the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Jun 2026
Understanding ecosystem gross primary productivity, evapotranspiration, and water use efficiency of maize using ISBA-A-gs land surface model over temperate and tropical semi-arid climates
Abstract. Water use efficiency (WUE), a key ecohydrological indicator linking carbon assimilation and vegetation water loss, is critical for understanding ecosystem responses under changing hydro-climatic conditions. Process-based land surface models (LSMs) are widely used to represent carbon-water interactions; however, their ability to simulate ecosystem-scale WUE across contrasting climates remains limited. This study evaluates the performance of the Interactions between Soil-Biosphere-Atmosphere model with A-gs photosynthesis scheme (ISBA-A-gs) implemented within the SURFEX land surface modelling platform in simulating gross primary productivity (GPP), evapotranspiration (ET), and WUE (GPP/ET) for maize grown under temperate (France, FR-Lam) and tropical semi-arid (India, Ind-IITH) climates. The model was driven by site-specific meteorological and vegetation variables across six growing seasons under sprinkler irrigation at FR-Lam, and two seasons (monsoon and winter) under alternate furrow irrigation (AFI) at Ind-IITH. Model calibration revealed that FR-Lam is characterized by relatively higher cuticular conductance and pronounced atmospheric control on stomatal behaviour, whereas at Ind-IITH, AFI-induced adjustments in mesophyll conductance and soil moisture stress thresholds. At FR-Lam, ISBA-A-gs simulated the seasonal mean cumulative GPP, ET, and WUE of 1039 ± 20 gC m-2, 610 ± 31 kg H2O m-2, and 1.70 ± 0.10 gC kg-1 H2O, respectively, as compared to measured values of 1026 ± 30 gC m-2, 562 ± 42 kg H2O m-2, and 1.82 ± 0.11 gC kg-1 H2O correspondingly. At Ind-IITH, the model simulated the seasonal mean cumulative GPP, ET, and WUE of 766 ± 15 gC m-2, 567 ± 30 kg H2O m-2, and 1.35 ± 0.11 gC kg-1 H2O, respectively, as compared to measured values of 793 ± 11 gC m-2, 522 ± 20 kg H2O m-2, and 1.51 ± 0.12 gC kg-1 H2O correspondingly. Further, the diagnostic analysis using the GPP·VPD0.5-ET relationship revealed that ISBA-A-gs realistically captures the coupling between carbon assimilation and transpiration-driven water loss. Overall, ISBA-A-gs demonstrates strong capability in simulating carbon and water fluxes of maize, particularly in representing WUE dynamics under contrasting climate regimes.
- Preprint
(1832 KB) - Metadata XML
-
Supplement
(144 KB) - BibTeX
- EndNote
Status: open (until 28 Jul 2026)
- RC1: 'Comment on egusphere-2026-3119', Anonymous Referee #1, 29 Jun 2026 reply
-
RC2: 'Comment on egusphere-2026-3119', Anonymous Referee #2, 05 Jul 2026
reply
Reviewer Comments
This manuscript evaluates the performance of the ISBA-A-gs land surface model in simulating gross primary productivity (GPP), evapotranspiration (ET), and water use efficiency (WUE) of maize across two contrasting agroecosystems: a temperate site in France and a tropical semi-arid site in India. The study is based on a rich observational dataset, including eddy covariance measurements, meteorological observations, vegetation variables, and irrigation information. The modelling results are generally encouraging, particularly for GPP, and the comparison between conventional and alternate furrow irrigation is potentially useful for understanding carbon-water trade-offs in irrigated maize systems.
However, I find that the manuscript in its current form does not yet clearly articulate its main scientific contribution. The dataset is valuable and the simulations appear promising, but the manuscript reads largely as an extensive model calibration and evaluation exercise rather than a tightly framed contribution to understanding ecosystem carbon-water coupling. I therefore recommend major revision.
