Assessing the Impact of Temperature and Precipitation Trends of Climate Change on Agriculture Based on Multiple Global Circulation Model Projections in Malta

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I was excited to read this study initially, as climate change is expected to greatly influence the Mediterranean agriculture production of Greece, Italy and the other Basin countries. There is enough data here to give statistical likelihood production outcomes on the three chosen crops: potatoes, forage and vineyards. This is the "payoff" of the whole study, but it is missing! Instead, we get English words and not mathematical predictions. The reader
understands that statistical production predictions would only be guess-estimates, but it is the meat of the study. It is disappointing that this large effort has fallen short for its readers, when it really is an exemplary opportunity to do statistical analysis. If the authors choose other countries in future studies, it is recommended that the climate prediction models get folded in with the crop requirements to give statistical estimates of reduced or enhanced yields.

 

 

Author Response

Dear Reviewer,

Thank you for your thoughtful and constructive feedback. We greatly appreciate your suggestion to integrate climate prediction models with crop requirements to provide statistical estimates of reduced or enhanced yields. This is an excellent point, and we fully acknowledge that such an approach would add significant value to the study.

However, we would like to highlight that there is currently a substantial data and knowledge gap in this area. Conducting a comprehensive statistical analysis of production outcomes would require extensive high-resolution climate and agricultural data, along with additional time and computational resources, which were beyond the scope of our current study. Given these constraints, our focus was to identify key trends and potential impacts rather than provide detailed statistical projections.

That said, we fully recognize the importance of quantitative predictions and appreciate your perspective on how they could enhance our study. In future research, we aim to explore these statistical models further, potentially expanding the study to other regions and crops, provided sufficient data is available.

To address your suggestion, we have now explicitly acknowledged this in the recommendations section of our paper, where we state:

Line 902 - "Since this study focused on analysing climate trends and their potential implications for crop production without directly quantifying projected yield variations, future studies should incorporate climate prediction models alongside crop modelling to provide statistical estimates of potential yield changes under different climate scenarios. Utilising established methodologies, such as those used in Galdies and Vella [35], in which they modelled the crop evapotranspiration (ETo) flux using an ETo calculator created by the Food and Agriculture Organization, could offer a more quantitative assessment of how climate change may impact agricultural productivity. This would enhance the predictive value of such research and support more informed decision-making for policymakers and stakeholders in Mediterranean agriculture."

Once again, we sincerely appreciate your valuable insights and the time you have taken to review our work. Your feedback has been instrumental in shaping our future research directions, and we are grateful for your contribution.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

This study uses the CMIP5 climate model to assess the impacts of climate change. The study has contributed to the growing concerns about agricultural sustainability. However, the paper needs to be significantly revised, and the following comments must be addressed before moving forward.

 

Major suggestion:

1. The paper focused on CMIP5 models, although CMIP6 provides better outcomes and is used in the latest IPCC reports. The authors addressed CMIP6 but did not justify why the CMIP5 model was selected in this study and what the advantages of the selected model are.

 

2. I would suggest the authors should address the selections of the models and what are the advantages and/or disadvantages, etc.

 

3. The study did not address and justify how the method is to be applied and practiced in the real world. I would suggest that the authors discuss the practical and regional applications in the discussion section more.

 

4. The study did not justify the methodology (design science), like why you selected the study method, why the model was selected, and why you used the proposed framework should be addressed. Methodology is different from Method. I would also recommend changing the title of section 2 to Methodology instead of Method.

 

5. The structure of the article needs to be reorganized. For example, do you really need nine sub-sections for the Result section? The same issue is in the Discussion section.

 

Minor suggestions:

6. The authors need to double-check the format, like line 57, 148.

 

7. The format of the table needs to meet the requirements of the journal.

 

8. More attention needs to be paid to the citations in the article. For example, in line 103, if [12] affirmed a statement or findings, the authors need to be shown before [12], like Galdies et al. [12] …

 

9. Typically, the abbreviation section should be named “Nomenclature.”

 

10. The figure resolution also needs to be revised, like in Figure 2-3.

 

11. The figures’ format needs to be revised. Why do some figures have boundaries while others do not?

 

12. Double-check every sentence, edit the format to match the journal, and remove redundancy if necessary.

Author Response

Comment 1 - Major suggestion:

1. The paper focused on CMIP5 models, although CMIP6 provides better outcomes and is used in the latest IPCC reports. The authors addressed CMIP6 but did not justify why the CMIP5 model was selected in this study and what the advantages of the selected model are.

