the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Sep 2024
Climate Change and Farmer Livelihoods in Wayanad, India: A Livelihood Vulnerability Index Assessment
Abstract. This study investigates the intricate relationship between climate change and livelihood vulnerability in the Wayanad district of Kerala, employing a Livelihood Vulnerability Index (LVI) to assess the household-level vulnerability of local farmers. A total of 41 indicators were used to construct the vulnerability index for the farmers of the Western Ghat region, with 16 indicators related to sensitivity, 7 to exposure, and 18 to adaptive capacity, and the index is administered to the farmers of Wayanad, Kerala. The results indicate a high level of vulnerability among most farmers, with exposure and sensitivity to climate risks, such as floods and droughts, significantly outweighing their adaptive capacity. The findings reveal that a substantial proportion of rural households are highly exposed to adverse climate change risks and lack the social, physical, and financial capital to effectively mitigate these challenges. The region's geographic and climatic conditions further exacerbate these vulnerabilities. Given the heavy reliance on agriculture for livelihoods in Wayanad, these results underscore the urgent need for targeted policy interventions to enhance the resilience of these communities.
- Preprint
(884 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on nhess-2024-155', Pradeepkumar Thayyil, 24 Oct 2024
Farmers in Wayanad district are exposed to vagaries in climate which seriously hamper the crop and live stock production . This article is making a serious attempt to factor the issues and will be useful for further analysis .
Citation: https://doi.org/10.5194/nhess-2024-155-CC1 -
AC3: 'Reply on CC1', Nikhil k. s., 11 Nov 2024
Thank you for recognizing the study's focus on climate-related challenges farmers face in Wayanad. This research aims to provide a valuable foundation for further analysis and targeted interventions in the region by addressing the impacts of climate variability on crop and livestock production.
Citation: https://doi.org/10.5194/nhess-2024-155-AC3
-
AC3: 'Reply on CC1', Nikhil k. s., 11 Nov 2024
-
RC1: 'Comment on nhess-2024-155', Anonymous Referee #1, 04 Nov 2024
The study is of immense importance considering the recent occurrence of landslides in the study locale. The authors have done due justice to the research objectives. The methodology adopted for studying the vulnerability is also sound. I would like to make few minor suggestions/comments-
Selection of respondents -The criteria adopted for selection of respondents has not been mentioned. Was there any relation with holding a focus group discussion and selection of respondents. Please justify under the heading (Section 110).
In section 190, the need for crop diversification has been stressed but in the results the crop diversity ratio was found to be 0.9. Is there a further need for diversification considering the unique terrain and limited land available for cultivation?
There is mismatch in the year mentioned for the citation Raj and Sofia as compared to what is given in text. (section 580)
Citation: https://doi.org/10.5194/nhess-2024-155-RC1 -
AC4: 'Reply on RC1', Nikhil k. s., 11 Nov 2024
Thank you for your constructive feedback and keen observations. I appreciate your suggestions, which will help improve the study's clarity and depth.
- I will add details on the respondent selection criteria, clarifying any link with the focus group discussions in Section 110.
- Regarding the crop diversity ratio, I will address whether further diversification is advisable given the terrain and land limitations.
- I will also correct the citation year discrepancy for Raj and Sofia in Section 580.
Citation: https://doi.org/10.5194/nhess-2024-155-AC4
-
AC4: 'Reply on RC1', Nikhil k. s., 11 Nov 2024
-
RC2: 'Comment on nhess-2024-155', Anonymous Referee #2, 07 Nov 2024
The manuscript requires a complete revision, as its current structure and presentation are too disorganized to allow for a meaningful evaluation of the study. Given the scope of necessary revisions, I recommend rejection and resubmission. The paper is challenging to read and interpret, as the results are entirely disconnected from the announced objective and are further complicated by methodological issues.
The LVI construction lacks a solid foundation in existing studies. The framework is flawed, as it selectively includes four of the five Sustainable Livelihood capitals without justification, notably omitting ‘natural capital,’ which is arguably the most relevant when studying farmers. There is also no explanation for using PCA for variable weighting, with statistical terminology parachuted and without adequate clarification. The choice of indicators is poorly justified, and the normalization and aggregation method (41 indicators with different units) contradicts the study's intended purpose of “highlighting the critical need for targeted interventions to address the unique vulnerabilities of Wayanad households” (L72 onwards). Furthermore, the central concept of this study, "livelihood vulnerability," as defined in L39, is impossible to understand.
