Review Reports

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

Thank you for the author’s detailed response and tireless efforts. Overall, the resubmitted manuscript shows substantial improvements compared with the previous version. Nevertheless, several issues remain: inconsistent internal study data (precipitation for La Llave in 2024), missing data for FJ 6X6 T33, the optimal‑performing genotype based on the STI index, and the absence of the key recommended genotype G69 SEL 16 FM in Table 4. Although the authors are evidently aware of most limitations of this study, the following points need further addressing before acceptance:

1. Unify the 2024 precipitation values for the La Llave site. The 2024 precipitation data for La Llave are contradictory between Section 4.2 in the Methods and Table 1 in the Results (277.2 mm versus 377.2 mm). Please consult the original observational records from the La Llave weather station to confirm and standardize the cumulative precipitation for 2024. If the correct value is 277.2 mm, revise the statement that “precipitation doubled without a corresponding increase in yield”.
2. G69 SEL 16 FM (STI = 1.558, DSI = 0.785) is highlighted for recommendation in the Conclusions but is not included in Table 4. Table 4 omits two genotypes repeatedly mentioned in the Discussion and Conclusions: G69 SEL 16 FM and FJ 6X6 T33. Please add FJ 6X6 T33 and G69 SEL 16 FM to Table 4, and complete their GMP, STI, DSI metrics as well as yield data under the two water‑regime treatments. Alternatively, explicitly state in the main text the rationale for excluding them from the table. The threshold “rain‑fed yield > 1450 kg ha⁻¹” set in the Abstract should be consistent with the final list of selected genotypes.
3. The citation numbers for references related to Lin 68‑84 are out‑of‑sequence (e.g., reference [21] appears before [16‑20], and [28] appears before [19, 26, 27]). Please correct the citation numbering.

Author Response

We thank the reviewer for their constructive comments, which helped improve the manuscript.

Comments 1. Unify the 2024 precipitation values for the La Llave site (277.2 mm in Section 4.2 vs. 377.2 mm in Table 1). If the correct value is 277.2 mm, revise the statement that "precipitation doubled without a corresponding increase in yield".

Response 1: We thank the reviewer for detecting this inconsistency. We checked the original records of the La Llave weather station, and the correct value for 2024 is 377.2 mm, which corresponds to the sum of the monthly precipitation from July to October reported in Table 1 (99.6 + 105.6 + 164.8 + 7.2 mm). The value of 277.2 mm was a typographical error and has been corrected throughout the manuscript (Sections 2.1, 2.6, and 4.2), which now consistently reports 377.2 mm. Because the corrected value is about twice the 2023 total (377.2 vs. 186.8 mm), the statement that precipitation doubled without a proportional increase in yield remains valid and has been retained.

Comments 2. G69 SEL 16 FM and FJ 6X6 T33 are mentioned in the Discussion and Conclusions but are not included in Table 4. Please add them, with their GMP, STI and DSI values and yield under both water regimes. The threshold "rainfed yield > 1450 kg ha⁻¹" in the Abstract should be consistent with the final list of selected genotypes.

Response 2: Done. Table 4 now includes FJ 6X6 T33 (rainfed: 1,801.6 kg ha⁻¹; rainfed + rescue irrigation: 3,045.3 kg ha⁻¹; GMP = 2,342.3 kg ha⁻¹; STI = 2.169; DSI = 1.122) and G69 SEL 16 FM (1,677.6 and 2,349.1 kg ha⁻¹; GMP = 1,985.1 kg ha⁻¹; STI = 1.558; DSI = 0.785), together with their classification. The table now includes all eight genotypes with STI > 1.30. We also verified that the threshold stated in the Abstract (STI > 1.30 and rainfed yield > 1,450 kg ha⁻¹) corresponds exactly to the seven genotypes listed there. The only genotype with STI > 1.30 not included in that list is Tto 5 FJ-17 4X4 402 L-19 FJ (STI = 1.616), because its rainfed yield (1,440.2 kg ha⁻¹) falls below the threshold. The numerical format of Table 4 was also standardized.

Comments 3. The citation numbers in lines 68–84 are out of sequence (e.g., reference [21] appears before [16–20], and [28] appears before [19, 26, 27]).

Response 3: Done. References have been renumbered according to their order of first appearance in the text, and the reference list has been updated accordingly.

 

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Authors, 

This paper is very well written. Your work is vital to advancements in improved drought tolerant common bean. I have a few comments.

Table 2. I am curious why DM went from 80.1 (rainfed) to 90.1 ( rain + irrigation ) in 2023 to just the opposite direction in 2024. Could the irrigation management been a factor in this case?

Line 143 100SW was also highest in La Lave both years since they were not significantly different. So the sentence needs to be reworded.

Table 3. Even though significant the G and GxE were small in magnitude compared to E for DF, DM, 100SW. I think you mention this in the text. Seems you can ignore GxE for GY. 

Line 167 Even though highly significant, they are small in magnitude compared to E. 

Section 2.4 do you have data in a table to show where these numbers appear? 

Why did you use only a subset of your data to run your indices?  

Author Response

We thank the reviewer for their constructive comments, which helped improve the manuscript.

Comment 1. Table 2. I am curious why DM went from 80.1 (rainfed) to 90.1 (rain + irrigation) in 2023 to just the opposite direction in 2024. Could the irrigation management have been a factor in this case?

Response 1: We thank the reviewer for this observation, which allowed us to clarify the interpretation.

• 2023: physiological maturity at Celaya 2023 was recorded at the treatment level. Rainfed and rescue-irrigated plots reached maturity approximately 30 and 40 days after flowering, respectively, and these intervals were applied to the flowering date of each plot (Section 4.5.1). The 10-day difference is therefore a treatment-level observation, as now stated in Section 2.2. It is consistent with the acceleration of maturity under the September moisture deficit (14.7 mm) in the rainfed plots, whereas the rescue irrigation applied 55 days after sowing restored soil moisture during pod filling.
• 2024: we agree with the reviewer that an effect of irrigation management cannot be ruled out. Section 2.2 now states that the 2024 difference (96.7 vs. 92.4 days) was accompanied by high variability among genotypes (SD = 12.2 and 14.6 days), suggesting that genotypic phenological responses to the microenvironment of each treatment (G×E) contributed to it, although an effect of irrigation management cannot be excluded.

Comment 2. Line 143. 100SW was also highest in La Llave both years since they were not significantly different. So the sentence needs to be reworded.

Response 2: We thank the reviewer for this comment; however, we respectfully consider the sentence to be correct. Based on the least significant difference for comparing environment means (LSD₀.₀₅ = 1.1 g; Table 2), the two Celaya 2024 environments (30.9 and 30.8 g) did not differ from each other, but both exceeded La Llave 2024 (29.5 g; difference of 1.3–1.4 g) and La Llave 2023 (28.3 g; difference of 2.5–2.6 g). Both differences are larger than the LSD. In addition, the two La Llave environments differed from each other (1.2 g). The 100-seed weight was therefore significantly higher at Celaya 2024 than at La Llave. We retained the wording and updated the means according to Table S2.

Comment 3. Table 3. Even though significant, the G and G×E were small in magnitude compared to E for DF, DM, 100SW. I think you mention this in the text. Seems you can ignore G×E for GY.

Comment 4. Line 167. Even though highly significant, they are small in magnitude compared to E.