Response 1- Dear Reviewer,

Thank you for your insightful comments regarding the choice of climate models used in this study. We appreciate your concern and recognize the importance of clearly justifying our selection of CMIP5 over more recent model iterations such as CMIP6 and CMIP7.

To address this, we have expanded the introduction to provide a more detailed explanation of our choice. Specifically, we highlight that while CMIP6 aims to improve climate process accuracy and update emissions scenarios, CMIP5 remains widely adopted in climate change research and policy frameworks. Additionally, at the time of data collection in 2023, high-resolution downscaled CMIP6 datasets were not readily available at the spatial scale required for our study. We also discuss how CMIP5 offers a 1 km downscaled dataset, which is particularly critical for small-scale regional analyses, such as our study on Malta.

The revised introduction now includes the following justification:

Line 128 - "Following the release of Coupled Model Intercomparison Project Phase 5 (CMIP5), more advanced multi-model ensemble datasets from CMIP6 [17] and CMIP7 [15] have been introduced. While CMIP6 models aim to enhance climate process accuracy and employ updated emissions scenarios [18], the choice to utilise CMIP5 for this study is based on its widespread adoption in existing climate change research and policy frameworks. Much of the current knowledge and ongoing climate change risk assessments and adaptation strategies are still rooted in CMIP5, making its projections highly relevant for continuity in research and practical applications [19].

At the time of data collection in 2023, high-resolution downscaled CMIP6 datasets were not readily available at the spatial resolution required for this study. Currently, the finest resolution available for CMIP6 from Copernicus datasets is 12 km, with the most widely accessible datasets offering a coarser resolution of approximately 50 km [20]. In contrast, CMIP5 is the only model framework that provides a 1 km downscaled dataset, which is particularly important for small-scale regional studies, such as those focusing on Malta. Furthermore, recent research [21] highlights that CMIP5 exhibits lower inter-model variability and a more balanced precipitation distribution when compared to CMIP6, making it a more stable choice for assessing climate impacts. While both CMIP5 and CMIP6 indicate rising temperatures and increased precipitation, CMIP6 projects stronger warming trends and a more uneven annual precipitation distribution, with wetter wet seasons and drier dry seasons."

Furthermore, we have expanded the section following the model justifications to further clarify our reasoning. We now include a comparison of Equilibrium Climate Sensitivity (ECS) and Transient Climate Response (TCR) values between CMIP5 and CMIP6, emphasizing how CMIP6 may introduce greater uncertainty in climate projections. This is particularly relevant for studies requiring consistency and lower variability in projections. The following text has been added:

Line 184 - "Additionally, the study by Meehl et al., [23] showcased how the average ECS and TCR values across CMIP5 models project less significant warming trends when compared to CMIP6. For context, CMIP5 had an ECS and TCR of 3.2 °C and 1.8 °C, respectively, whereas in CMIP6, the corresponding values are slightly higher at 3.7 °C and 2.0 °C. This suggests that CMIP6 may introduce greater uncertainty in climate projections, which could be a limitation for studies requiring consistency and lower variability. Given these factors, this study employs CMIP5 as the primary climate modelling framework."

We hope that these additions sufficiently address your concerns and provide a clear rationale for our methodological choices. Once again, we sincerely appreciate your valuable feedback, which has helped strengthen the clarity and justification of our study.

 

Comment 2 - I would suggest the authors should address the selections of the models and what are the advantages and/or disadvantages, etc.

Response 2 - To enhance transparency, we now reference the comprehensive assessment by (Line 150 – 190) Flato et al. [22], which outlines the key improvements in CMIP5 compared to previous generations. These advancements contribute to more reliable climate projections, particularly for assessing agricultural impacts in the Maltese Islands.

Additionally, we have introduced Table 2, which presents a comparative overview of key climate sensitivity parameters, including Effective Radiative Forcing (ERF), Equilibrium Climate Sensitivity (ECS), Transient Climate Response (TCR), Climate Sensitivity Parameter (CSP), and Climate Feedback Parameter (CFP) for the selected CMIP5 models. This table provides a clearer understanding of model variability and their respective strengths and limitations.

