Farmer’s Perception of Climate Change and Factors Determining the Adaptation Strategies to Ensure Sustainable Agriculture in the Cold Desert Region of Himachal Himalayas, India

Abstract

Agricultural practices in the cold desert region of the Himalayas are frequently affected by climate-induced uncertainty in the past few decades. This research work aimed to examine the following questions: (a) Are there any significant climatic changes in the cold desert region of Himachal Himalayas? (b) How do the local farmers perceive climate change? (c) What and how indigenous and modern climate sensitive resilience measures/practices are being adapted by farmers for risk mitigation? A modified Mann–Kendall (m-MK) test and anomaly index were used to examine the changes in climatic variables over the cold desert region. Data on the observed changes in climatic variables were investigated through gridded products provided by the Indian Meteorological Department (IMD) and farmer perception, and their adaptation measures were collected by an extensive primary survey using a semi-structured questionnaire. The results indicate that farmers’ perceptions of changing rainfall, temperature, and seasons were consistent with historical climatic data. The drying water resources and crop damage were the most pressing concerns for farmers due to climate change activity. The farmers are adapting to climate change by altering their farming practices for agricultural risk management. The binary logistics regression (BLR) model was used to investigate the influence of different variables on the adopting farmer’s decision. The result revealed that various factors like landholding size, accessibility of transport, awareness of climate change, availability of water, and distance from market were responsible for choosing suitable climate resilience adaptation measures. This research contributes to recalibrating appropriate strategies across the cold desert region for designing sustainable agricultural practices.

1. Introduction

Climate change has become a much-talked-about subject on both national and international platforms. Climate change is posing significant challenges to the Himalayan region, impacting its glaciers, water resources, and ecosystems [1,2,3,4]. The Himalayan region, a vast and diverse geographic area spanning several countries, has long been at the forefront of the global climate change discourse. The impacts of climate change are not uniformly distributed; they are especially severe in the agrarian and ecological systems of mountainous areas [5,6]. The mountain region has witnessed above-average warming in the twentieth century [7]. Specifically, the Himalayas region at higher altitudes experiences warming phenomena almost three times the global average [8]. The study reported that the Himalayas are much warmer (0.06 °C/year) than the global average during 1982 to 2006 [9]. The historical climate trends provide a concerning indication of climate change in the western Himalayas. For the majority of the seasons, the mean temperature from 1951 to 2010 showed a noticeably rising trend, viz., monsoons (0.03 °C/year), post-monsoon (0.02 °C/year), and winters (0.02 °C/year), but summer (0.01 °C/year) did not exhibit a statistically significant increase [10]. Most of the studies documented that the consequences of climate change are visible in the western Himalayan regions and have a viable impact on the nation’s economy and food security [11,12]. Reduced water availability, a decline in productivity of food grains and apples, a shortening of rabi season (October to March), a shift of temperate fruit belt, more extreme weather events (drought/flood), and increased incidence of pest/disease outbreaks are some of the adverse effects of climate change associated with the western Himalayas [13,14,15].

According to FAO data, about 39% of people living in mountains in developing nations experience problems with food security. These populations largely depend on a mix of farming activities, including agriculture, horticulture, and livestock. Agriculture is severely affected by climate dynamics, and there is an anticipated decline in productivity that jeopardizes food security [16]. Changes in rainfall and temperature are expected to distress the agricultural cycle, increase the risk of landslides, floods, and soil erosion, and result in drought stress and lower crop yields [17]. The fact that 70% of the Himalayan population is dependent on agriculture shows their susceptibility [18]. Effective climate resilience measures are hampered by the paucity of comprehensive research on the local-level effects of climate change on Himalayan agriculture [19].

In the western Himalayas, the significance of farmers’ perception of climate change impacts has been widely recognized in the literature [12,18,20,21]. However, farmers’ perceptions of climate change are multi-faceted in nature and vary depending on an individual, community, and regional basis [22,23]. Farmers’ responses to climate change may vary across regions based on agroecological contexts, socioeconomic factors, existing infrastructure, and capacity [24,25]. It needs to understand not only the observable changes in climate but also the attitudes and behaviors of farmers in response to these quantifiable changes. Farmers need this information to recognize patterns, threats, and the necessary preventative measures to deal with the effects of climate change. Seasonal climate forecasts have always been a major concern for farmers, as they provide very little time to respond and limit the ability to implement efficient adaptation measures [26]. Farmers give more weightage to recent climatic scenarios because the past experience and external sources can differ significantly [27]. It is mainly due to climatic uncertainty. On the other hand, inaccurate perceptions mislead farmers into adopting inappropriate measures [21]. As such, local farmers’ perceptions and behaviors towards adaptation measures help to ensure sustainable agricultural development in the face of a changing climate.