Major comments:
1. The novelty and central contribution need to be clarified.
The manuscript currently presents its novelty in several different ways: (i) the first application of ISBA-A-gs in its MEB configuration to a semi-arid Indian maize system, (ii) an ecosystem-scale WUE assessment using a locally calibrated LSM under contrasting climates, partly motivated by a keyword-based literature gap, and (iii) the effect of irrigation management (CFI vs AFI) on carbon-water coupling. In my view, the third aspect is the most scientifically compelling contribution, but it is not developed as the main thread of the paper. A first application of an existing model to a new site is useful, but it is mainly incremental and may not provide sufficient novelty on its own. I strongly encourage the authors to define a single central research question, for example: can a process-based LSM reproduce and mechanistically explain irrigation-driven divergence in ecosystem WUE, and does this behavior differ between temperate and semi-arid maize systems? The manuscript should then be restructured around this question, with the CFI/AFI analysis and its mechanistic interpretation placed at the center.
2. The manuscript is too long and repetitive.
The manuscript would benefit from substantial streamlining. The Introduction alone is very long (over 150 lines) and currently reads more like a broad survey than a targeted motivation for the study; it should be shortened and focused on the specific knowledge gap that this paper addresses. Similar redundancy occurs throughout the manuscript. Much of the Results section restates, in prose, values that are already shown in the figures and tables, including year-by-year and treatment-by-treatment GPP, ET, WUE, and performance metrics. The Discussion also frequently repeats the Results before developing broader interpretation. In addition, the site and instrument descriptions are highly detailed and partly overlap with information available in previous publications. I recommend that the authors condense the Introduction, move routine methodological or site details to tables, figures, appendices, or the Supplement, and focus the main text on the patterns, anomalies, and mechanistic interpretation that directly support the central research question. The two diagnostic analyses should either be more clearly differentiated in terms of their scientific purpose or streamlined if they lead to similar conclusions. A substantial reduction in length would greatly improve the clarity and readability of the manuscript.
3. The calibration/validation strategy at Ind-IITH is unclear, and the dataset is insufficient for strong transferability claims.
The two sites are strongly imbalanced in data volume. FR-Lam includes six growing seasons, whereas Ind-IITH, which is presented as a key novel component of the study, includes only one year, with monsoon and winter seasons under CFI and AFI (Table 2). This limits the assessment of interannual variability at the Indian site and makes strong claims about model transferability across contrasting hydro-climatic regimes difficult to support. These claims should be tempered unless additional years are included or the associated uncertainty is discussed more explicitly.
More importantly, the calibration and validation design for Ind-IITH is unclear. Sect. 2.4 states that simulations were conducted for “four treatments (2 seasons)” at Ind-IITH during calibration, but also states that the optimized parameter set was validated using “Ind-IITH: two seasons” not included in the calibration phase. In addition, Table 6 reports separate optimized parameter values for CFI-Monsoon, AFI-Monsoon, CFI-Winter, and AFI-Winter. The authors should specify exactly which observations were used for calibration and which were used for validation, explain how independence between calibration and validation was ensured, and adjust the wording of the conclusions accordingly.
4. The quantitative reporting needs to be made consistent.
Some key numbers are reported inconsistently across the manuscript. First, the sign convention for GPP is confusing. In the Abstract, GPP is reported as a positive cumulative value, for example 1039 gC m⁻² at FR-Lam and 766 gC m⁻² at Ind-IITH (p. 2, lines 33-38). However, in the Results, FR-Lam GPP is reported as negative, for example −8 to −11 gC m⁻² day⁻¹ in Sect. 3.2 and −900 to −1500 gC m⁻² in Sect. 3.4 (p. 28-37, lines 614-617 and 778-780). At the same time, WUE is defined as GPP/ET and is always reported as positive. The authors should use one sign convention for GPP throughout the text, figures, and equations.
Second, some cumulative ET values do not match across sections. For example, FR-Lam ET is reported as about 610 kg H₂O m⁻² in the Abstract, 500-800 kg H₂O m⁻² in Sect. 3.4, and 800-1150 kg H₂O m⁻² in the Discussion (p. 2, lines 33-35; p. 37, lines 788-789; p. 45, lines 1014-1015). If these numbers refer to different time periods or calculation methods, this should be clearly explained. Otherwise, the values should be corrected. Overall, the authors should check all key GPP, ET, and WUE values in the Abstract, Results, and Discussion for consistency.