Furthermore, we have expanded the discussion to highlight the range of ECS values (2.8°C to 4.7°C) and their implications for warming projections. We also examine variations in TCR, CSP, CFP, and ERF, which influence how different models balance radiative forcing and temperature changes. This discussion helps clarify why these models were chosen and how their characteristics may affect the study’s findings.

We hope these additions adequately address your comments and enhance the methodological transparency of our study. Once again, we sincerely appreciate your constructive feedback, which has been instrumental in improving the clarity and robustness of our analysis.

Comment 3 - The study did not address and justify how the method is to be applied and practiced in the real world. I would suggest that the authors discuss the practical and regional applications in the discussion section more.

Response 3 - To address this comment, we have significantly expanded the recommendations section (Section 4.4) to include more Malta-specific applications. One key addition is a discussion on precipitation patterns across different districts, emphasising how projected variations in rainfall may influence water availability and agricultural planning. We highlight that while southern areas, including the South Eastern District and southern parts of the Western District, are expected to receive higher precipitation, the Northern District, Gozo, and Comino may face lower rainfall levels. This underscores the need for localised water management strategies, particularly in drier regions, to ensure sustainable agricultural and urban resilience planning.

To provide further practical insights, we reference Papadimitriou et al. [69], whose work underscores the importance of region-specific solutions for water resource management in Malta. Our study now discusses optimising irrigation techniques and adopting innovative water management strategies that align with Malta’s unique climatic and agricultural conditions. Additionally, we stress the necessity of a multi-stakeholder approach, where farmers, industry stakeholders, and policymakers collaborate to implement effective and sustainable solutions within existing regulatory frameworks.

Furthermore, in response to concerns about crop resilience under climate stress, we have expanded our recommendations to include alternative potato strains that may perform better under Malta’s changing climate. Specifically, we suggest the adoption of Tetyda and Finezja, which are more heat-tolerant and efficient under restricted irrigation compared to currently used cultivars like Alpha and Arran Banner. These crop adaptations could enhance food security and agricultural sustainability under future climate conditions.

We hope these additions effectively address your comments by strengthening the real-world applicability of our findings. Once again, we sincerely appreciate your insightful suggestions, which have helped improve the practical relevance of our study.

Comment 4 - The study did not justify the methodology (design science), like why you selected the study method, why the model was selected, and why you used the proposed framework should be addressed. Methodology is different from Method. I would also recommend changing the title of section 2 to Methodology instead of Method.

Response 4 - To address your concerns, we have:

1. Renamed Section 2 from "Method" to "Methodology" to better reflect its scope.
2. Expanded the justification for our methodological approach by explicitly explaining why we selected this study design, model framework, and analytical techniques.

We now state in the Methodology section:

Line 222 - “This study adopts a design science approach to analyse the impact of climate variables on agricultural production in the Maltese Islands. This methodology was chosen to integrate spatial analysis, agricultural census data, and climatic projections in a structured, reproducible manner, ensuring a comprehensive understanding of the impact of climate on agriculture. A district-level analysis was selected due to the availability of high-resolution Agricultural Census data from 2010 and 2020, which provide detailed insights into land use and irrigation patterns across Malta. To assess potential climate impacts, we utilised CMIP5 climate projections instead of CMIP6 or CMIP7, as CMIP5 has undergone extensive validation, and at the time of the study, it provided the only downscaled dataset at a 1 km resolution, which is essential for a small-scale regional study. Instead of using raw temperature and precipitation values, we selected bioclimatic variables (BIOs), as they offer an ecologically relevant representation of climate trends affecting agriculture [28]. Spatial mapping was carried out using QGIS, with official Maltese district shape files sourced from the MITA Geoportal, enabling the classification and ranking of districts based on their exposure to climate impacts. The following sections outline the data sources, processing techniques, and analytical framework employed to achieve the study’s objectives.”

These revisions provide a clearer methodological justification, aligning with your recommendations.

We appreciate your valuable insights, which have strengthened the transparency and rigor of our study.

Comment 5 - The structure of the article needs to be reorganized. For example, do you really need nine sub-sections for the Result section? The same issue is in the Discussion section.