Adaptation to climate change refers to an adjustment in ecological, economic, and social systems in response to observed or predicted changes in climatic stimuli and their effects and implications for mitigating the adverse impacts of change or taking advantage of new opportunities [28]. Previous studies have identified various adaptation strategies adopted by the farmers in the Himachal Himalayas region [29,30,31,32,33]. These adaptation strategies include techniques related to crops, such as growing high-value crops, mixed cropping, crop calendar adjustments, introducing drought-resistant varieties of seed, and cultivating water-sensitive crops [33,34]. Other strategies involve improving irrigation facilities, soil conservation, practicing modern technology, and diversification of income [29,30,31,35]. It is important to note that climate change’s impact on agriculture varies significantly within the same administrative unit due to altitudinal differences [5]. Consequently, farmers’ adaptation strategies also differ significantly based on their distinct physical and socioeconomic backgrounds. On the other hand, farmers’ perceptions of climate change do not guarantee that they will take adaptation measures. Farmers’ household characteristics and many socioeconomic factors may influence farmers’ decisions to implement some preventive measures in their agricultural practices [36,37]. Those factors could either operate as a barrier or as a facilitator to take the adaptation decision. Previous literature has reported that farmer’s ability to adapt decisions depends on their socioeconomic variables [38], demographic factors [39], psychological behavior [40], social capital [41], institutional factor [42], and physical characteristics of the region [22]. However, location-specific adaptation strategies as well as influencing factors behind farmers’ adaptation decisions should be prerequisite for the comprehensive understanding about the sustainable agricultural practices of the region. Without effective adaptation, initial climate change-related concerns might have severe consequences for farmers’ livelihoods in these challenging, harsh climatic conditions.

Researchers have conducted several studies on the impacts of climate change phenomena in the cold desert region [43,44,45]. The majority of the works indicate that climate change will happen, and agriculture will be affected by these quantifiable changes [34,46]. However, there is limited research focusing on the determinants responsible for the climate change adaptation mechanisms in the study region. To the best of our knowledge, this study is the first of its kind in the cold desert region of Himachal Himalayas aiming to provide essential information from the field to guide policymakers and development partners towards sustainable agriculture practices. The study aimed to address the following research questions, such as (a) Are there significant climatic changes in the cold desert region of Himachal Himalayas? (b) How do local farmers perceive climate change? (c) To what extent do changes in agricultural practices occur via climate adaptation? (d) What are the major determinants that influence farmers’ adoption of climate change adaptation strategies? Therefore, this research enhances knowledge of the impacts of climate change on agricultural practices in the cold desert region. It also helps to design and develop effective adaptation strategies to ensure sustainable agriculture and strengthen the resilience of local farmers in changing climate scenarios.

2. Study Area

Lahaul and Spiti is a cold desert area in the Trans-Himalayan Mountain region. With a total area of 13,835 square kilometers, Lahaul-Spiti is the largest district in Himachal Pradesh, covering 24.85% of the state’s total land area (Figure 1). The harsh climatic conditions of the district act as a barrier to the development of human settlement. Thus, the district is sparsely populated with a population density of almost two people per square kilometer as per the 2011 census report. The district has a total population of 31,564, including a male population of 16,588 and female population of 14,976. Out of a total of 521 villages, 280 villages are inhabited. The altitude in the district ranges from 2650 to 5600 m with two great valleys, Lahaul and Spiti. According to a census report (2011), the district’s economy is predominantly agrarian, and 58.18 percent of workers in the district are engaged in the primary sector. The limited availability of cultivable land in cold desert regions reflects the challenging natural environment [46]. Cultivable land is only found in a few suitable locations due to the rugged topography. The district is well-known for producing high-quality potatoes, barley, buckwheat, and peas, which are in high demand in other parts of the country.

3. Materials

3.1. Climatic Data

In this study, rainfall and temperature data were extracted from the IMD-endorsed gridded dataset available on the website (www.imdpune.gov.in, accessed on 10 March 2024). The data set includes daily gridded rainfall and maximum and minimum temperatures of Lahaul and Spiti from the period of 1971 to 2023. A total of 18 gridded rainfall points at 0.25° × 0.25° spatial resolution covered the study region. On the other hand, 9 gridded temperature points at 1° × 1°spatial resolution were taken into consideration for analyzing the temperature trend of the study and its surrounding region.

3.2. Field Survey Information

For this study, the assessment of farmer’s perception of climate change and its impact on agricultural activities were gathered through an intensive household questionnaire survey. The field survey was conducted during May and June of 2023 and 2024. This region is sparsely populated, with villages that are often isolated and difficult to access. To ensure representation from diverse geographical areas and socioeconomic conditions within the region, we employed a random sampling strategy and selected 22 inhabited villages in the Lahaul and Spiti regions. The demographic statistics about the sample villages were tabulated in the Supplementary Section (Supplementary Table S1). The primary data collected through semi-structured questionnaires, which can be used to measure farmers’ perceptions of changes in climatic variables and the methods they use to cope with climate change impacts. A total of 215 farmers were randomly selected based on their availability, and each household took an average 45 min to 1 h to provide the information. The information we gathered included demographic characteristics, farming systems, agricultural activities, perception of climate change issues, and practices adopted by the farmers. Some of these factors have been reported to influence the farmers’ decisions to adopt adaptation strategies. The details of the research framework are presented in Figure 2.