The paper lacks a coherent conceptual framework, and there is insufficient reference to existing studies to support the methodology. Additionally, there are two pages of equations that could be summarized in a simple table; statistical methods are presented abruptly and without rationale. Figures 2 and 3 merely visualize climate data and do not add new insights, while Figure 5 also provides little that is not already covered in the text. There is no substantial discussion of the results, further obscuring the study's contribution.
To summarize, the indicator selection lacks appropriate justification, is fundamentally flawed, and is aggregated too broadly to allow for any meaningful conclusions. The chaotic structure of the manuscript makes it impossible to understand the rationale of the study and to identify the authors' original contributions versus data already available. For these reasons, I recommend rejection and resubmission, conditional on the authors following a clear structure, grounding their methodology in a solid conceptual framework, justifying statistical analyses, clarifying the study’s objectives, and providing a meaningful explanation of the results.
Citation: https://doi.org/10.5194/nhess-2024-155-RC2 -
AC1: 'Reply on RC2', Nikhil k. s., 10 Nov 2024
1. The LVI construction lacks a solid foundation in existing studies. The framework is flawed, as it selectively includes four of the five Sustainable Livelihood capitals without justification, notably omitting ‘natural capital,’ which is arguably the most relevant when studying farmers.
2. The paper lacks a coherent conceptual framework, and there is insufficient reference to existing studies to support the methodology.
3. The choice of indicators is poorly justified, and the normalization and aggregation method (41 indicators with different units) contradict the study's intended purpose of “highlighting the critical need for targeted interventions to address the unique vulnerabilities of Wayanad households” (L72 onwards). Furthermore, the central concept of this study, "livelihood vulnerability," as defined in L39, is impossible to understand
The LVI framework applied in this study is grounded in established methodologies for assessing livelihood vulnerability. The literature consistently supports the use of the LVI for household vulnerability analysis, with foundational studies by Cutter et al. (2000), IPCC (2007), Cannon and Müller-Mahn (2010), Pandey et al. (2017), Jamshidi et al. (2019), and Sarker et al. (2019). This framework, structured around the core dimensions of exposure, sensitivity, and adaptive capacity, aligns closely with IPCC guidelines for vulnerability assessment. The conceptual foundation of this study is based on the IPCC vulnerability model, which categorizes vulnerability into exposure, sensitivity, and adaptive capacity. These elements have been adapted within the Sustainable Livelihood Framework to assess livelihood vulnerability in Wayanad comprehensively, accounting for factors that shape farmers' resilience to climate impacts in the Western Ghats. I also appreciate the feedback on conceptual clarity, which will enhance the paper's coherence and rigor. The indicator selection was guided by the need to reflect Wayanad's specific socio-economic and environmental vulnerabilities, grounded in frameworks like the Sustainable Livelihood Framework and the IPCC model. Indicators were carefully chosen to represent well-documented aspects of exposure, sensitivity, and adaptive capacity relevant to Wayanad farmers, ensuring a scientifically sound and context-sensitive analysis. I have defined it in this study as the collective impact of climate and socio-economic factors on households' capacity for sustainable livelihoods under climate stress (L 38-56)
The selection of indicators was tailored to capture the unique socio-economic and environmental characteristics of the Western Ghats and Wayanad district, utilizing data from surveys, field inspections, and focus group discussions in rural Wayanad, enriched by a literature review and expert input. Furthermore, the methodology and indicator selection draw heavily from prior studies on the LVI, including those by Omerkhil et al. (2020), Rani and Tiwari (2024), and Gerlitz et al. (2016). These studies informed the selection, normalization, and aggregation techniques applied in this research. Recognizing that these references could have been emphasized more fully, I will revise the manuscript to better integrate these foundational sources, providing clarity on the rationale behind the methodological choices. To standardize diverse indicator units, normalization was employed, facilitating aggregation into a composite index that captures livelihood vulnerability holistically. Although this study aims to inform targeted interventions, normalization, and aggregation help identify vulnerability patterns across the district, highlighting areas needing policy attention.
In response to the reviewer's concern about the treatment of natural capital, it is important to clarify that indicators traditionally associated with natural capital—such as family involvement in agriculture, average land size, slope of farmland, type of farming, and landholding—were embedded within relevant subcomponents, rather than treated as a standalone capital. This integration underscores the interconnected nature of natural capital with other livelihood components in rural farming contexts like Wayanad. If necessary, these indicators could be restructured or re-indexed to emphasize their role. By embedding natural capital in this manner, the framework remains contextually sensitive and scientifically robust, aligning with widely accepted methodologies in LVI-based research. To further clarify the term "livelihood vulnerability,".