Response 5 - To address this, we have:

1. Reduced the total number of sub-sections in both the Results and Discussion sections, consolidating related content where appropriate.
2. Eliminated redundant information, particularly in these sections, to improve clarity and maintain a more streamlined narrative.

These revisions ensure a more concise and logically structured manuscript while preserving all essential findings and interpretations.

We sincerely appreciate your constructive suggestion, which has helped enhance the overall readability and organization of our study.

 

Minor suggestions:

1. The authors need to double-check the format, like line 57, 148.

 

1. The format of the table needs to meet the requirements of the journal.

 

1. More attention needs to be paid to the citations in the article. For example, in line 103, if [12] affirmed a statement or findings, the authors need to be shown before [12], like Galdies et al. [12] …

 

1. Typically, the abbreviation section should be named “Nomenclature.”

 

1. The figure resolution also needs to be revised, like in Figure 2-3.

 

1. The figures’ format needs to be revised. Why do some figures have boundaries while others do not?

 

1. Double-check every sentence, edit the format to match the journal, and remove redundancy if necessary.

Thank you for your detailed comments and minor suggestions. We appreciate your thorough review, which has helped us refine the manuscript further.

Response to minor comments - To address your feedback:

1. Formatting Issues (Lines 57, 148, etc.) – We have carefully reviewed and corrected any formatting inconsistencies throughout the manuscript.
2. Table Formatting – We have adjusted the tables to align with the journal’s requirements. Specifically, for Tables 5 and 6, we have:
• Reinserted them at their original resolution to ensure clarity.
• Adjusted the layout to landscape for better readability.
• Ensured the legends are appropriately placed alongside the tables.
• However, we encountered a spare page before the table that we could not remove. We will continue to troubleshoot this issue.
3. Citation Format (e.g., Line 103) – We have revised in-text citations where necessary, ensuring that references are cited in accordance with journal guidelines. For example, instances like “[12]” have been modified to “Galdies et al. [12]” where appropriate.
4. Abbreviation Section – We have renamed it to "Nomenclature", following standard academic conventions.
5. Figure Quality & Formatting – We have:
• Revised the resolution of Figures 2 and 3 to enhance clarity.
• Standardized figure formatting so that all images have a consistent appearance.
• However, we encountered a glitch with Figures 1 and 2, where an automatic border appears on the right side. Since we were unable to remove it, we will provide these figures as supplementary material to ensure clarity.
6. Final Proofreading & Format Consistency – We have conducted a thorough review to correct any remaining formatting inconsistencies, remove redundancy, and ensure compliance with the journal’s style guide.

We appreciate your insightful suggestions, which have significantly improved the clarity and presentation of our work. Thank you for your time and effort in reviewing our manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

Overall Assessment

This manuscript evaluates the potential impacts of climate change on agricultural production in Malta using multiple Global Circulation Models (GCMs) from CMIP5. While the topic is relevant and timely, I appreciate authors effort, but the manuscript suffers from flaws that I want authors to consider revising their work and improve its quality.

Major Concerns

1. Methodological Limitations

The authors rely exclusively on CMIP5 models despite acknowledging the availability of more advanced CMIP6 and emerging CMIP7 datasets. While they attempt to justify this choice (lines 149-157), their explanation that "much of the current knowledge and ongoing climate change risk assessments and adaptation strategies are still rooted in CMIP5" is insufficient. My concern is that using outdated modeling frameworks undermines the scientific validity and practical utility of their conclusions. I suggest you can strengthen your justification claims.

The statistical analysis is inadequate. The hierarchical clustering approach used to assess model agreement lacks rigorous validation beyond the computation of cophenetic correlation coefficients. The authors fail to demonstrate how well their selected models represent historical climate conditions in Malta specifically, which is crucial for establishing confidence in future projections.

2. Data Interpretation Issues

The manuscript presents abundant data but often fails to adequately interpret their significance. For example, in section 3.3, the authors present temperature and precipitation trends but provide minimal analysis of why specific models diverge or converge. The physical mechanisms driving climate change impacts on Maltese agriculture are poorly explained.

The authors do not sufficiently address the limitations of applying global climate models to a small island state like Malta. The spatial resolution of the GCMs used is likely inadequate for capturing microclimatic variations critical to the diverse agricultural zones in Malta's limited land area.