4. Methodology

4.1. Historical Climatic Data Analysis

4.1.1. Annual and Seasonal Trends of Rainfall and Temperature

To identify the annual and seasonal trends of historical climatic data (including rainfall and temperature) from 1971 to 2023, the modified Mann–Kendall test (m-MK) was applied. Seasonal trends were assessed for the following seasons: pre-monsoon (March to May), monsoon (June to September), post-monsoon (October to December), and winter (January to February). The conclusions drawn from this analysis are based on the assumption that the null hypothesis indicated no trend or had a normal distribution, whereas the alternative hypothesis implies the existence of either a positive (upward) or negative (downward) trend. The modified Mann–Kendall test (m-MK) approach utilized in this study follows the methodology outlined by Yue and Wang (2004) [47], which provides a more robust approach for trend detection in time series data affected by serial autocorrelation. Serial autocorrelation is a prevalent problem in the time series data. To overcome this issue, the modified Mann–Kendall (m-MK) test is utilized, which employs a variance correction approach [47]. This test is a modified variant of the Mann–Kendall test. The m-MK test assesses the autocorrelation coefficient of the time series data at different confidence limits. Before performing the trend test, the m-MK version incorporates a pre-whitening procedure that eliminates the impact of autocorrelation.

With the help of Equation (1), the MK statistic ($S$) was calculated.

$$S={\displaystyle \sum _{k=1}^{n-1}{\displaystyle \sum _{j=k+1}^{n}sign(R_j-R_k)}}$$

(1)

where n is the number of instances or observations, and R_j and R_k are denoted as the rank of kth and jth observations.

The modified variance for the $S$ statistic is presented in Equation (2).

$$V^{*}(S)=V(S)\times {\displaystyle \frac{n}{n^{*}}}$$

(2)

where ${\displaystyle \frac{n}{n^{*}}}$ is denoted as a correction factor, and $V(S)$ is calculated as in the original Mann–Kendall test.

4.1.2. Anomaly Index for Rainfall and Temperature

Anomaly is a widely used metric to measure extreme events in long-term historical datasets [48,49]. By analyzing anomalies in the climatic data over time, we can identify trends in climatic variables and assess whether the variables are changing over time. With the help of Equation (3), anomalies in the climatic data were calculated.

$$A_i={\displaystyle \frac{{\rm X}-\overline{{\rm X}}}{SD}}$$

(3)

where $A_i$ indicates the anomaly index; X denotes climatic data of the particular year (rainfall and temperature); $\overline{{\rm X}}$ is its long-term mean; and SD denotes the standard deviation.

The anomaly index was used to measure the changes in historical climatic data from 1971 to 2023. A minimum of 30 years of data were necessary to accurately identify the anomalies [50]. Shorter time-scales could miss the signs of climatic fluctuation.

4.2. Farmer’s Perception About Climate Change

The first part of the questionnaire covers the details of the demographic information about farmers. Following that, it collects the farmers’ perceptions about climate change, including changes in rainfall, temperature, and seasons. The responses are then processed and analyzed through the statistical software. The degree of association between the sociodemographic characteristics of the farmers and their awareness of the effects of climate change was assessed using Chi-squared test statistics.

4.3. Farmers’ Perceptions of Climate Change Induced Agricultural Activity

The second section of the questionnaire focused on how agriculture is affected by climate change. The questions about the impacts of climate change on agriculture were open-ended, while the specific questions about the impacts of climate change and potential adaptation strategies were presented sequentially. The responses were coded into different categories. The main goal of the survey was to gather the information necessary for in-depth analysis of the research questions. After gathering the information, we also evaluated the main agriculture impacts brought by climate change using the correlation evaluation method (CAE). The correlation evaluation method (CAE) is a statistical method used to evaluate the relationship between variables in a given dataset [51]. The importance of parameters is determined by analyzing their correlation with the outcome variable. It focuses on evaluating the correlation between agricultural problems and their importance in adopting adaptation strategies.