Response to Comments 3 and 4: We partially agree with the reviewer. A sentence was added to Section 2.3 acknowledging that, although significant for all variables, the G×E interaction was smaller in magnitude than the environmental effect for days to flowering, days to maturity, and grain yield. The pattern, however, is not the same for all variables:

• 100-seed weight: genotype was the main source of variation (56.0% of the total sum of squares), and G×E (20.8%) exceeded the environmental effect (16.5%).
• Days to maturity: the three sources contributed similar proportions (E = 32.7%, G = 30.4%, G×E = 31.3%).

We respectfully consider that G×E should not be ignored for grain yield. Although smaller than E (22.5% vs. 44.1%), it was nearly twice the genotypic effect (12.5%) and significant against the error term (F = 2.85, p < 0.01). This indicates changes in genotype ranking across environments, which is precisely what matters for selection: the environmental main effect affects all genotypes equally and does not alter selection decisions, whereas G×E determines whether the best genotype in one environment is also the best in another. This justifies the GGE biplot analysis (Section 2.7), which identified genotypes with broad adaptation (SCR 16) and specific adaptation (the P. SaltilloXTA lines at La Llave 2023). The magnitude of G×E observed here is also consistent with that reported for common bean in Mexico (23% in López-Salinas et al. [32]).

Comment 5. Section 2.4. Do you have data in a table to show where these numbers appear?

Response 5: Done. Section 2.4 now indicates the source table for each value:

• The yield increases under rescue irrigation (FM Anita +564.8%, Actopan +355.6%, Bayo INIFAP +320.0%, Bayo Berrendo +243.1%) are calculated from the yields reported in Table 4.
• The 2024 values now appear in the text with reference to Table S2: SCR 16 (2,879 kg ha⁻¹), FR21 G 69 11 L7 FJ (2,671 kg ha⁻¹), and G 9703 (2,546 kg ha⁻¹).
• Table S2 (Supplementary Material) contains the means of all 100 genotypes in the six environments, together with their drought tolerance indices.

Comment 6. Why did you use only a subset of your data to run your indices?

Response 6: We thank the reviewer for this question. GMP, STI, and DSI require comparing the yield of each genotype with and without stress at the same location and season, so that the difference reflects only moisture availability. Celaya 2023 was the only environment that met this condition, as explained in Section 4.6:

• Celaya 2024: no moisture stress occurred. Mean rainfed yield (1,333.8 kg ha⁻¹) slightly exceeded that under supplemental drip irrigation (1,289.6 kg ha⁻¹), giving a negative stress intensity index (SI = −0.034) for which DSI cannot be meaningfully interpreted.
• La Llave: both years were evaluated under rainfed conditions only, because the site lacks irrigation infrastructure, so no non-stressed control was available.

The remaining four environments were not discarded. They were used as an independent validation of the classification obtained at Celaya 2023: STI was significantly associated with mean genotype yield across those four environments (Spearman's ρ = 0.495, p < 0.001; Sections 2.5 and 4.7). In addition, the GGE biplot analysis integrates grain yield data from all six environments. The analysis therefore makes use of the complete dataset.

