moisture (SMsurf) was the most critical hydrothermal driver, exhibiting nearly opposite effects and significant threshold effects on R_max and T_opt. Causality diagnosis was further employed to construct direct and indirect causal networks of hydrothermal factors affecting R_max and T_opt across different vegetation types and drought gradients. This study highlights the weakened coupling between WUE and drought, suggesting that vegetation's carbon-water trade-off is evolving toward drought adaptation, which is crucial for understanding the adaptive strategies of vegetation in response to climate change.

Received: 12 Dec 2025 – Discussion started: 14 Jan 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 3063 KB)

Supplement (7248 KB)

Download & links

Zijun Wang, Rong Wu, Yangyang Liu, Zhaoying Zhang, Zhongming Wen, Zhenqian Wang, Stephen Sitch, and Wenping Yuan

Status: open (until 03 Apr 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2025-6195', Anonymous Referee #1, 11 Feb 2026 reply
General Comments
This manuscript investigates the long-term changes in water use efficiency (WUE) under water stress, a classic and important topic in ecohydrology. The study contributes to understanding vegetation carbon–water trade-off strategies during drought conditions. However, the manuscript would benefit from substantial improvements in writing clarity and in the mechanistic interpretation of the findings.
The current version lacks sufficient mechanistic explanations and reads more as a statistical description rather than a process-based analysis.

Different types of drought (e.g., soil moisture drought vs. atmospheric drought) involve distinct mechanisms that influence vegetation responses. The manuscript should clearly specify the drought type throughout and discuss their differing impacts on WUE.

The importance of the timing of WUE changes under drought is not sufficiently justified. The authors should clarify why the temporal evolution of WUE under drought matters from a physiological or ecohydrological perspective.

Specific Comments
Abstract
The types of data used in this study (e.g., satellite products, flux data, reanalysis, etc.) should be clearly stated in the abstract.

The stated knowledge gap (L19–20) is not entirely accurate, as several previous studies have already examined WUE under drought conditions. The authors should better position their contribution relative to existing literature and clarify the physical interpretation of the proposed drivers of WUE changes.

The abstract contains too many abbreviations, which reduces readability. Consider minimizing abbreviations to improve clarity.

The findings derived from the causal analysis should be explicitly summarized.

Introduction
Lines 46–71 list a large number of studies without clear synthesis. This paragraph should be restructured to highlight key insights from previous research. Based on the cited literature, what are the dominant drivers of WUE under drought?

Data and Methods
The descriptions of datasets and metrics are mixed in the current “Data” section. It would be clearer to move the definition and calculation of WUE to the Methods section and separate data description from methodological explanation.

Results
The figure presentation should be improved for readability. Some text elements are too large, while others are too small. Additionally, abbreviations in figure captions should be reduced.

The labeling in Fig. 2 is unclear and should follow a consistent alphabetical format (e.g., a, b, c). Moreover, the color schemes in Fig. 2a(1–2) and Fig. 2b(1–2) are too similar, making global patterns difficult to distinguish.

The manuscript lacks a physical explanation for how elevated CO₂ concentrations lead to a decoupling between GPP and ET during drought conditions. A clearer eco-physiological interpretation is necessary.

Reply
Citation: https://doi.org/10.5194/egusphere-2025-6195-RC1
RC2: 'Comment on egusphere-2025-6195', Anonymous Referee #2, 07 Mar 2026 reply