4.4. Farmer’s Perception on the Adaptation Strategies

4.4.1. Methods for Evaluating Local Adaptation Strategies

Similarly, to evaluate the effectiveness of various coping strategies implemented in the villages, farmers were requested to provide details about the coping mechanisms or strategies they had used to mitigate their losses due to the impacts of climate change. Farmers’ various adaptation strategies were grouped into five different categories, including crop and variety adaptation, agricultural land-related adaptation, technological adaptation, irrigation/water resource-related adaptation, and income diversification. A rating scale of 1–5 was used to rate the responses of the farmers. The responses were rated as follows: 1 = most common, 2 = common, 3 = quite common, 4 = less common, and 5 = least common. The survey questions focused on the farmers experiences or beliefs, including their perceptions of the likelihood of causes, consequences, and adaptation measures used to lesser the effects of climate change. Finally, in order to evaluate the hierarchy of the different adaptation strategies used by the farmers, a weighted average mean was applied.

The mathematical expression of the weighted average mean is presented in Equation (4).

$${\rm X}={\displaystyle \frac{{\displaystyle {\sum }_{j=1}^n{w}_j{x}_j}}{{\displaystyle {\sum }_{j=1}^n{w}_j}}}$$

(4)

4.4.2. Description of Independent Parameters Influencing Farmers’ Adaptation Decisions

The decisions of farmers to adapt was influenced by a number of independent variables, which are detailed in

Table 1. Key parameters and their potential impacts on farmers’ adaptation decisions.

Parameter	Variable	Measurement	Possible Effect
P1	Gender	Gender of farmers in a binary format, where 1 = Male; and 0 = Female	Positive/Negative
P2	Education	Education status of the farmers noted as years of schooling	Positive
P3	Age	Age of the responding farmers (Years)	Negative
P4	Farming experiences	Years of farming experience	Positive
P5	Landholding size	The size of land measured in hectares	Positive
P6	Number of livestock	The quantity of livestock	Positive
P7	Altitude	Location of farms expressed as altitude in meters	Negative
P8	Agricultural purpose	The purpose of activity expressed as 1 = market oriented; 0 = subsistence	Positive
P9	Contact with extension agents	Number of extension agent visits annually (continuous)	Positive
P10	Exposure to media	Number of mass media outfits accessed (continuous)	Positive
P11	Household size	Number of members in the household (continuous)	Positive
P12	Engagement in the non-agricultural sector	Number of household members engaged in non-agricultural activity	Negative
P13	Accessibility of transport	Connectivity of the road network	Positive
P14	Awareness of climate change	Awareness or perception of climate change	Positive
P15	Monthly income	Total monthly income of the respondent	Positive
P16	Distance to market	Distance to the local market (kilometers)	Positive
P17	Distance to home	Distance to home to the agriculture location measured in kilometers	Positive
P18	Availability of water	Nearby water availability for agriculture	Positive

. These independent variables were identified through a review of the literature on factors affecting farmers’ adaptations and also considering the unique characteristics of the study region. The independent parameters may have either a positive or negative impact on the farmers’ adaptation decisions. The 18 influencing parameters were considered for this study. The independent parameters were evaluated for statistical issues like multicollinearity. The multicollinearity test was performed to check the independent effect of each of the influencing parameters [52]. The two most widely used indices were tolerance and variance inflation factor [53]. These techniques were frequently used to assess the level of correlation between dummy explanatory parameters. If the tolerance value was less than 0.1 or the VIF was greater than 10, it denoted that there was serious collinearity among the influencing parameters.

4.4.3. Determination of the Parameters Influencing Farmers’ Decisions

After compiling and classifying the data, we used the binary logistic regression model to determine the factors influencing farmers’ decisions to adjust to climate change. The decision to adopt any adaptation strategies can be seen as a yes or no response, so it can be considered as a discrete binary dependent variable [54]. The decision of the farmers to either adapt or not adapt was dichotomous in nature. When dealing with binary variables, the BLR model is the best method to analyze the factors influencing the decision [55]. This statistical method is popular due to its simplicity and straightforward interpretation. The binary logistics model (BLR) is suitable in this scenario because it takes into account the association between a set of independent factors and a binary dependent factor [56,57].

In its reduced form, the binary logistics model is expressed as follows:

$$Y_x=f\left(P_1,P_2,P_3\dots ,P_{18}\right)$$

(5)

where, $Y_x$ is the probability of making an adapt decision (0 = farmers who did not adapt; 1 = farmers who adapted).