4. There is also no explanation for using PCA for variable weighting, with statistical terminology parachuted and without adequate clarification
Thank you for your insightful comment regarding the use of Principal Component Analysis (PCA) for variable weighting. PCA is a widely accepted and robust statistical method that prioritizes components based on their contribution to the total variance in complex, multidimensional datasets, making it especially useful in studies involving vulnerability and livelihood assessments. By applying PCA, we derive weights for indicators in a data-driven way, which enhances both the robustness and interpretability of our results. Although PCA’s extensive use in similar research led to the assumption that a detailed explanation might not be necessary, I recognize that providing a brief, clear discussion on its application here would benefit readers. I will revise the manuscript to include a concise explanation of why PCA was selected, highlighting its role in generating unbiased and statistically sound weights for the vulnerability components. This addition will improve methodological transparency and address any potential concerns regarding the weighting process.
5. Additionally, there are two pages of equations that could be summarized in a simple table; statistical methods are presented abruptly and without rationale.
Thank you for the feedback on the presentation of equations. I understand the concern regarding clarity and the potential for conciseness, but I felt that providing the equations fully was important to ensure transparency in constructing the index. The detailed equations allow readers to fully understand the methodology and replicate the process if needed, which is particularly important in studies involving index construction with multiple indicators and components. Also, I will give a brief rationale for the PCA, and other statistical methods used, explaining its relevance to the study objectives as mentioned above.
6. Figures 2 and 3 merely visualize climate data and do not add new insights, while Figure 5 also provides little not already covered in the text. There is no substantial discussion of the results, further obscuring the study's contribution
Regarding Figures 2, 3, and 5, and the discussion of results, the mean temperature and precipitation are major components of LVI (Refer Table 1in the article).
Figures 2 and 3 present baseline climate data essential for understanding the specific climate-related risks faced by Wayanad’s farming households. While these figures may appear primarily descriptive, they play a crucial role in contextualizing the vulnerability assessment by illustrating the climatic trends and patterns that contribute to exposure and sensitivity—two key components of the Livelihood Vulnerability Index (LVI) framework applied in this study. Explanations for these figures are provided in the results and discussion section, linked to observed vulnerabilities in the district.
For instance, in the results and discussion, it is noted that "a study by Tomson (2020) in Wayanad district reported that 41.4% of coffee farmers, 32.7% of black pepper growers, and 30% of coconut farmers experienced production losses ranging from 20–50%, 30–50%, and 20–50%, respectively, due to the flood in 2019–2020." This highlights the impact of extreme precipitation events on agricultural production in Wayanad. Additionally, recent extreme rainfall followed by landslides has caused substantial human casualties and agricultural losses. Figure 2 demonstrates that the frequency of extreme rainfall events has been increasing since 1980, underscoring its relevance to this study.
Similarly, Figure 3(a) shows the distribution of maximum temperatures in Wayanad district from 2013 to 2023, revealing an increase in extreme temperatures over the study period. Figure 3(b) presents the minimum temperature distribution, which shows a trend similar to the maximum temperature. These figures indicate a growing disparity between minimum and maximum temperatures over the past decade, with many studies indicating that crop yields are highly sensitive to heat stress. The effect of temperature range variation on the yield of various crops is presented in a table, uploaded here as a supplementary file.
Moreover, the increase in diurnal temperature (i.e maximum day temperature -minimum day temperature) also has a significant negative impact on crop yield (Rao et al., 2008). We will include the figures relevant to each crop yield and temperature increase in the revised manuscript.Rao, G.S.L.H.V. & Mohan, H. & Gopakumar, C.s & Rishnakumar, K.N. (2008). Climate change and cropping systems over Kerala in the humid tropics. 286-291.
Therefore, Figure 3 is highly relevant to the objectives of this study.
Figure 5 is a visual representation of the study's core findings, specifically illustrating the disparity among the three key components of livelihood vulnerability: exposure, adaptive capacity, and sensitivity. The asymmetry of the triangle in Figure 5 is intentional, as it emphasizes the imbalance between these components in the context of Wayanad. Exposure and sensitivity to climate risks, such as floods and droughts, significantly outweigh the region’s adaptive capacity, which is the central finding of this study. This visual is designed to succinctly encapsulate the results explored in more detail throughout the results and discussion sections. The figure provides a clear synthesis of the vulnerability assessment’s outcomes and serves as a visual tool to reinforce the discussion. While the results are covered in depth in the text, Figure 5 offers a concise summary of the study's findings, highlighting the critical need for targeted interventions to address these imbalances. I understand that the figure might appear to overlap with the text, but I believe it serves an important role in providing a visual summary of complex results. To enhance the clarity and impact of the figure’s contribution, I am happy to provide a more detailed explanation of its significance within the revised manuscript, further emphasizing how it encapsulates the study’s key messages.