This study systematically reveals the spatiotemporal evolution and driving mechanisms of the coupling relationship between global water use efficiency (WUE) and drought through the integration of multi-scale and multi-source datasets combined with model ensembles. The study represents a significant innovation in ecohydrology and possesses considerable scientific value for understanding carbon-water cycles under a changing climate. Although the methodology is rigorous and the results are scientifically meaningful, there is still room for improvement in terms of presentation and technical details. Therefore, I recommend the acceptance of this manuscript after the following revisions are addressed.
Major comments:
1. Section 3.1: Why is Rmax defined as the maximum value of the correlation coefficient rather than the maximum absolute value of the correlation coefficient? The authors should provide a reasonable explanation for this choice. In addition, a more comprehensive explanation of both Rmax and Toptshould be provided in this section.
2. Section 3.3: The authors employed XGBoost and Shapley values for attribution analysis; however, essential technical details are lacking, such as key hyperparameters (e.g., max_depth, learning_rate). These should be provided to improve the clarity and reproducibility of the manuscript.
3. Section 3.3: The authors state that “only the seven models that were consistent with the remote sensing results” were used. Please clarify the criteria used to select these seven models. If the selection follows approaches used in previous studies, appropriate references should be provided
4. From the figures, it appears that the analysis was conducted in regions with permanent vegetation. The authors should clarify how grid cells classified as permanent vegetation areas were selected.
5. In the Results section, many geographically specific terms referring to relatively small regions are used, such as the Chersky Range, Chad Basin, Katanga Plateau, and Karaganda Basin. I believe that such detailed geographical descriptions may hinder readability. I suggest replacing them with broader regional descriptions, such as “central Africa”.
6. In the Results section, the authors repeatedly state expressions similar to “the results remain consistent under the 16-year and 20-year moving window analyses.” I understand that the authors conducted extensive work to ensure the robustness of the results; however, this repeated wording appears redundant. It is unnecessary to reiterate this statement multiple times, and I suggest placing it once in the Methods or Discussion section instead.
7. The Discussion section should further elaborate on the physiological mechanisms. The Results indicate that atmospheric CO2 is the most important driver of the weakening coupling between WUE and SPEI, yet the manuscript provides insufficient mechanistic discussion regarding this key conclusion. The authors are encouraged to consult relevant literature and expand the discussion to better elucidate the underlying ecohydrological mechanisms.
Minor comments:
Please carefully check the grammar, tense, and other issues throughout the full text.
1. Line 49: Forests.
2. Line 67: Grammatical error; revise to applied machine learning methods and found.
3. Line 80: increased - Check the tense.
4. Ling 168: Equations.
5. Line 196: were.
6. Line 208: varying the width of the moving window: Here directly state the width of other moving window used.
7. Line 224: The standard expressions for “training/test sets” in scientific papers are training and test subsets, not “testing”.
8. Line 248, Line 249: Was.
9. Line 266-Line 272: According to aridity index, the globe is classified into arid, semi-arid, sub-humid, and humid. These are dryness or aridity gradients, instead of drought gradients.
10. Line 332: Provide an explanation of the sub figuresin the figure caption.
11. In the discussion section: academic papers require a clear statement of the study limitations to guide future research, but this subsection is missing in the manuscript. It is recommended to add a subsection entitled “Study Limitations” to objectively analyze the shortcomings of the present work and propose directions for future research.

Reply

Citation: https://doi.org/10.5194/egusphere-2025-6195-RC2

Zijun Wang, Rong Wu, Yangyang Liu, Zhaoying Zhang, Zhongming Wen, Zhenqian Wang, Stephen Sitch, and Wenping Yuan

Supplement

https://doi.org/10.5194/egusphere-2025-6195-supplement

Zijun Wang, Rong Wu, Yangyang Liu, Zhaoying Zhang, Zhongming Wen, Zhenqian Wang, Stephen Sitch, and Wenping Yuan

Viewed

Total article views: 355 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
236	101	18	355	39	17	32

HTML: 236
PDF: 101
XML: 18
Total: 355
Supplement: 39
BibTeX: 17
EndNote: 32

Views and downloads (calculated since 14 Jan 2026)

Month	HTML	PDF	XML	Total
Jan 2026	115	41	9	165
Feb 2026	68	24	6	98
Mar 2026	53	36	3	92

Cumulative views and downloads (calculated since 14 Jan 2026)

Month	HTML	PDF	XML	Total
Jan 2026	115	41	9	165
Feb 2026	68	24	6	98
Mar 2026	53	36	3	92

Viewed (geographical distribution)

Total article views: 342 (including HTML, PDF, and XML) Thereof 342 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 21 Mar 2026

Short summary

Ecosystem water use efficiency exhibited a reduced correlation with drought and a delayed response timing from 1982 to 2018, a pattern primarily driven by CO₂ fertilization. Surface soil moisture emerged as the dominant hydroclimatic driver, exerting contrasting influences and exhibiting pronounced threshold effects. This study highlights the divergent vegetation adaptation strategies across arid and humid regions, underscoring the need to reassess the water use efficiency–drought relationship.


Total:	0
HTML:	0
PDF:	0
XML:	0