5. Result and Analysis

5.1. Climate Change Scenario and Trend Analysis

Figure 3 depicts the rainfall trend for all 18 rainfall grids over the Lahaul and Spit regions. The seasonal trend of rainfall was analyzed for long-term rainfall data from 1971 to 2023. During pre-monsoon season, all 18 grids were detected as having a downward trend in the last 54 years. Among them, 10 grids recorded a negative rainfall trend at a confidence level of 90%.None of the grids showed an increasing rainfall trend in the pre-monsoon season. The negative trend for the pre-monsoon season was measured as −2.90 to −0.95 as per the m-MK test result (Supplementary Table S2). As the result of the m-MK trend detection techniques for the monsoon season, the Z value varied from −0.58 to 4.18 using IMD gridded rainfall data over the Lahaul and Spiti regions. Out of 18 rainfall grids, 6 grids experienced an increasing rainfall trend at the confidence level of 90%. Meanwhile, 6 grids in the monsoon season detected an insignificant increasing trend (p > 0.10) for the period of 1971–2023. The remaining 6 grids showed an insignificant decreasing trend in the monsoon season. Out of the total, 8 gridded data sets in the post-monsoon season detected a decreasing trend (p < 0.10), whereas 9 grids recorded an insignificant decreasing trend (p > 0.10). Only one grid showed an insignificant upward trend for long-term rainfall trend analysis (1971–2023). The Z value varied from −4.90 to 0.13 in the post-monsoon season. In terms of rainfall, the winter seasons showed a falling trend across all the meteorological grids (Figure 3). The winter rainfall had a statistically significant downward trends in most grids (13 grids), while trends in the remaining 5 stations had an insignificant decreasing trend (p > 0.10). Note that the rainfall trends were statistically mixed in all the grids, with their magnitude ranging from −12.60 to −0.53. The outcomes of the m-MK test reported that, for annual rainfall data, among the 18 rainfall grids in the Lahaul and Spiti regions, 8 grids displayed a downward trend at a confidence level of 90%, whereas 7 grids were identified as having an insignificant negative trend (p > 0.10). The remaining 3 grids demonstrated an insignificant positive trend (p > 0.10) for annual rainfall from 1971 to 2023. As reported by Jaiswal et al. (2015) [58], the majority of the population of Himachal Pradesh relies on agriculture and horticulture as their primary livelihoods. Therefore, any decrease in rainfall within the state will significantly affect farmers’ livelihoods. The investigation of various patterns of rainfall is a crucial factor for the agriculture sectors and imperative in understanding annual and seasonal rainfall trends, which is essential for determining precise water needs. However, studying the rainfall patterns in mountainous regions poses a difficult task due to the data availability and varying topographical features.

The m-MK results of the average temperature at a 10% level of significances are presented in Figure 4. The outcome shows that almost all temperature grids witnessed a rise in temperature, with the majority of the grids being statistically significant on annual and seasonal scales in last 54 years (1971–2023). For a better understanding, please refer to Supplementary Table S3 for further details. These temperature trends align with previous observations of warming trends in the Western Himalayas [59]. The rising temperature has direct and indirect impacts on both agricultural production and the economy of the regions [34]. The upward trend in temperature over the region may impact the rate of evapotranspiration, which leads to the need of additional water for agricultural crops and other sectors [60]. Therefore, the current research can serve as a full package, vital for policymakers in formulating strategies for both minor and major geographical areas.

5.2. Calculation of Anomalies for Examining Long-Term Climatic Trend

For analysis of extreme climatic events, anomaly calculation was carried out for both rainfall and temperature data. The findings suggest that there may be a decreased rainfall pattern over the years in the study region. The anomaly index value for rainfall ranged from −1.58 to 6. However, the cold desert region experienced sharp rising temperatures and severe to dry conditions in the period (1971 to 2023). The temperature anomaly index value ranged from −3.02 to 3.48. The outcomes of anomaly calculation were visually compared to determine if there were any significant changes in the long-term position for both sets of climatic data (Figure 5). This information is very essential for developing preparedness planning for economic activity, particularly depending on rainfall and temperature.

5.3. Perception of Farmers About Climate Change

5.3.1. Sociodemographic Information of the Farmers

In this study, 215 farmers were randomly selected as respondents from the Lahaul and Spiti regions. They covered 18 villages in the study region. These farmers were interviewed to gather information about various parameters.

Table 2. Farmers’ characteristic across the study area.