3. Structural and Organization Problems

The paper appears to be lengthy because it has some redundant information. Section 4.3 ("Characteristics and implications of projected variables on agricultural crops") contains substantial repetition across subsections on potatoes, forage, and vineyards. This section could be condensed by at least 30% without loss of content.

The discussion frequently shifts between focusing on climate projections and agricultural impacts without clear transitions, making it difficult for readers to follow the logical flow of the argument.

4. Inadequate Uncertainty Analysis

Despite acknowledging model uncertainties in section 4.6, the authors fail to systematically quantify these uncertainties or incorporate them into their recommendations. This is particularly problematic given the high stakes of agricultural policy decisions that might be informed by this research.

The conclusions about agricultural impacts are presented with greater certainty than the underlying data warrant. For example, the statement that "forage-producing Districts may expect an increase in their forage yield" (lines 668-669) does not adequately reflect the substantial uncertainties in both climate projections and crop response models.

5. Limited Originality

Many of the findings echo existing literature on Mediterranean climate change impacts without substantial novel contributions because the recommendation has been suggested by other researchers. The recommendations provided in section 4.5 are generic and lack the specificity needed for practical implementation in the Maltese context.

I advise authors to rewrite the conclusion section and move recommendation and limitation of study into this section. The conclusions should emanate and present only the findings of the study succinctly and recommendations tailored to the study’s result and not generic. Remove citation here, be specific to work alone. Limitation should be summarized also. I suggest authors revisit the limitation for using CMIP5 line 806-816. It is countering your justification previously presented in introductory section

Minor Issues

• The meaning of GCM should be written full in the title.
• Provide the sources for figure 1 & 2
• Improve the abstract with major findings of the study.
• The manuscript contains incorrect intext citation, grammatical errors and awkward phrasings that impede clarity. Eg line 96, 150, 524. Authors should do a thorough proofread.
• Move the abbreviation before the introductory section so readers can have an idea of what various acronym stand for before encountering it. Improve on the use of abbreviation intext.
• Several figures lack clear legends and are difficult to interpret (especially Tables 4 and 5).
• The manuscript uses inconsistent terminology, sometimes referring to "Districts" by number and sometimes by name.

Conclusion

This manuscript addresses an important topic with potential policy implications for Maltese agriculture, I recommend a major revision.

Comments on the Quality of English Language

see report

Author Response

Overall Assessment

This manuscript evaluates the potential impacts of climate change on agricultural production in Malta using multiple Global Circulation Models (GCMs) from CMIP5. While the topic is relevant and timely, I appreciate authors effort, but the manuscript suffers from flaws that I want authors to consider revising their work and improve its quality.

Major Concerns

1. Methodological Limitations

Comment 1 - The authors rely exclusively on CMIP5 models despite acknowledging the availability of more advanced CMIP6 and emerging CMIP7 datasets. While they attempt to justify this choice (lines 149-157), their explanation that "much of the current knowledge and ongoing climate change risk assessments and adaptation strategies are still rooted in CMIP5" is insufficient. My concern is that using outdated modeling frameworks undermines the scientific validity and practical utility of their conclusions. I suggest you can strengthen your justification claims.

Response 1 - Dear Reviewer,

Thank you for your valuable feedback regarding the methodological limitations and the justification for using CMIP5 models. We understand the importance of ensuring that our model selection is scientifically rigorous and practically justified.

To strengthen our justification, we have expanded our discussion in both the Introduction and Model Justification sections, highlighting key reasons for selecting CMIP5 over CMIP6 and CMIP7:

• Line 137 - At the time of data collection (2023), high-resolution downscaled CMIP6 datasets were not available at the spatial scale required for this study. The finest available CMIP6 dataset from Copernicus is 12 km, with widely accessible datasets at 50 km resolution, whereas CMIP5 provides a 1 km downscaled dataset. This finer resolution is crucial for small-scale regional studies like Malta.
• CMIP5 has been shown to exhibit lower inter-model variability and a more balanced precipitation distribution compared to CMIP6. While both CMIP5 and CMIP6 indicate rising temperatures and increased precipitation, CMIP6 projects stronger warming trends and more extreme seasonal precipitation variation (wetter wet seasons, drier dry seasons). This introduces greater uncertainty in regional projections, making CMIP5 a more stable choice for assessing climate impacts.
• Despite the advancements in CMIP6, CMIP5 remains widely used in climate change risk assessments, adaptation strategies, and policy frameworks, ensuring continuity with existing research. Many ongoing studies, particularly those used for policy applications, still rely on CMIP5 projections.
• Line 184 - We have expanded our discussion to include Meehl et al. [23], which highlights differences in model sensitivity. CMIP5 models exhibit a more moderate climate sensitivity (ECS = 3.2°C, TCR = 1.8°C), while CMIP6 models tend to amplify warming trends (ECS = 3.7°C, TCR = 2.0°C). The greater sensitivity in CMIP6 models may introduce higher uncertainty, which could be a limitation for studies requiring consistent and lower-variability projections.