Round 2

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

I thank the authors for their careful responses and for addressing the reviewers' concerns and questions. In the revised manuscript, the authors have improved the relevant statements in the Results and Discussion in accordance with the review suggestions, and have also adopted more cautious and open-ended phrasing for the results and conclusions. Therefore, I agree to accept this manuscript.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,
The manuscript addresses a relevant agronomic problem: the identification of common bean genotypes with stable yield and adaptation to rainfed conditions in the Bajio region of Mexico. The evaluation of 100 genotypes across six environments, including contrasts between years, locations, and water management regimes, provides a potentially useful experimental basis for breeding programs. The approach of combining yield, drought-response indices, and GGE analysis is appropriate for studies of genotype-by-environment interaction.
However, the manuscript requires substantial improvement before it can be considered for publication. The main limitations relate to the consistency of the drought-tolerance calculations, the description and statistical analysis of the lattice design, the interpretation of the GGE biplot, the quantification of environmental stress, and the strength of some of the conclusions. I detail below the points that I consider essential.
Major Comments
1. Recalculate and thoroughly revise the DSI values and tolerance classifications.
This is the most critical issue in the manuscript. The authors define DSI using the formula cited in the manuscript and classify genotypes with DSI below 0.5 as highly tolerant. However, using the values reported in Table 4 and the mean stress values reported for Celaya 2023, the DSI values do not appear to be reproducible.
For example, when the reported stress and potential yields are entered into the stated DSI formula, the resulting value for GPO 69 1E 18 FM is approximately 1.15, not 0.304. Similarly, the recalculated value for Dalia is approximately 1.72, not 0.458.
This changes the biological interpretation: several genotypes with high STI and high GMP may be productive, but not necessarily less susceptible according to DSI. The classification of genotypes as highly tolerant in Table 4 must therefore be recalculated. If a different formula or a different reference mean was used, this must be explicitly stated. Since the abstract and conclusions depend on these indices, the entire manuscript should be updated after correcting this point. This is essential for the interpretability and scope of the results.
2. Conceptually separate productivity under stress, relative tolerance, and response to irrigation.
The manuscript tends to group three distinct phenomena under tolerance: high yield under rainfed conditions, low relative yield reduction, and strong response to irrigation. These are not equivalent. A genotype may show high yield under stress and still have a strong relative reduction compared with irrigated conditions. I recommend reorganizing the interpretation into three categories: absolute yield under stress, relative stability or susceptibility, and plasticity or positive response to favorable environments.
This is particularly important for genotypes such as Dalia, FM Anita, Actopan, and Bayo INIFAP, which show strong responses to irrigation but do not necessarily show relative tolerance to water deficit.
3. Water stress should be quantified more rigorously.
The manuscript interprets differences between years and sites as effects of water stress. However, the evidence presented mainly characterizes environmental conditions, such as cumulative precipitation, rainfall distribution, number of dry days, and response to irrigation, rather than the stress status of the plants. This distinction is important: stress is not an intrinsic property of the environment, but a physiological condition of the organism. An environment with low precipitation, high evaporative demand, or soil limitations may trigger stress in one species, genotype, or variety, but not necessarily in another. Therefore, the use of the term water stress requires direct or indirect evidence of the plant water and functional status, not only a description of the environment.
In the climatic section, the authors show clear differences in rainfall distribution between Celaya 2023 and Celaya 2024, especially during September, coinciding with sensitive reproductive stages. However, no parameters are reported that would allow physiological or biochemical confirmation that the plants were under water stress, such as soil moisture, leaf water potential, stomatal conductance, photosynthetic rate, canopy temperature, evapotranspiration, vapor pressure deficit, water balance, stress indices, relative water content, osmolyte accumulation, oxidative damage, or biochemical markers associated with water deficit.
The interpretation of the La Llave environments requires particular caution. In 2024, higher precipitation was recorded than in 2023, including 164.8 mm in September, but this increase was not translated into a proportional increase in yield. This suggests that the low yield performance at La Llave should not be attributed exclusively to drought or water stress. Edaphic and agronomic factors could also be involved, including soil texture, low organic matter, lower water-holding capacity, effective soil depth, crop establishment, nutritional limitations, heat stress, biotic pressure, or management differences.
For these reasons, I recommend that the authors reformulate the language of the manuscript and clearly distinguish among environments with limiting water conditions, low-precipitation environments, yield-restrictive environments, and plants subjected to water stress. In the absence of physiological, biochemical, or plant water-status parameters, it would be more appropriate to refer to limiting environmental conditions, contrasting rainfed environments, or potentially stress-inducing environments, avoiding the presentation of water stress as a demonstrated condition. Alternatively, if the authors wish to retain the term water stress, they should incorporate additional data documenting soil and/or plant water status during critical crop periods.
4. The comparison between the irrigated treatments in 2023 and 2024 is not fully equivalent.
In Celaya 2023, a single 30 mm rescue irrigation was applied by gravity irrigation, whereas in 2024 drip irrigation was used, with two weekly 10 mm events during the reproductive stage, totaling approximately 80 mm. These treatments differ not only in the amount of water applied, but also in frequency, application technology, and physiological timing. Therefore, comparisons between irrigated 2023 and irrigated 2024 should be treated with caution. I suggest naming the treatments more precisely as rainfed, rainfed plus rescue irrigation, and rainfed plus supplemental drip irrigation, or otherwise justifying why they can be analyzed as equivalent.
5. Revise the statistical analysis of the lattice design.
The manuscript states that a simple 10 by 10 lattice design with three replications per environment was used, but the analysis presented does not clearly indicate whether the means were adjusted for incomplete blocks, whether the efficiency of the lattice was evaluated against an RCBD, or how blocks were incorporated into the combined ANOVA. The model mentions the block effect nested within replication and environment, but Table 2 does not report the block effect or its degrees of freedom.
The authors should clarify:
- whether the results are based on raw or adjusted means;
- whether genotype was treated as fixed or random, and why;
- how environment, replication, block, and genotype-by-environment effects were estimated;
- whether BLUEs or BLUPs were used for the GGE analysis;
- whether multiple comparisons were applied to adjusted means;
- how the percentages of sum of squares reported in the text were calculated.
6. Do not interpret R2 as heritability.
The discussion states that phenotypic coefficients of determination suggest high heritability for phenological traits and seed weight. This interpretation is not appropriate. The model R2 is not equivalent to heritability. If the authors wish to discuss heritability, they should estimate variance components and calculate broad-sense heritability or repeatability across environments, ideally with confidence intervals. If heritability is not calculated, the statement should be removed or replaced with a more conservative interpretation regarding model precision or the proportion of explained variation.
7. Revise the interpretation of the GGE biplot.
The GGE biplot is useful, but the current interpretation needs greater precision. The manuscript states that genotypes close to the origin have broad stability and that SCR 16 shows broad adaptation because of its extreme position on PC1. However, in GGE analysis, the interpretation of stability and mean performance requires specific views, such as average environment coordination and mean versus stability. Proximity to the origin may indicate a low contribution to the interaction, but it may also correspond to non-outstanding mean yield. In addition, the caption indicates genotype-centered scaling, whereas the usual GGE approach is based on environment-centering to remove the main environmental effect.
I recommend that the authors specify the type of centering, scaling, and singular-value partitioning; use adjusted means by environment; include a which-won-where biplot; include a mean versus stability biplot; avoid declaring definitive mega-environments with only two years of data; and use tentative environment groupings rather than mega-environments, unless additional cycles or independent validation are available.
8. Correct the erroneous reference to Figure 1 in the climatic section.
In Section 2.1, the manuscript refers to Table 1 and Figure 1 when discussing climatic conditions, but Figure 1 corresponds to the GGE biplot, not to a climatic figure. The authors should either add an actual climatic figure or remove this reference.
9. Moderate the conclusions regarding interspecific breeding.
The results obtained with lines derived from Phaseolus acutifolius are interesting, especially under severely limiting conditions at La Llave. However, the manuscript uses language that is too strong, such as validating the value of interspecific breeding or validation of interspecific breeding. With the current data, the study supports the evaluation of these lines as promising resources, but it does not validate introgression mechanisms or physiological tolerance. I recommend using more cautious wording, such as supports the potential value, suggests that these lines merit further testing, or provides field-based evidence that selected interspecific-derived lines may be useful under severely limiting environments.
10. Include complete genotype-by-environment data as supplementary material.
The current supplementary material appears to be limited to the description of the 100 genotypes. However, for reproducibility and scientific value, the authors should include a complete table with yield, days to flowering, days to maturity, and 100-seed weight for each genotype in each environment, ideally with adjusted means, standard errors, and environment codes. The data availability statement indicates that the original contributions are included in the article and supplementary material, but the complete phenotypic data do not appear to be available in the revised version.
11. Review the consistency of genotype names.
There are several inconsistencies among the abstract, tables, text, and supplementary material. For example, the abstract uses FM-69 1E-18 and FJ-T9-61, whereas the tables list them as GPO 69 1E 18 FM and Trat 9 SEG 61 FJ. Variants such as P. Saltillo-245 and P. SaltilloXTA-245 also appear. These names should be strictly standardized, because the manuscript has applied value for breeding and germplasm registration.
12. Review Table 3: possible errors in standard deviation or standard error.
Table 3 reports, for example, 100-seed weight of 30.8 plus or minus 0.1 for Celaya rainfed plus irrigation 2024, which seems unusually low if it corresponds to the standard deviation among 100 genotypes. The authors should clarify whether the values are SD, SE, or residual error. If they are standard deviations among genotypes, some values appear unlikely and should be verified.
13. Revise Table 5 and the selection criteria.
Table 5 combines heterogeneous selection criteria: high yield, drought tolerance, response to irrigation, broad stability, extreme adaptation, grain quality or size, and earliness. The table is useful, but it should be based on reproducible and corrected criteria. In particular, the broad stability category should be supported by a formal stability indicator, not by DSI or by presence in the top 20 without showing the underlying data. In addition, the note on Dalia is confusing and should be reformulated after recalculating the indices.
14. Avoid excessive methodological claims.
The conclusion states that the integrated framework may serve as a model for other multi-environment evaluation programs in legumes. Since the approach combines already established tools, namely tolerance indices and GGE biplot analysis, the methodological contribution should not be presented as a robust framework or general model, but rather as a useful regional application. The main value of the study is the multi-environment dataset and the preliminary identification of promising genotypes, not methodological innovation.
Minor Comments
1. Change multi-environmental evaluation to multi-environment evaluation.
2. Use either Bajio or El Bajio consistently, rather than alternating between the two without a clear rationale.
3. Remove unnecessary quotation marks around La Llave after the first mention.
4. Review unit formatting: kg ha-1, degrees C, mm, and 100-seed weight.
5. Ensure that Phaseolus vulgaris and Phaseolus acutifolius are italicized throughout the manuscript.
6. Remove the Patents section if it does not apply, or state Not applicable according to the journal format.
7. Revise the phrase phenological sensitive stages; it should be phenologically sensitive stages.
8. Review redundant phrases such as nutritional limitations and water availability in the discussion.
9. Include a brief explanation of why GMP, STI, and DSI were selected instead of other complementary indices.
10. If Tukey HSD is retained, specify exactly which comparisons it was used for, since with 100 genotypes across 6 environments it may not be the most informative strategy.
11. Improve the resolution and readability of Figure 1; the genotype labels are small and the panel is overcrowded.
12. Clearly separate in the discussion what was directly observed in this study from what is inferred from previous physiological literature.
In summary, the manuscript has applied potential, but it requires substantive corrections. The absolute priority is to recalculate the tolerance indices, revise the statistical model, correct the interpretation of the GGE biplot, and moderate the conclusions. Once these issues are resolved, the study could represent a useful contribution to common bean breeding programs targeting rainfed environments and variable water-limiting conditions.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript, the authors analyzed the yield stability and adaptability of 100 common bean genotypes across six rain-fed environments in Mexico’s Bajío region, using drought tolerance indices (GMP/STI/DSI) and GGE biplot. A comprehensive multi-environment evaluation framework was established, and two extremely drought-tolerant interspecific lines were screened out. Reproductive-stage precipitation timing, rather than total rainfall, was verified as the key yield-limiting factor, and three breeding subregions were classified. Despite evaluating only basic agronomic traits without exploring drought-related physiological and molecular mechanisms, this established multi-environment evaluation method can provide references for stress-resistance breeding of legumes and other crops. Therefore, the study possesses certain theoretical and practical value; however, several deficiencies remain to be improved.
Abstract: The authors are suggested to add the following statement: "The research area was divided into three distinct major breeding environments via GGE biplot, clarifying the suitable planting regions for different genotypes."
Introduction: The introduction lacks detailed local production information of the Bajío region, including planting area, yield level, drought frequency and economic losses, as well as the drought tolerance defects of local dominant cultivars. It is recommended to supplement the above agricultural background and clarify that the combination of multiple indices and GGE biplot is superior to a single drought index for accurately screening high-yield and drought-tolerant genotypes.
Results: It is advised to supplement GGE stability parameters (e.g., genotypic stability variance), explain potential sources of residual variation, and provide the 2024 drought tolerance data as well as two-year correlation analysis to validate data reliability.
Discussion: The non-promotive yield effect of increased precipitation in La Llave requires further explanation. Soil nutrient data should be supplemented to analyze the interactions between low organic matter, high sand content and water stress.
In addition, this study was merely conducted at two sites in the Bajío region over a two-year trial period, with only water stress assessed, resulting in single-stress evaluation and insufficient verification of long-term genetic stability. The authors should add a section of research limitations to explicitly clarify the deficiencies.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript evaluates 100 common bean genotypes across six environments in Mexico’s Bajío region using drought tolerance indices, multienvironment trials, and a GGE biplot to identify genotypes with broad or specific adaptation under rainfed conditions. The study is timely, methodologically relevant, and supported by extensive field data. The environmental characterization is strong, and the integration of indices such as STI, GMP, and DSI adds value. However, several issues related to clarity, methodological transparency, statistical interpretation, and presentation should be addressed to strengthen the manuscript.