-
RC3: 'Reply on AC1', Anonymous Referee #2, 12 Nov 2024
Dear authors,
I will clarify my previous comments.
I will begin by reinforcing my argument that the paper lacks a strong conceptual framework. The framework of exposure, sensitivity, and adaptive capacity is indeed widely known and widely applied. Therefore, using this framework in itself is not the most innovative part. However, the choice of indicators that fit under these three categories for the Wayanad area is the primary result and potentially the most innovative part of this study—yet these choices are not well justified. It appears as though the interviews and data were collected somewhat arbitrarily and then assigned randomly to these three categories.
This is why the construction of the Livelihood Vulnerability Index (LVI), specifically the sub-indices of the LVI, should be supported by a clearly defined conceptual framework that goes beyond the basic exposure/sensitivity/adaptive capacity structure, which already serves as the foundation of the LVI in other studies. There is a need to justify why the selected indicators are specifically appropriate for these three categories and, importantly, for the Wayanad area.
When I mention a lack of a conceptual framework, I am referring to the absence of an overarching rationale that justifies each decision, from creating categories of indicators to determining their final values. Relying solely on the exposure/sensitivity/adaptive capacity framework because it is widely accepted is not sufficient to support the purpose of this study, nor does it clarify many other methodological choices made here. I will elaborate on these points below.
Regarding the sustainable livelihood (SL) framework and the exclusion of natural capital: thank you for incorporating this argument in your author’s response, but it is lacking in the manuscript where it should be central. To further support my point on the absence of a conceptual framework: you do not provide a clear justification for using the SL framework, which should be grounded in more than simply “because other studies did so.” The lack of a conceptual framework is also evident in your choice to include only 4 out of 5 capitals without explaining the rationale in the manuscript, making this selection appear arbitrary, as if elements of the framework were chosen “à la carte” to fit the study.
Regarding the PCA: while the use of PCA for statistical purposes is widely accepted, its application for weighting local indicators is unsupported and missing from your already absent conceptual framework. PCA is designed to identify patterns of relationships between variables, creating a new, reduced set of variables to make data easier to analyze or visualize. Consequently, I do not see how conducting a PCA serves as a valid justification for weighting the indicators—especially as these indicators were chosen without justification. Using PCA and relying on eigenvalues would most likely reinforce inherent biases in the data, stemming from the arbitrary selection of indicators (arbitrary because not detailed in the manuscript) or from sampling and interview biases. This defeats the study’s purpose of having its findings “enable the stakeholders to identify priority areas for intervention”.
What is the purpose of weighting the indicators? (This, too, is missing from a conceptual framework.) If the goal is to give more visibility to the local context by highlighting indicators that might otherwise be diluted in an average, which aggregates data and downplays certain factors, then this purpose should be explicitly stated. Weighting sub-indicators in composite indicators serves to assign importance to each sub-indicator, reflecting its contribution to the overall measure. In this sense, I understand that a PCA might be conducted to statistically identify these important variables—but this would be a result in itself.
A major limitation of using PCA for weighting indicators collected through interviews is that PCA inherently assigns higher weights to indicators with greater variance, assuming they contribute more to the overall information. However, with locally focused, interview-based data, an indicator with low variance—such as a universally valued community practice—might be very significant but receive a low weight if it does not vary much across responses. Thus, PCA may undervalue consistent or universally critical aspects of the local context. There is no reflection on the validity of PCA or how it affects the reliability of the results, as there is no separate discussion section addressing these issues. Additionally, there is no reflection on the limitations of the study (such as the PCA, the sampling, and the interviewing methods).
To achieve the assumed purpose of creating a local index that represents local reality, what seems more appropriate to me is to use local knowledge, with experts in the relevant field assigning weights based on their knowledge. This could be achieved, for example, by involving stakeholders in weighting through participatory methods, like fuzzy-cognitive modeling, which is also used extensively in many LVI studies. I am offering this as a balancing example, to show an alternative methodology of equal value, based on the same reasoning the authors used: its wide acceptance in other studies. To clarify, I am not asking the authors to explain why such methods would not be considered; I am asking them to justify their specific methodological choice for constructing an LVI in the Wayanad area.