Given these considerations, we have clarified and strengthened our justification for using CMIP5 as the primary climate modeling framework. We sincerely appreciate your feedback, which has helped improve the transparency and methodological rigor of our study.

 

Comment 1.2 - The statistical analysis is inadequate. The hierarchical clustering approach used to assess model agreement lacks rigorous validation beyond the computation of cophenetic correlation coefficients. The authors fail to demonstrate how well their selected models represent historical climate conditions in Malta specifically, which is crucial for establishing confidence in future projections.

Response 1.2 - Thank you for your insightful feedback regarding the statistical analysis and validation of model agreement. While this aspect extends slightly beyond the primary scope of the study, we recognize its importance in strengthening the credibility of our projections. Despite the tight time constraints, we have incorporated additional analyses to enhance model validation and transparency.

To address this concern (Line 481 – 546), we have added Tables 10–13, which present a detailed breakdown of model values for six key bioclimatic variables. These tables highlight inter-model variability, allowing readers to assess how different models converge or diverge in their climate projections. This addition complements the hierarchical clustering approach, reinforcing the model agreement assessment.

Key Enhancements:

• Model Agreement and Robustness:
• Under RCP 4.5 (2050 & 2070), models AC, BC, CC, and CN demonstrate higher agreement, clustering together with moderate temperature increases and stable precipitation trends.
• Models MR and GF, in contrast, form a separate cluster, projecting higher warming and drier conditions, which suggests lower confidence in their projections.
• Scenario Sensitivity and Divergence Under RCP 8.5:
• The divergence becomes more pronounced under RCP 8.5, with GF and MR consistently projecting the strongest warming trends and driest conditions.
• In 2050, GF reaches 35.34°C (BIO 5), with MR close behind at 34.29°C, reinforcing their tendency to overestimate warming compared to other models.
• By 2070, GF projects 23.17°C (BIO 1) and 342 mm (BIO 12), confirming its status as the warmest and driest model.
• Scenario Sensitivity of AC:
• AC behaves as a transitional model, aligning with robust clusters under RCP 4.5 but shifting toward warmer and drier models under RCP 8.5. This suggests that its projections are scenario-dependent, requiring careful evaluation based on emission trajectories.

By explicitly showcasing model variability and clustering patterns, these additional tables and discussions provide a more transparent validation of the hierarchical clustering approach.

 

Comment 2 - Data Interpretation Issues

The manuscript presents abundant data but often fails to adequately interpret their significance. For example, in section 3.3, the authors present temperature and precipitation trends but provide minimal analysis of why specific models diverge or converge. The physical mechanisms driving climate change impacts on Maltese agriculture are poorly explained.

Response 2 - To address this, we emphasize that the key parameters of the CMIP5 models used in our study (outlined in Table 2) are crucial for understanding model divergence and convergence. These parameters, including Effective Radiative Forcing (ERF), Equilibrium Climate Sensitivity (ECS), and Transient Climate Response (TCR), directly influence how different models project temperature and precipitation trends. For instance, the variation in ECS values—from 2.8°C in BCC-CSM1-1 to 4.7°C in MIROC-ESM—helps explain discrepancies in warming projections, which, in turn, affect precipitation estimates.

Additionally, we clarify that our study focuses on climate projections rather than a physical assessment of climate mechanisms. While we recognize the importance of explaining climate change impacts on Maltese agriculture, our primary aim is to identify the most robust models for national-level policymaking. However, we will consider incorporating a brief discussion to enhance clarity in linking model outputs to their broader implications.