Parameter	Variables	Category	Frequency	Parameter	Variables	Category	Frequency
P1	Gender	Male	129 (60)	P10	Exposure to media	Mean (count)	2.52 ± 1.97
		Female	86 (40)	P11	Household size	Mean (count)	6.13 ± 2.52
P2	Education	No Schooling	31 (14.42)	P12	Engagement in the non-agricultural sector	Mean (count)	1.87 ± 1.56
		Primary	85 (39.53)	P13	Accessibility of Transport	Yes	111 (51.63)
		Secondary	68 (31.63)			No	104 (48.37)
		Higher Secondary	21 (9.77)	P14	Awareness of climate change	Yes	167 (77.67)
		Graduation	10 (4.65)			No	48 (22.33)
P3	Age	Mean (years)	38.21 ± 9.32	P15	Monthly Income (Rupees)	Below 5000	42 (19.53)
P4	Farming experiences	No experience	39 (18.14)			5000–10,000	81 (37.67)
		1–5 years	72 (33.49)			10,000–15,000	62 (28.84)
		5–10 years	63 (29.30)			15,000–20,000	22 (10.23)
		10–15 Years	24 (11.16)			Above 20,000	8 (3.72)
		Above 15 years	17 (7.91)	P16	Distance to market	Below 500 m	74 (34.42)
P5	Landholding size	No own land	28 (13.02)			500 m–1000 m	79 (36.74)
		Below 1 ha	59 (27.44)			1000–1500 m	53 (24.65)
		1–2 ha	51 (23.72)			1500 m–2000 m	4 (1.86)
		2–5 ha	48 (22.33)			above 2000 m	5 (2.33)
		above 5 ha	29 (13.49)	P17	Distance to home	Below 100 m	6 (2.79)
P6	Number of livestock	Mean (count)	2.54 ± 1.91			100 m–500 m	74 (34.42)
P7	Altitude	Mean (meter)	3164.58 ± 426.71			500–1000 m	58 (26.98)
P8	Farming purpose	Market-oriented	147 (68.37)			1000–1500 m	70 (32.56)
		Subsidence	68 (31.63)			above 1500 m	7 (3.26)
P9	Contact with extension agents	Mean (count)	2.87 ± 1.95	P18	Availability of water	Yes	181 (84.19)
						No	34 (15.81)

provides an overview of the respondents’ characteristics. Out of the total respondents, 60% were male, and the remaining 40% were female. The mean age of the farmers was 38 years with a standard deviation of 9 years. A total of 7.91% of respondents had worked in agriculture for more than 15 years indicating their substantial climatic experience over the region. The respondents’ average household size was 6 persons. The survey results also revealed that 14.42% of the respondents did not receive any formal education, while 39.53% of respondents had completed primary school, 31.63% had completed secondary school, 9.77% of the respondents attained higher secondary education, and only 4.65% of respondents had graduated (Table 2). Approximately 68.37% of the respondents reported being engaged in agriculture as their market-oriented occupation. Due to differences in elevation, the impact of climate can vary for the communities living in similar administrative units. The average altitude of the sample villages was about 3165 m above the mean sea level (AMSL). The respondents in the sample had a consistent attitude towards climate change. Most of the respondents (77.67%) agreed that climate change was happening and had an impact on agricultural activities. Almost 85% of the respondents reported that nearby water availability were major sources to continue the agricultural activity. This demonstrates that changes in climatic variables affected water availability, which can hamper the agricultural activities of the region. On average, each household owned about 2.54 livestock across the study region. The majority of farmers had family members engaged in non-agricultural activities. Although off-farm income assisted farmers in sustaining their livelihoods, it was not sufficient for adapting to climate change. Most of the farmers belonged to middle economic income class, even though they focused on agriculture as their main income source. Thus, any kind of changes happening in the climate indicators will hamper the farmers’ financial condition.

5.3.2. Farmers’ Perceptions About Climate Change Indicators

The study aimed to understand how local farmers perceived climate change in the study region. The indicators of climate change were grouped into four categories, i.e., changes in rainfall, temperature changes, changes in season, and changes in overall climate scenario. The categorization of rainfall changes included changes in amount, intensity, and timing and the temperature changes comprised the rise, fall, and extreme events. Seasonal change was considered as a single variable. The majority of the farmers (77.68%) reported perceiving at least one climate change indicator in recent years. The Chi-square test results demonstrated the association between the sociodemographic characteristics of the farmers (gender, education status, and age) and their understanding of climate change impact. The findings showed (

Table 3. Gender-wise perception of climate change indicators.

Parameter	Perception Category	Male (n = 129)	Female (n = 86)	p-Value (Chi-Square Test)
Changes in climate (n = 167; f = 77.68%)	Yes	79.07	75.58	0.547
	No	20.93	24.42
Changes in Rainfall (n = 161; f = 74.88%)	Yes	63.57	60.47	0.646
	No	36.43	39.53
Changes in Temperature (n = 156; f = 72.56%)	Yes	65.12	72.09	0.283
	No	34.88	27.91
Changes in Season (n = 130; f = 60.47%)	Yes	57.36	54.65	0.2
	No	42.64	45.35

) no significant association between the gender of the farmers and the perception of climate change for each category (p > 0.05).

Table 4. Farmer’s level of education and perception about climate change indicators.

Parameter	Perception Category	Education Status					p-Value (Chi-Square Test)
		No Schooling	Up to Primary	Up to Secondary	Up to Higher Secondary	Graduation and Above
Changes in climate	Yes	6.45	77.65	100	100	100	<0.001
	No	93.55	22.35	0	0	0
Changes in Rainfall	Yes	6.45	70.59	100	100	100	<0.001
	No	93.55	29.41	0	0	0
Changes in Temperature	Yes	3.23	70.59	94.12	100	100	<0.001
	No	96.77	29.41	5.88	0	0
Changes in Season	Yes	3.23	40.00	94.12	100	100	<0.001
	No	96.77	60.00	5.88	0	0

shows a significant association between the farmers’ level of education and their perception of respective climate change indicators (changes in climate, rainfall, temperature, and season). Farmers with higher education levels were more likely to recognize the changes in climate patterns. This is mainly because people with higher education might be more exposed to information about climate change, leading to a higher perception of its existence.