Minor comments:
- Please check the use of “correspondingly” in the Abstract. “respectively” would be more appropriate in most cases.
- Line 138: the citation “(Boone et al., 2017)” should be Boone et al., (2017).
- Please check whether “leaf are index” in the caption of Figure 4 should be “leaf area index”.
- The initial value of mesophyll conductance (gm) is given as 7.53 × 10⁻² m s⁻¹ in the text (Sect. 3.3) but as 7.53 × 10⁻³ m s⁻¹ in Table 5. Please reconcile.
- The unit for WUE is written inconsistently, e.g., "g C kg H₂O⁻¹" in Eq. (3) versus "gC kg⁻¹ H₂O" elsewhere. Please standardize unit and symbol notation throughout the text, equations, tables, and figures (including spacing, e.g., "gC" vs "g C").
- The manuscript would benefit from a thorough English-language edit; there are recurrent issues with articles, tense, and phrasing that occasionally impede readability.
Citation: https://doi.org/10.5194/egusphere-2026-3119-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 6 | 3 | 0 | 9 | 1 | 1 | 1 |
- HTML: 6
- PDF: 3
- XML: 0
- Total: 9
- Supplement: 1
- BibTeX: 1
- EndNote: 1
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript titled “Understanding ecosystem gross primary productivity, evapotranspiration, and water use efficiency of maize using ISBA-A-gs land surface model over temperate and tropical semi-arid climates” evaluates the performance of the ISBA-A-gs model in simulating key ecohydrological fluxes (, , and ) across two distinct climate regimes (temperate France and tropical semi-arid India). The authors successfully calibrated site-specific parameters to capture carbon-water dynamics under different irrigation practices. While the application of a process-based model to cross-climatic regimes is interesting and relevant, the manuscript is excessively lengthy and dense, which occasionally makes it difficult to understand the core findings. Streamlining the text and restructuring the presentation of results will significantly improve readability and impact.
The Introduction provides an excellent, comprehensive background on , maize agro-ecosystems, and the history of the ISBA model framework. However, it is excessively long, making it difficult to understand the overarching research gap until lines 196–201. I recommend streamlining the introduction to establish a more direct line of sight to the study's novel contributions. Furthermore, the alignment between the stated objectives and the results should be tightened. Specifically, it is not immediately clear how objectives (iii) and (iv) are systematically isolated and resolved in the text. Please restructure the Results and Discussion sections to explicitly map back to these objectives.
The Results section reads more like an extended narrative and contains text that belongs elsewhere. For example: Section 3.1 (Site Meteorology and Vegetation): This section describes the environmental forcing data rather than new experimental or model findings. To tighten the manuscript, this text should be significantly condensed and integrated into Section 2.1 (Experimental sites description) within the Materials and Methods section.
Section 3.3 (Model Calibration): This section is redundant as it essentially reiterates the exact numerical values already clearly presented in Table 5 and Table 6. The text should be condensed to focus strictly on the biophysical interpretation of why these parameters shifted between the sites and irrigation treatments, rather than repeating the data tables.
Line 444: The description of the optimization algorithm requires more technical clarity. The authors state that a "one-at-a-time" sensitivity analysis was performed followed by 50 simulations per parameter. It is unclear if the final parameter selection was achieved via automated iterative optimization or manual tuning. Please explicitly state the exact mathematical or algorithmic approach used to isolate the final "optimal values" to ensure reproducibility.
Figures 6 and 8: In the scatter plots comparing daily observations against simulations, the authors have plotted a best-fit regression line. While a 1:1 line allows the reader to visually assess under- or over-estimation (bias) immediately. I highly recommend modifying these figures to display a 1:1 line.
Line 199: Please explain what is meant by "climate-specific parameterization".
Minor correction: Please correct the narrative citation style. When the authors' names are part of the sentence structure, only the year should be in parentheses (e.g., change "developed by (Noilhan and Planton, 1989)" to "developed by Noilhan and Planton (1989)"). This occurs in several places in the text (e.g., Calvet et al., 1998; Boone et al., 2017).