Comment 2.2 - The authors do not sufficiently address the limitations of applying global climate models to a small island state like Malta. The spatial resolution of the GCMs used is likely inadequate for capturing microclimatic variations critical to the diverse agricultural zones in Malta's limited land area.

Response 2.2 - To address this, we have explicitly justified our choice of CMIP5, as it offers (Line 141) a 1 km downscaled resolution—an option not yet available in CMIP6 or CMIP7. This high-resolution dataset represents the most advanced model currently accessible for Malta and significantly improves the applicability of global climate projections to a small island setting.

We recognize, however, that despite this advancement, global and regional climate models still face inherent challenges in accurately representing microclimates within Malta’s limited land area. While a country-specific climate model would offer an ideal solution, no such tailored model currently exists. Future research should prioritize the development and application of finer-scale climate models specifically designed for small island states to enhance the precision of climate impact assessments.

 

Comment 3 - Structural and Organization Problems

The paper appears to be lengthy because it has some redundant information. Section 4.3 ("Characteristics and implications of projected variables on agricultural crops") contains substantial repetition across subsections on potatoes, forage, and vineyards. This section could be condensed by at least 30% without loss of content.

The discussion frequently shifts between focusing on climate projections and agricultural impacts without clear transitions, making it difficult for readers to follow the logical flow of the argument.

Response 3 - Thank you for your constructive feedback. We acknowledge your concerns regarding redundancy and structural coherence, particularly in Section 4.3. In response, we have streamlined this section by reducing repetitive content across the subsections on potatoes, forage, and vineyards, condensing it by approximately 30% while retaining all critical insights.

Additionally, we have restructured both the results and discussion sections to minimize overlap and enhance logical flow. The transitions between climate projections and agricultural impacts have been clarified to ensure a more coherent narrative, making it easier for readers to follow the argument.

 

Comment 4- Inadequate Uncertainty Analysis

Despite acknowledging model uncertainties in section 4.6, the authors fail to systematically quantify these uncertainties or incorporate them into their recommendations. This is particularly problematic given the high stakes of agricultural policy decisions that might be informed by this research.

The conclusions about agricultural impacts are presented with greater certainty than the underlying data warrant. For example, the statement that "forage-producing Districts may expect an increase in their forage yield" (lines 668-669) does not adequately reflect the substantial uncertainties in both climate projections and crop response models.

Response 4 - To address this, we have expanded Section 4.1 to include a structured uncertainty analysis by incorporating climate model clustering. This approach highlights model divergence, particularly under RCP 8.5, where greater variability in temperature and precipitation projections underscores the need for cautious interpretation of results. By presenting temperature and precipitation estimates across different model clusters, we provide a more transparent assessment of the range of possible climate outcomes.

Additionally, we have explicitly linked this uncertainty to agricultural policy decisions. Policymakers and farm managers are advised to consider inter-model variability when designing adaptation strategies, ensuring that projections are not taken at face value but rather assessed within the context of potential deviations. The inclusion of smaller clusters further reinforces confidence in the robustness of warming and precipitation trends while acknowledging the limitations of individual model outputs.

To align with this uncertainty framework, we have also refined statements regarding agricultural impacts to better reflect the inherent variability in climate projections. For example, the previous claim that "forage-producing Districts may expect an increase in their forage yield" has been revised to emphasize the conditional nature of this projection, considering both climate and crop response uncertainties.

 

Comment 5 - Limited Originality

Many of the findings echo existing literature on Mediterranean climate change impacts without substantial novel contributions because the recommendation has been suggested by other researchers. The recommendations provided in section 4.5 are generic and lack the specificity needed for practical implementation in the Maltese context.

I advise authors to rewrite the conclusion section and move recommendation and limitation of study into this section. The conclusions should emanate and present only the findings of the study succinctly and recommendations tailored to the study’s result and not generic. Remove citation here, be specific to work alone. Limitation should be summarized also. I suggest authors revisit the limitation for using CMIP5 line 806-816. It is countering your justification previously presented in introductory section

Response 5 - To address this, we have revised the recommendation section to include more tailored, practical implementations, such as the introduction of thermo-tolerant potato strains and specific water management strategies for different districts in Malta. This ensures that our recommendations are not generic but instead provide actionable insights relevant to Malta’s agricultural and climatic context.