In Lahaul and Spiti regions, there was a significant association between farmers’ age and their perception about climate change indicators (

Table 5. Age of farmers and their perception of climate change indicators.

Parameter	Perception Category	Respondent Age					p-Value (Chi-Square Test)
		Below 20 Years	20–30	30–40	40–50	Above 50 Years
Changes in Climate	Yes	0	15	92	96	93	<0.001
	No	100	85	8	4	7
Changes in Rainfall	Yes	0	10	73	78	78	<0.001
	No	100	90	27	22	22
Changes in Temperature	Yes	0	10	76	88	85	<0.001
	No	100	90	24	12	15
Changes in Season	Yes	0	2	39	79	63	<0.001
	No	100	98	61	21	37

). The p-value (<0.05) from the Chi-square test showed a statistically significant positive association between them. The results indicate that older people were more likely to believe that climate change was happening. It is also crucial to understand that older people may have more lived experience and may have witnessed the effects of climate change over time, which may not be the case with younger people.

Figure 6 shows the comparison between the farmers’ perception and observed temporal trend of rainfall and temperature derived from the IMD dataset. Highly matched refers to the condition when the observed trend was significantly positive (negative) in the nearest station, and the perception of more than 50% of the respondents also showed an increase (decrease). Matched refers to the condition when the observed trend was positive (negative) in the nearest station, and the perception of more than 50% of the respondents also showed an increase (decrease). Unmatched refers to the condition when the observed trend was positive or negative in the nearest station, but the perception of more than 50% of the respondents was just the opposite. Highly unmatched refers to the condition when the observed trend was significantly positive or negative in the nearest station, but the perception of more than 50% of the respondents was just the opposite.

5.4. Farmers’ Perception on Climate Change-Induced Agricultural Problems

Table 6. Farmers’ views on climate change-induced agricultural problems.

Climate Change Impact on Agriculture	Average Merit (AM)	Standard Deviation (SD)
Drying of water resources (n = 135; f = 62.79%)	0.668	0.015
Crop damage (n = 85; f = 39.54%)	0.668	0.018
Reduced crop yield area (n = 60; f = 27.91%)	0.501	0.011
Loss of soil moisture (n = 107; f = 49.77%)	0.494	0.023
Pests and disease (n = 40; f = 18.61%)	0.417	0.010
Delayed in time of sowing (n = 29; f = 13.49%)	0.207	0.020
Increase mortality of seedlings (n = 17; f = 7.91%)	0.186	0.011

provides a summary of the numerous impacts of climate change on agriculture perceived by farmers in the Lahaul and Spiti regions. The results showed that 62.79% of the farmers agreed that climate change contributed to the drying up of water resources; on the other hand, 49.77% strongly agreed that soil moisture loss was a result of climate change. Additionally, 39.54% reported that crop damage was also a result of climate change. The findings also revealed that 27.91%, 18.61%, 13.49%, and 7.91% of the farmers claimed that decreased crop yields, increased pests and diseases, delayed sowing time, and higher seedling mortality were results of climate change in the Lahaul and Spiti regions. The outcome showed that climate change is occurring and putting a strain on farmers due to its continuous impacts upon agricultural activities. Interestingly, farmers’ comprehension of climate change and its unrelenting effects on agriculture activities is remarkable. Their understanding about climate change is linked to their experiences with the impacts of different stages of agricultural activities. This indicates that the farmers have been significantly affected by the changing climate conditions over the years. Therefore, it is expected that farmers’ adaptation measures directly target the specific challenges perceived by farmers’ in their region.

Simultaneously, in this study, we also evaluated the major agricultural problems faced by the local famers of the Lahaul and Spiti regions and their importance in adopting adaptation strategies using the correlation attribute evaluation (CAE) method. In this context, we analyzed agricultural problems perceived by the farmers with relation with adaptation or remedies taken by farmers. The results were sorted in descending order according to the importance or average merit of the problems that took some adaptation measures by the farmers (Table 6). The study identified that drying water resources and crop damage were the most pressing concerns for farmers, promoting them to develop strategies to cope with these issues. It was interesting to notice that losing soil moisture was frequently reported by many farmers in the Lahaul and Spiti regions. Despite this, when it came to prioritizing adaptation strategies, addressing reduced crop yields took precedence over tackling soil moisture loss. The result seems counterintuitive, and it also suggests that while farmers frequently experience losing soil moisture in the agricultural fields, the immediate consequence is reduced crop yield area. Thus, it requires the most urgent attention. The findings highlight the need for adaptation strategies that simultaneously address many issues, such as water resource management or drought-resistant crop varieties.