1. The manuscript provides a general description of the 10×10 simple lattice design and the six environments, but the explanation lacks sufficient detail for full reproducibility. The authors do not clearly specify the exact plot size, plant density, fertilizer management, or pest and disease control practices, all of which can influence yield under rainfed conditions. The description of the supplemental irrigation regime is also incomplete, as the timing, volume, and method of irrigation are not explicitly stated.
2. Although the authors use GMP, STI, and DSI, the rationale for choosing these specific indices over others commonly used in drought research is not adequately discussed. The manuscript would benefit from a clearer explanation of why these indices were prioritized and how they complement each other in capturing different dimensions of drought tolerance. Additionally, the authors do not address the limitations of DSI, which can be sensitive to environmental mean yield and may misclassify genotypes under heterogeneous stress intensities.
3. The GGE biplot analysis is presented, but the interpretation remains superficial and does not fully exploit the analytical potential of the method. The authors mention three mega‑environments but do not discuss the biological or agronomic implications of these clusters. The manuscript would be strengthened by a deeper explanation of why certain environments grouped together, how this relates to rainfall patterns or soil characteristics, and what this means for breeding strategies targeting specific adaptation. The discussion should also address the stability vs. responsiveness trade‑off more explicitly.
4. The manuscript frequently compares mean yields across environments but does not provide measures of variability such as confidence intervals for environmental means. For example, the statement that supplemental irrigation increased yield by 57.3% in 2023 is compelling, but without variability estimates, it is unclear whether this difference is statistically meaningful or influenced by outlier genotypes.
5. The authors state that La Llave environments were the most limiting and that certain genotypes showed specific adaptation, yet the discussion does not adequately explain why 100‑seed weight remained relatively high under severe stress. The manuscript attributes this to compensatory effects but does not provide supporting evidence or references.
6. Several genotype names appear inconsistent or unclear, such as “FR21 G 69 21 L7 FM” and “Trat 9 SEG 61 FJ,” which may confuse readers unfamiliar with the breeding program. The manuscript would benefit from a standardized naming convention or a supplementary table explaining the origin, pedigree, and classification (landrace, elite line, interspecific line) of each genotype. This is especially important for the interspecific lines derived from Phaseolus acutifolius, which are highlighted as key findings.
7. The manuscript briefly mentions that La Llave has a Haplic Phaeozem with low organic matter, but this important factor is not integrated into the interpretation of results. Soil texture, water‑holding capacity, and fertility strongly influence drought response, and the authors should discuss how these edaphic differences contributed to the observed genotype × environment patterns.
8. Although the introduction references recent advances in understanding physiological responses to drought, the discussion does not connect these mechanisms to the observed field results. For example, traits such as pod retention, root architecture, or osmotic adjustment are not discussed in relation to the high‑performing genotypes.
9. The introduction would benefit from a concise overview of the regulatory mechanisms and breeding strategies underlying crop drought resistance. Consult and cite this article (https://doi.org/10.1016/j.ncrops.2024.100029). Including a brief description of key molecular, physiological, and genetic pathways, together with established and emerging breeding approaches for improving drought tolerance, would provide essential context and strengthen the scientific foundation of the study. This also partly applies to the discussion section. Strengthening this link would enhance the biological relevance of the findings and provide clearer guidance for breeders.
10. The discussion section tends to emphasize positive findings without sufficiently addressing limitations. For example, the authors highlight the strong performance of certain genotypes but do not discuss potential trade‑offs such as late maturity, susceptibility to diseases, or variability across years. A more balanced discussion acknowledging uncertainties, limitations of the dataset, and potential confounding factors would improve the scientific rigor of the manuscript.

The manuscript presents valuable multienvironment data and identifies promising genotypes for drought‑prone regions, making it a potentially strong contribution to common bean breeding research. However, several aspects require improvement, including clearer methodological descriptions, deeper interpretation of GxE interactions, stronger physiological grounding, and more polished presentation of results. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Yield Stability and Adaptation of Common Bean (Phaseolus vulgaris L.) Genotypes in Rainfed Environments of the Bajío, Mexico presents an interesting study and followings are to be considered for improving the  manuscript.

This study based on 100 genotypes, for a G×E analysis, the population is very small and conclusions that have been drawn using the results could be highly biased. Without having a validation population, the result can’t be confirmed. This is the major drawback I have seen in this study. Authors may need to include these limitations in the discussion.

Other than that, materials and methods:

Please provide the coordinates of the locations.

Kindly provide detail information of how you conducted variance analysis. What are the software and versions.

The principal component analysis needed to be explained clearly.

In the results, if is good, if you can provide information about variation of phenotypic data using density pots to understand the diversity of the genotypes.

Before providing the results of variance analysis, if the descriptive statistics were included it is easy to identify the mean values of the traits and the variations.

Depending on the experiment design large proportion of residual variance can be detected. can you include that output in the result section.