This brings me to the structure of the results and discussion section: the current organization makes it impossible to distinguish what comes from this study and what is drawn from the literature. I would recommend that the authors, when revising their manuscript, separate these two distinct sections. The results section should focus solely on the values of the sub-indicators and highlight the most noticeable findings, with the discussion section providing explanations.
I would also like to emphasize again that the figures do not belong in the results section. As the authors themselves state, “Figures 2 and 3 present baseline climate data essential for understanding the specific climate-related risks faced by Wayanad’s farming households. While these figures may appear primarily descriptive, they play a crucial role in contextualizing the vulnerability assessment by illustrating the climatic trends and patterns that contribute to exposure and sensitivity—two key components of the Livelihood Vulnerability Index (LVI) framework applied in this study.” If these figures are intended to contextualize the study, they belong in the Data and Methods section. Following a clear manuscript structure is essential: contextual elements should precede the results section. The announced goal of this study is to build a local LVI, and this should be the focus of the results, not descriptive or contextual elements, which currently receive disproportionate attention relative to the actual findings.
Thus, once again, I strongly recommend that the authors entirely restructure their manuscript, following a clear Intro-Methods-Results-Discussion-Conclusion structure, and develop a complete conceptual framework that details and justifies all methodological choices. This will allow the results derived from these choices to be presented clearly. Only then can a thorough assessment of the manuscript be conducted objectively, which is currently not possible in its present form.
The purpose of this review process is ultimately to support the authors in developing a study and results that can be successfully shared. The review aims to improve the quality of the manuscript. Right now, the focus is on restructuring the manuscript to follow the standard Intro-Methods-Results-Discussion-Conclusion format and on ensuring a complete conceptual framework that justifies each methodological choice and allows the replication of the study elsewhere. These are basic steps necessary to allow for a more in-depth review of the study. However, at its current stage, the manuscript requires extensive major and structural revisions, which is why I suggest the authors resubmit the study after having improved it following these guidelines.
Citation: https://doi.org/10.5194/nhess-2024-155-RC3
-
RC3: 'Reply on AC1', Anonymous Referee #2, 12 Nov 2024
-
AC1: 'Reply on RC2', Nikhil k. s., 10 Nov 2024
-
CC2: 'Comment on nhess-2024-155', KS Krishnamohan, 09 Nov 2024
This study presents an insightful analysis of the impact of climate change on the livelihoods of farmers in the Western Ghat region of Kerala. I would like to offer a few suggestions regarding the exposure component of the vulnerability assessment.
The research investigates changes in climate variables such as precipitation and temperature, along with the effects of extreme weather events. Recently, Kerala has been facing increased occurrences of extreme precipitation, which is captured in the analysis. Additionally, the authors highlight a rise in temperature extremes (lines 215 to 230). I recommend that the authors conduct a more detailed examination of temperature changes to better understand the impacts.
Numerous studies have documented the effects of temperature stress on plants, which can adversely affect crop yields (e.g., Hasanuzzaman et al., 2013; DOI: 10.3390/ijms14059643). For instance, the present study indicates that in several recent years, temperature have exceeded the critical threshold of 35°C. This finding is particularly relevant for crops like rice, as studies such as Xie et al. (2023; DOI: 10.1016/j.envpol.2017.02.030) shows that the maximum growth temperature for rice is below 35 °C. Hasanuzzaman et al. (2013) demonstrate that even a 1 °C increase in seasonal average temperature can result in a reduction in cereal grain yields by 4.1% to 10.0%. Several other studies also quantify the impact of temperature stress on various crops mentioned in this research.
I suggest assessing threshold temperature changes in detail (such as the number of days per year above a temperature threshold) and discuss how this could influence yields of different crops mentioned in this study.
Citation: https://doi.org/10.5194/nhess-2024-155-CC2 -
AC2: 'Reply on CC2', Nikhil k. s., 11 Nov 2024
Thank you for your thoughtful suggestions on refining the exposure component, particularly regarding temperature changes. I agree that a detailed assessment of threshold temperature events, such as the frequency of days exceeding critical crop-specific thresholds, would enhance our understanding of climate impacts on yields. I will incorporate this analysis to provide a more comprehensive view of temperature stress effects on crop productivity in the Western Ghats.
Citation: https://doi.org/10.5194/nhess-2024-155-AC2
-
AC2: 'Reply on CC2', Nikhil k. s., 11 Nov 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 190 | 95 | 69 | 354 | 5 | 39 |
- HTML: 190
- PDF: 95
- XML: 69
- Total: 354
- BibTeX: 5
- EndNote: 39
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1