Additionally, we have refined the conclusion to focus solely on summarizing the study’s findings without citations, ensuring clarity and coherence. The limitations section has been revamped to align with the study’s justification for using CMIP5, avoiding contradictions while concisely summarizing the constraints of the research.

While we have kept the limitations and recommendations as separate sections, they have been streamlined to eliminate redundancy and enhance readability. These improvements ensure that the study presents novel contributions, not just reiterations of existing literature, by emphasizing region-specific strategies for agricultural adaptation in Malta.

 

Minor Issues

• The meaning of GCM should be written full in the title.
• Provide the sources for figure 1 & 2
• Improve the abstract with major findings of the study.
• The manuscript contains incorrect intext citation, grammatical errors and awkward phrasings that impede clarity. Eg line 96, 150, 524. Authors should do a thorough proofread.
• Move the abbreviation before the introductory section so readers can have an idea of what various acronym stand for before encountering it. Improve on the use of abbreviation intext.
• Several figures lack clear legends and are difficult to interpret (especially Tables 4 and 5).
• The manuscript uses inconsistent terminology, sometimes referring to "Districts" by number and sometimes by name.

 

Response to minor comments - Thank you for your detailed suggestions. We have carefully addressed all the minor issues raised to improve clarity, consistency, and overall readability.

• Title Update: The full meaning of GCM has been included in the title for clarity.
• Figures 1 & 2: The sources for these figures have now been provided.
• Abstract Enhancement: The abstract has been revised to include the study's major findings, ensuring a concise yet informative summary.
• Proofreading & Grammar: The manuscript has undergone thorough proofreading to correct grammatical errors, awkward phrasings, and in-text citation inconsistencies, including the specific issues noted (e.g., lines 96, 150, 524).
• Abbreviations: While MDPI format requires abbreviations to remain at the end, we have improved their in-text usage to ensure clarity for readers encountering them for the first time.
• Tables & Figures: Tables 5 and 6 have been reformatted in landscape orientation for better readability, with legends clarified. The high-resolution versions of these tables have been reinserted, though a spare page issue remains unresolved.
• Terminology Consistency: We have standardized the use of "Districts," ensuring they are consistently referenced by either number or name throughout the manuscript.

 

Conclusion

This manuscript addresses an important topic with potential policy implications for Maltese agriculture, I recommend a major revision.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors addressed the issue and I think it can be accepted. 

Author Response

Dear Reviewer,

Thank you very much for your positive feedback. We are glad to hear that our revisions have satisfactorily addressed the concerns. We greatly appreciate your time and effort in reviewing our manuscript and your recommendation for acceptance.

Reviewer 3 Report

Comments and Suggestions for Authors

authors have done a good job on the revision. however, write a summary of major finding in the conclusion. 

Author Response

Dear Reviewer,

Thank you very much for your valuable feedback and positive assessment of our revision. We truly appreciate your constructive suggestion.

As per your suggestion, we have now included a summary of the major findings in the conclusion section to enhance the clarity and impact of the paper. Kindly find the summary of the findings below [Line 984 - 1003].

"It demonstrates clear spatial gradients in the patterns of temperature and precipitation bioclimatic indices.  Specifically, the Western District and certain areas in the Northern District consistently exhibit lower temperatures, while the Northern and South Eastern periphery of Malta are expected to experience higher temperatures. Focusing on the Western District, temperature projections consistently indicate cooler conditions across different scenarios, namely RCP 4.5 and RCP 8.5, with a notable difference of approximately 1 °C less than other districts.

Regarding precipitation, an examination of annual precipitation and precipitation during the wettest quarter respectively, reveals a distinct pattern. The southern regions of Malta, including the southern areas of the Western District and the South Eastern district, are projected to receive the highest levels of rainfall. In contrast, the northern regions, specifically Gozo and Comino and the Northern District, are expected to experience comparatively lower levels of precipitation. As for the BIO 17 index, an average value of 8 mm is estimated across all timeframes and RCPs.  However, a more detailed analysis reveals that the Western District is projected to have slightly higher rainfall index (BIO 17), ranging from approximately 9 to 9.2 mm when compared to other Districts, which range from 8.4 to 7 mm. This discrepancy is particularly significant when compared to Gozo and Comino and the periphery of the Northern District as it highlights the regional perspective, indicating a slightly elevated level of precipitation in the Western District within the context of projections for the driest quarter."