Discussion can be improved by discussing your results with the existing literature

 

 

 

Comments on the Quality of English Language

English can be improved by maintaining the flow of the information. for i.e. in the introduction, authors talked about the genetic diversity and then straightly moved to the multi-environmental evaluations.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Comments to the Authors
Dear Authors,
I have carefully reviewed the manuscript entitled "Yield Stability and Adaptation of Common Bean (Phaseolus vulgaris L.) Genotypes in Rainfed Environments of the Bajío, Mexico". The study addresses an agronomically relevant question for common bean improvement in the Bajío region, particularly the identification of genotypes with productive performance, adaptation, and stability under variable water availability. The evaluation of a large set of genotypes, the comparison across contrasting seasons and sites, and the integration of drought-response indices with GGE biplot analysis provide a potentially useful basis for breeding-oriented interpretation.
The manuscript has several strengths, including the size of the evaluated germplasm panel, the use of multi-environment field data, and the effort to distinguish broad adaptation, specific adaptation, and response to water management. However, several methodological, statistical, and interpretive issues need to be addressed so that the conclusions remain fully aligned with the experimental design and the analyses actually performed. My comments below are intended to help strengthen the manuscript, clarify the scope of inference, and improve reproducibility.
Major Comments
1. Alignment between title, abstract, and experimental scope
The title and parts of the abstract emphasize rainfed environments, but the study also includes rescue irrigation or supplemental irrigation treatments at Celaya. The manuscript should more clearly reflect this structure in the title, abstract, objectives, and conclusions. The comparison of water regimes is mainly structured at Celaya, whereas La Llave appears to have been evaluated under limiting rainfed conditions. This distinction should be made explicit early in the manuscript to avoid giving the impression that all six environments correspond strictly to rainfed production.
The authors should also reassess broad statements about regional recommendation or adaptation. The dataset is valuable, but the experimental network is limited to two sites and two growing seasons; therefore, regional extrapolation should be framed cautiously.
2. Consistency between the declared field design and the statistical analysis
A central methodological issue is the relationship between the declared field design and the statistical model used for analysis. The manuscript states that a "10×10 simple lattice design with three replications per environment" was used, but the statistical analysis is described as a randomized complete block model with replications nested within environments. This apparent mismatch must be clarified.
The authors should explain whether the incomplete-block lattice structure was incorporated into the analysis or whether it was used only as a field arrangement. If the final analysis was performed as an RCBD, the rationale for not using the lattice structure should be provided. In a field trial with 100 genotypes, this point is important because the handling of spatial heterogeneity may affect genotype means, ranking, and subsequent GGE interpretation.
The manuscript should also be checked for consistency regarding the use of unadjusted means, adjusted means, BLUEs, or any other estimator. The statistical methods, table footnotes, figures, and supplementary material should all describe the same analytical basis.
3. Statistical model, inferential scope, and multiple comparisons
The statistical methods should provide a clearer justification for treating genotype and environment as fixed effects. This may be appropriate if the authors intend to compare only the specific genotypes and environments evaluated. However, if the manuscript aims to support broader inference about adaptation within the Bajío or other rainfed systems, the limitations of a fixed-effect framework should be acknowledged or the analysis should be complemented accordingly.
The use of Fisher’s LSD for comparisons among 100 genotypes should also be reconsidered or explicitly justified. Given the number of possible comparisons, the authors should address the risk of overinterpretation. The manuscript would benefit from a clearer explanation of how statistical significance, agronomic relevance, and genotype selection were balanced.
4. Interpretation of drought-response indices
The use of GMP, STI, and DSI is appropriate for describing genotype performance under contrasting moisture conditions. However, these indices appear to have been calculated from the Celaya 2023 rainfed versus rescue-irrigation contrast. Their interpretation should therefore be restricted to that specific stress scenario.
The authors should avoid presenting these indices as a general or stable classification of drought tolerance across all environments. The other environments do not reproduce the same water-regime contrast, and the irrigation treatments differ between years in method, volume, and timing. The manuscript should consistently distinguish between a field-based response under a particular contrast and broader drought tolerance.
The supplementary material should also allow readers to trace the full index calculation across all evaluated genotypes, not only the selected outstanding materials. This would strengthen transparency and make the selection of highlighted genotypes easier to assess.
5. Distinguishing high yield, relative tolerance, and irrigation responsiveness
The manuscript makes an important distinction between high absolute yield, lower proportional susceptibility, and strong response to irrigation. This distinction is agronomically useful, but it should be applied consistently throughout the abstract, results, discussion, and conclusions.
Some genotypes appear to maintain high absolute yield under limiting conditions but also show a larger proportional yield reduction relative to the population mean. Others combine competitive yield with lower relative susceptibility. These categories should not be conflated. The manuscript should avoid implying that high yield under stress is necessarily equivalent to drought tolerance unless the supporting index or analysis justifies that interpretation.
6. GGE biplot interpretation and environmental groupings
The GGE biplot analysis is a useful component of the manuscript. Nevertheless, the interpretation of environmental groupings should be more cautious. The term "mega-environments" may be too strong for a dataset comprising two sites, two years, and water treatments that are not fully balanced across locations.
The authors should consider presenting these patterns as preliminary environmental groupings or response patterns rather than definitive mega-environments. Similarly, claims about suitable planting regions should be moderated unless they are supported by a broader testing network.
The interpretation of genotype stability from the GGE biplot should also be handled carefully. If stability is a central conclusion, it would be useful to provide additional stability metrics or, at minimum, to clarify that the GGE interpretation is exploratory and based on the evaluated environments.
7. Additional support for genotype stability and adaptation claims
Several genotypes are identified as broadly adapted, specifically adapted, or stable. These claims would be stronger if supported by additional metrics or tabular evidence. The authors should consider including a genotype-by-environment ranking table, coefficients of variation by genotype, or established stability parameters.
If the authors choose not to add further stability analyses, the language around stability and broad adaptation should be adjusted so that it does not exceed what can be inferred from the GGE biplot and the available environments.
8. Interpretation of La Llave as a multi-stress environment
The interpretation of La Llave is one of the interesting aspects of the manuscript. The consistently low yield despite increased precipitation in 2024 suggests that additional constraints may have contributed to poor productivity. The discussion of alkaline pH, low organic matter, low iron availability, and limited phosphorus availability is plausible and agronomically relevant.
However, the design does not experimentally separate the effects of water deficit, soil constraints, nutrient availability, and temperature. The authors should therefore distinguish clearly between direct evidence, agronomic inference, and mechanistic hypothesis. Any causal language regarding edaphic constraints should be carefully moderated unless supported by additional measurements.
The methods should also provide more detail about the soil analysis, including sampling depth, timing, number of subsamples, analytical methods, and laboratory or institutional source. This information is important because the interpretation of La Llave relies substantially on soil constraints.
9. Agronomic management and possible confounding factors
The crop-management section should be expanded or clarified. The manuscript mentions herbicide applications, mechanical/manual control, and insecticide applications. The authors should describe timing, doses, uniformity across environments, and whether phytotoxicity or differential pest/weed pressure was observed.
This is relevant because yield differences in field trials may be influenced by factors other than moisture availability or genotype. The manuscript should clarify whether these factors were adequately controlled or whether they may have contributed to the observed variation.
10. Reproducibility and data availability
The conclusions depend on ANOVA, genotype × environment interaction, drought-response indices, and GGE biplot analysis. Therefore, the level of data availability should be sufficient to reproduce the main analyses.
The authors should review whether the supplementary material provides enough information for reproduction. Ideally, data should be available at the plot or replication level, with identifiers for environment, replication, block or incomplete block if applicable, genotype, and measured traits. Analytical code or a detailed analytical workflow would also improve transparency.
If only genotype means are provided, the authors should explain the limitations this imposes on reproducibility and specify exactly how those means were obtained.
11. Genotype nomenclature and traceability
The manuscript should be checked carefully for consistency in genotype names. Some key genotypes appear with similar but non-identical names across the abstract, text, tables, and figures. This may affect traceability and could create confusion for readers or breeders interested in the highlighted materials.
The authors should verify all genotype names against a master list and ensure that each genotype has one unique identifier used consistently throughout the manuscript and supplementary material.
12. Figure quality and readability
The GGE figures are central to the manuscript and should be improved for readability. Some genotype and environment labels are difficult to read, and some plotted points or vertices are not sufficiently informative without a clear key.
The authors should improve resolution, font size, contrast, abbreviation consistency, and label strategy. If numerical genotype codes are retained in any figure, the supplementary material should provide an unambiguous correspondence between codes and genotype names.
13. Literature support and mechanistic interpretation
The discussion includes several statements on genotype × environment interaction, drought physiology, Phaseolus acutifolius traits, seed-weight compensation, and stability. The authors should verify that each reference directly supports the point for which it is cited.
Mechanistic explanations should be clearly distinguished from traits measured in the present study. For example, if physiological mechanisms such as stomatal regulation, osmotic adjustment, or root architecture were not measured, they should be discussed as plausible interpretations from previous literature rather than as direct findings of the present work.
14. Editorial consistency and manuscript structure
The manuscript would benefit from a final editorial revision for consistency of units, abbreviations, scientific names, genotype names, table headings, figure captions, and journal-template elements. The discussion should also be reviewed to reduce repetition and ensure that the limitations are directly reflected in the conclusions.
Minor Comments
•    Check consistency of yield, seed-weight, precipitation, and temperature units throughout the manuscript.
•    Use environment abbreviations consistently across text, tables, figures, and supplementary material.
•    Ensure that all genotypes highlighted in the abstract are clearly traceable in the main or supplementary tables.
•    Verify that all numerical values in the abstract match the corresponding values in tables and results.
•    Clarify the source of climate data and the spatial proximity of weather stations to each experimental site.
•    Distinguish clearly between rescue irrigation and supplemental drip irrigation when interpreting irrigation response.
•    Avoid extrapolating the Celaya 2023 stress contrast as a general drought-tolerance ranking.
•    Review the use of terms such as "drought tolerance", "stability", "broad adaptation", and "mega-environments" so that each term is supported by the relevant analysis.
•    Strengthen the connection between the limitations section and the final conclusions.
•    Ensure that supplementary tables are clearly labeled, complete, and cited at the appropriate points in the main text.
Overall Comment to the Authors
The manuscript addresses a relevant problem and contains a dataset with clear agronomic value. To strengthen the scientific contribution, the authors should focus on the coherence between field design and statistical analysis, the inferential limits of the drought-response indices, the interpretation of environmental groupings, the reproducibility of the analysis, and the consistency of genotype identification. A revised version should ensure that all conclusions are strictly supported by the experimental design and the analyses presented.

Author Response

Comments 1: Alignment between title, abstract, and experimental scope

"The title and parts of the abstract emphasize rainfed environments, but the study also includes rescue irrigation or supplemental irrigation treatments at Celaya. The manuscript should more clearly reflect this structure in the title, abstract, objectives, and conclusions."

Response 1: We agree with this observation. The title, abstract, introduction, and conclusions have been revised to accurately reflect the experimental structure, which includes both rainfed and supplementally irrigated conditions at Celaya and exclusively rainfed conditions at La Llave.

Changes made:

• Title revised from "...in Rainfed Environments..." to "...under Contrasting Moisture Regimes..."
• Abstract: We added “two moisture regimes at Celaya (rainfed and rainfed with supplemental irrigation)" and “rainfed and water-limited conditions”.
• Introduction (last paragraph): revised to “including rainfed and supplementally irrigated conditions at Celaya and exclusively rainfed conditions at La Llave"
• Conclusions (first paragraph): revised from "under field rainfed conditions" to "under contrasting moisture conditions"

Comments 2: Consistency between field design and statistical analysis

"The manuscript states that a 10×10 simple lattice design was used, but the statistical analysis is described as a randomized complete block model. This apparent mismatch must be clarified."

Response 2: We thank the reviewer for this observation. We clarify that the 10×10 simple lattice was used as the field arrangement to facilitate management of the large number of genotypes during sowing, data collection, and harvesting. However, the combined multi-environment statistical analysis was conducted using a randomized complete block design (RCBD) model with replications nested within environments, as implemented in PROC GLM (SAS v9.4). The incomplete-block structure was not incorporated into the analytical model. This approach was adopted because (i) with 100 genotypes evaluated across 6 environments with 3 replications (1,800 observations total), the RCBD model provided 1,188 degrees of freedom for the error term, ensuring robust and stable estimates of genotypic means; and (ii) for combined multi-environment analyses with selection objectives, the RCBD with fixed genotype and environmental effects is a standard and well-supported analytical approach in plant breeding trials. Unadjusted genotypic means were used consistently in all comparisons and as inputs for the GGE biplot analysis.

Changes made: The distinction between the field layout and the statistical model has been explicitly clarified in Sections 2.3, 4.3, and 4.7:

"Although the field layout followed a 10×10 simple lattice arrangement for operational management purposes, the incomplete-block structure was not incorporated into the statistical model; unadjusted genotypic means were used in all comparisons."

The footnote of Table 3 was also updated to state the following:

"Unadjusted genotypic means (environment-specific) were used as input for GGE biplot analysis.".

 

Comments 3: Statistical model, inferential scope, and multiple comparisons

"The statistical methods should provide a clearer justification for treating genotype and environment as fixed effects. The use of Fisher's LSD for comparisons among 100 genotypes should also be reconsidered or explicitly justified."

Response 3: We agree and have added explicit justifications in Sections 2.3 and 4.7.

Genotype and environment were treated as fixed effects because the objective of the study was to characterize the specific materials and environments evaluated, not to make inferences about a broader population of genotypes or environments. This is a standard approach in plant breeding trials focused on direct selection.

With respect to the LSD test with 100 genotypes, we acknowledge that with a large number of comparisons, the LSD increases the risk of Type I error in formal pairwise inference. We therefore explicitly stated that LSD comparisons were used as an exploratory tool to identify outstanding materials and that genotype selection for breeding purposes was based primarily on agronomic thresholds and the integrated interpretation of tolerance indices and GGE biplot results rather than on statistical significance alone.

Changes made: The following text was added to Sections 2.3 and 4.7:

“Genotype was treated as a fixed effect to allow direct mean comparisons among genotypes for selection purposes, as the objective of the study was to characterize the specific materials evaluated rather than to make inferences about a broader population of genotypes. Environment was likewise treated as a fixed effect, consistent with the goal of describing genotypic responses in the specific target environments included in the trial."

"Given the large number of genotypes evaluated (n = 100), LSD comparisons were used as an exploratory tool to identify outstanding materials rather than to make formal pairwise inferences among all possible combinations. Genotype selection for breeding purposes was based primarily on agronomic thresholds and the integrated interpretation of tolerance indices and GGE biplot results, rather than on statistical significance alone."

 

Comments 4: Interpretation of drought-response indices

"These indices appear to have been calculated from the Celaya 2023 rainfed versus rescue-irrigation contrast. Their interpretation should be restricted to that specific stress scenario."

Response 4: We fully agree. Throughout the revised manuscript, we have added explicit statements restricting the interpretation of GMP, STI, and DSI to the Celaya 2023 rainfed versus rescue irrigation contrast.

Changes made: Qualifying statements added to Sections 2.5, 3.3, and 4.6 and the Conclusions section. Representative examples:

"These indices reflect genotypic responses under this specific stress scenario and should not be interpreted as a general or stable drought tolerance ranking across all environments." (Section 2.5)

"These indices were calculated from a single stress contrast (Celaya 2023) and should not be extrapolated as stable tolerance rankings across environments or seasons." (Section 3.3)

"These indices reflect genotypic responses under the specific stress contrast available at Celaya 2023 and should not be interpreted as stable drought tolerance rankings without replication across additional stress events and seasons." (Section 4.6)

 

Comments 5: Distinguishing high yield, relative tolerance, and irrigation responsiveness

"The manuscript makes an important distinction between high absolute yield, lower proportional susceptibility, and strong response to irrigation. This distinction should be applied consistently throughout."

Response 5: We agree that this distinction is critical and have applied it consistently throughout the revised manuscript. Genotypes with high STI/GMP are explicitly described as having high absolute productive capacity, while DSI is consistently described as capturing relative susceptibility to yield reduction. The term "drought tolerance" has been replaced by "relative stress tolerance" or "relative moisture-deficit tolerance" where it referred specifically to DSI-based classification.

Changes made: Revised throughout Sections 2.5, 2.8, 3.2, 3.3, and Conclusions. A representative addition in Section 3.2:

"This distinction is agronomically relevant, as high absolute yield under stress does not necessarily coincide with low relative susceptibility to moisture deficit."

And in Conclusions:

"These two performance dimensions (high absolute yield and low relative susceptibility) are distinct and should be considered jointly when selecting materials for rainfed breeding programs."

 

Comments 6: GGE biplot interpretation and environmental groupings

"The term 'mega-environments' may be too strong for a dataset comprising two sites, two years, and water treatments that are not fully balanced across locations. The authors should consider presenting these patterns as preliminary environmental groupings."

Response 6: We agree entirely. All references to "mega-environments" have been replaced by "preliminary environmental groupings" or "preliminary environmental response patterns" throughout the manuscript, with explicit qualifications that these groupings are tentative pending validation across additional sites and seasons.

Changes made: Revised in the Abstract, Sections 2.7 and 3.4, and Conclusions. Representative text:

"The analysis revealed three preliminary environmental groupings, considered tentative given the limited number of sites and seasons evaluated, and pending validation across additional cycles." (Section 2.7)

"These groupings should be interpreted as exploratory response patterns based on the evaluated environments rather than as definitive mega-environments." (Section 2.7).

 

Comments 7: Additional support for genotype stability and adaptation claims

"Several genotypes are identified as broadly adapted or stable. These claims would be stronger if supported by additional metrics or tabular evidence."

Response 7: We acknowledge this limitation. Given the scope of the current dataset (two sites, two seasons), we chose to moderate the language around stability and broad adaptation rather than add formal stability parameters, which would require additional environments to be meaningful. All references to "broad adaptation" now include the qualifier "across the evaluated environments" to clarify the inferential scope.

Changes made: Language was moderated throughout Sections 2.7, 2.8, 3.2, and 3.4, and Conclusions. For example:

"...suggesting broad adaptation across the evaluated environments." (Section 2.7)

"...supporting the characterization of SCR 16 as a broadly adapted genotype within the evaluated network." (Section 3.4)

 

Comments 8: Interpretation of La Llave as a multi-stress environment

"The design does not experimentally separate the effects of water deficit, soil constraints, nutrient availability, and temperature. Causal language should be moderated. The methods should also provide more detail about the soil analysis."

Response 8: We agree. All causal language regarding edaphic constraints at La Llave has been moderated to reflect agronomic inference rather than experimental evidence. Additionally, we have added methodological details for the soil sampling and analysis in Sections 2.6 and 4.2.

Changes made:

Section 2.6 — language moderated:

"...suggesting that factors beyond precipitation may have constrained productivity independently of moisture availability; the alkaline soil pH (8.11), low organic matter content (1.63%), and limited iron availability (2.07 ppm) of the Haplic Phaeozem soil are consistent with this interpretation, although their independent contributions to yield limitation could not be quantified within the current experimental design."

The soil sampling details have been added to Section 2.6:

"Soil samples were collected at 0–30 cm depth prior to sowing at each site, with five subsamples per location composited into a single representative sample per site; analyses were performed by the soil laboratory of INIFAP-CEBAJ following standard methods (pH in water 1:2, organic matter by Walkley-Black, available Fe by DTPA extraction)."

The full soil analytical methods have been added to Section 4.2:

"Soil characterization was based on composite samples collected at 0–30 cm depth prior to sowing, with five subsamples per site composited into a single representative sample; analyses were performed by the soil laboratory of INIFAP-CEBAJ following standard methods (pH in water 1:2, organic matter by Walkley-Black, available Fe by DTPA extraction, available P by the Bray-Kurtz method, and cation exchange capacity by ammonium acetate at pH 7.0)."

Explicit inference disclaimer added to Section 3.1:

"However, the experimental design did not independently quantify the contributions of individual edaphic factors to yield limitation, and these interpretations should be considered as agronomic inference rather than direct experimental evidence."

A new citation was added to support the known effect of alkaline pH on iron availability in calcareous soils: López-Millán et al. (2013) [Reference 31].

 

Comments 9: Agronomic management and possible confounding factors

"The crop-management section should be expanded or clarified, including timing, doses, uniformity across environments, and whether phytotoxicity or differential pest/weed pressure was observed."

Response 9: We have expanded Section 4.4 to include the application timing for herbicides, confirmation of uniform management across environments and years, and explicit statements regarding the absence of phytotoxicity and differential weed interference.

Changes made: The following statements were added to Section 4.4:

"The same fertilization rate and formulation were applied uniformly across all environments and years."

"These [preemergence herbicide] applications were made at 3–5 days after sowing at both sites and in both years."

"Weed pressure was effectively controlled across all environments, and no differential weed interference among genotypes or environments was observed that could have confounded yield comparisons."

"The insecticide rotation protocol was applied uniformly across all environments; no phytotoxicity symptoms attributable to pesticide applications were observed in any environment."

The dose of cypermethrin, which was previously unspecified, was also added (25% a.i., 300 mL ha⁻¹).

 

Comments 10: Reproducibility and data availability

"The level of data availability should be sufficient to reproduce the main analyses. Ideally, data should be available at the plot or replication level."

Response 10: We have addressed this comment by clarifying the nature of the data provided in the Supplementary Material. Table S2 provides unadjusted genotypic means for all 100 genotypes across the six environments for all four evaluated traits. We have updated the description of Table S2 in the Supplementary Materials section and added an explicit statement in Section 4.7 specifying how these means were obtained:

"Genotypic means reported in Table S2 correspond to unadjusted arithmetic means across three replications within each environment, calculated from individual plot observations using PROC MEANS in SAS v9.4."

The description of Table S2 in the Supplementary Materials section was also updated to:

"Table S2: Unadjusted genotypic means (arithmetic means of three replications) for grain yield (kg ha⁻¹), days to flowering, days to maturity, and 100-seed weight for each of the 100 common bean genotypes across the six evaluation environments."

 

Comments 11: Genotype nomenclature and traceability

"The manuscript should be checked carefully for consistency in genotype names."

Response 11: All genotype names have been verified against the master list in Table S1 and are now consistent throughout the manuscript, tables, figures, and supplementary material. The following statement was added to Section 4.1:

"The complete list of genotypes, including unique identifiers, market class, growth habit, and origin, is provided in Table S1 of the Supplementary Material; genotype names used throughout the manuscript correspond exactly to those listed in Table S1."

 

Comments 12: Figure quality and readability

"The GGE figures should be improved for readability, resolution, font size, and label strategy."

Response 12: The GGE biplot figures (Figures 1 and 2) have been revised to improve resolution, font size, label readability, and abbreviation consistency. The legend in Figure 1 now explicitly defines the environment abbreviations: RF = rainfed; RF+I = rainfed + rescue irrigation (2023) or rainfed + supplemental drip irrigation (2024). A note has been added to the caption of Figure 1:

"Environmental groupings shown are preliminary and based on the six environments evaluated."

 

Comments 13: Literature support and mechanistic interpretation

"Mechanistic explanations should be clearly distinguished from traits measured in the present study."

Response 13: We have reviewed all the mechanistic statements in the Discussion and added explicit qualifications where the physiological traits were not directly measured in the present study.

Changes made: Representative additions in Section 3.2:

"These physiological traits were not directly assessed in the present study and are cited here as plausible mechanisms consistent with the superior field performance observed under severely limiting rainfed conditions at La Llave, rather than as direct findings of this work."

Similarly, in Section 3.3:

"These indices capture final yield outcomes but not the underlying physiological mechanisms conferring tolerance. Complementary characterization of physiological traits...were not assessed in the present study and represent a priority for future characterization."

 

Comments 14: Editorial consistency and manuscript structure

"The manuscript would benefit from a final editorial revision for consistency of units, abbreviations, scientific names, genotype names, table headings, figure captions, and journal-template elements."

Response 14: A comprehensive editorial revision was performed throughout the entire manuscript. Key actions taken include (i) standardization of moisture treatment terminology throughout (rainfed, rainfed + rescue irrigation, and rainfed + supplemental drip irrigation); (ii) consistent use of “preliminary environmental groupings” in place of "mega-environments"; (iii) verification of genotype names against Table S1; (iv) consistency of units and abbreviations across tables, figures, and text; and (v) alignment of the Discussion limitations section with the Conclusions.

 

Minor Comments:

All the minor comments have been addressed as follows:

• Units consistency: yield (kg ha⁻¹), seed weight (g), precipitation (mm), and temperature (°C) are consistent throughout.
• Environment abbreviations: standardized across text, tables, and figures (Celaya Rainfed, Celaya Rainfed + Rescue Irrigation, Celaya Rainfed + Supplemental Drip Irrigation, La Llave Rainfed).
• Genotypes in abstract traceable to tables: all four STI/GMP genotypes and two DSI genotypes mentioned in the abstract are reported in Table 4 and Table S2.
• Numerical values in abstract match tables: verified; STI values in the abstract (1.769, 1.679, 1.480, 1.322) match Table 4.
• Climate data source: clarified in Section 2.1—"Climate data were obtained from automatic weather stations located within each experimental site."
• Rescue irrigation vs. supplemental drip irrigation: consistently distinguished throughout the manuscript as two distinct treatments differing in method, volume, and timing.
• Celaya 2023 stress contrast not generalized: qualifying statements added in Sections 2.5, 3.3, 4.6, and the limitations section (3.5).
• Terminology review: "drought tolerance", "stability", "broad adaptation", and "mega-environments" have been moderated and qualified consistently throughout.
• Limitations connected to conclusions: The final paragraph of the Conclusions now explicitly states that environmental groupings and genotypic rankings should be validated before use as the basis for regional recommendations.
• Supplementary tables: Table S1 and Table S2 are cited at the appropriate points in the main text; descriptions have been updated.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the authors’ careful responses. The topic is relevant to stress-resistant breeding in rainfed agriculture and has both academic value and practical implications. Although the limitations have been supplemented, it is still recommended to add literatures for the selection of drought tolerance indices to further enhance the methodological rigor. I agree to accept the manuscript after minor revisions.

Author Response

Comments 1: "Thank you for the authors' careful responses. The topic is relevant to stress-resistant breeding in rainfed agriculture and has both academic value and practical implications. Although the limitations have been supplemented, it is still recommended to add literatures for the selection of drought tolerance indices to further enhance the methodological rigor. I agree to accept the manuscript after minor revisions."

Response 1: We thank the reviewer for the positive assessment and for this specific recommendation. We agree that additional literature support strengthens the methodological justification for the selection of GMP, STI, and DSI over other available indices.

 

In response, we have added an explicit sentence in Section 3.3 (Tolerance Indices as Selection Tools) and in Section 4.6 (Drought Tolerance Indices), acknowledging that the selection of these three indices was informed by their consistent performance in discriminating genotypic responses under contrasting moisture conditions in common bean and related legumes. Specifically, we incorporated citations to the following works, which directly support the use of these indices in Phaseolus vulgaris and related legumes under contrasting moisture regimes:

 

• Tay et al. (2025) [Reference 6] — evaluated terminal drought tolerance in common bean genotypes in Chile using DSI and complementary indices under rainfed vs. rescue irrigation contrasts.
• Cid-Ríos et al. (2025) [Reference 19] — identified bean lines with differential response to moisture availability in semiarid environments of north-central Mexico using combined index approaches.
• Mwale et al. (2023) [Reference 20] — applied STI and GMP in tepary bean (Phaseolus acutifolius) under contrasting stress conditions, confirming their discriminatory utility in the genus Phaseolus.
• Bhandari et al. (2024) [Reference 32, new] — evaluated the comparative validity of multiple stress tolerance indices including GMP and STI under contrasting stress levels in wheat and concluded that the GMP was consistently associated with yield performance under both moderate and severe stress conditions.

 

Changes made:

Section 3.3, the second paragraph, the following sentence was added:

"The selection of these three indices over other available alternatives was informed by their consistent performance in discriminating genotypic responses under contrasting moisture conditions in common bean and related legumes [6,19,20,32]."

 

Section 4.6, the introductory paragraph, the following sentence was added:

"These indices were selected on the basis of their complementary utility and consistent performance for discriminating genotypic responses under contrasting moisture conditions in common bean and related legumes [6,19,20,32]."

 

The full reference for the new citation is as follows:

Bhandari, R.; Paudel, H.; Alharbi, S.A.; Ansari, M.J.; Poudel, M.R.; Neupane, M.P.; Solanki, P.; Kushwaha, U.K.S. Evaluating stress tolerance indices for their comparative validity to assess terminal heat stress and heat drought tolerance of winter wheat (Triticum aestivum L.) genotypes. J. Agric. Food Res. 2024, 18, 101506. https://doi.org/10.1016/j.jafr.2024.101506

Reviewer 3 Report

Comments and Suggestions for Authors

 The authors have thoughtfully addressed the previous comments, strengthened the clarity and rigor of the work, and improved the overall quality of the presentation. I believe the manuscript is now suitable for publication.

Author Response

Comments 1. "The authors have thoughtfully addressed the previous comments, strengthened the clarity and rigor of the work, and improved the overall quality of the presentation. I believe the manuscript is now suitable for publication."

 

Response: We thank for this positive assessment and for the constructive feedback provided in the first round, which contributed to improving the manuscript.