5.5. Local Adaptation Strategies

According to the field survey data collected from 215 farmers, the majority (77.68%) reported observing changes in the climate, but only 56.74% of farmers reacted to these changes by implementing adaptation strategies in their agricultural operations. The farmers in the Lahaul and Spiti regions adopted various adaptation measures to deal with climate change, and they often used these strategies in combination. In this section, we analyzed how the farmers rated different groups of combined strategies on a scale of 1 to 5 according to their importance. There were five types of combinations identified, such as adaptation linked to crop varieties, adaptation related to agricultural land, technology-related adaptation, adaptation tied to irrigation/water resources, and income diversification. The findings highlight that 54.22% of farmers implemented one or more strategies to deal with the challenges brought by climate change. According to the weighted average mean method, most of the farmers commonly practiced irrigation and water resource-related adaptation measures, while the second most common practiced was crop and variety adaptation (

Table 7. Adaptation strategies adopted by farmers based on the ranking scale and weighted average method.

Adaptation Strategies	Ranking Scale					Weighted Average
	1 (Most Common)	2 (Common)	3 (Quite Common)	4 (Less Common)	5 (Least Common)
Crop and variety adaptation	28 (12.97)	27 (12.39)	29 (13.26)	26 (12.10)	13 (6.03)	22.37
Agricultural land-related adaptation	24 (11.10)	29 (13.44)	26 (12.07)	29 (13.44)	14 (6.70)	23.14
Technological adaptation	19 (9.01)	31 (14.47)	31 (14.47)	23 (10.83)	17 (7.95)	23.57
Irrigation/Water resource-related adaptation	34 (15.83)	34 (15.83)	24 (11.24)	15 (6.92)	15 (6.92)	20.57
Diversification of income	23 (10.64)	13 (6.20)	32 (15.11)	24 (11.07)	30 (13.73)	25.98

). Additionally, agricultural land-related adaptation and technological adaptation options received relatively high scores on the weighted average. It is worth noting that income diversification was the least commonly applied climate change response in the study area. These findings highlight that local farmers in the Lahaul and Spiti regions are at least taking steps or being forced to take action in response to the negative impacts of climate change.

5.6. Factors Affecting Farmers’ Adaptation Strategies

The decision to choose a certain adaptation strategy depends on a set of socioeconomic factors [53]. This study examined how several predictor variables influenced farmers’ decisions regarding climate change adaptation measures for agricultural practices. To eliminate the independent effects of each parameter, the multi-collinearity test is very crucial. The term collinearity implies that two variables are near close agreement with each other. When two variables are involved, then it is called multicollinearity. The results of the multicollinearity analysis for each of the independent variables are displayed in

Table 8. Multicollinearity test among the independent variables affecting farmers’ decisions.

Parameter	Tolerance	VIF
Gender	0.773	1.293
Education status	0.647	1.546
Age	0.210	4.765
Farming experience	0.245	4.088
Landholding size	0.451	2.217
Number of livestock	0.301	3.325
Altitude	0.436	2.291
Farming purpose	0.517	1.935
Contact with an extension agent	0.775	1.291
Exposure to media	0.7	1.428
Household size	0.232	4.32
Engagement in non-agricultural activity	0.347	2.883
Accessibility of transport	0.703	1.423
Awareness of climate change	0.459	2.178
Monthly income	0.465	2.152
Distance to market	0.481	2.08
Distance to home	0.91	1.099
Availability of water	0.751	1.332

. The outcome demonstrates that all tolerance levels exceeded 0.1, and all VIF values were below 10. This result indicates a low risk of collinearity and suggests that the variables can be reasonably relied upon to understand their influence on farmers’ adaptation decisions.

This study aimed to investigate the influence of various predictor variables on farmers’ decisions regarding climate change adaptation practices. Therefore, the study utilized the binary logistics regression (BLR) model to assess the impact of several independent variables on the farmers’ decisions. The outcomes of the BLR model showed that the χ2 value was 223.053, with a probability of 0.001, and the accuracy rate was 93.5%. These results indicate a statistically significant relationship between the predicator variables and dependent variable (adaptation measures by the farmers). The value of Nagelkerke’s R² was 0.866, which implies the suitability of the model in this study. The coefficients (β) for each independent variable were calculated, and a positive value indicated that the variable had a strong influence on the famers’ adaptation decisions, while a negative value suggested the opposite.

Table 9. Independent variables influencing farmers’ adaptation decisions.

Parameter	Estimate	Standard Error	p-Value
Gender	−0.065	0.705	0.927
Education status	0.295	0.466	0.527
Age	−0.11	0.092	0.231
Farming experience	0.515	0.644	0.423
Landholding size	1.274	0.423	0.003
Number of livestock	0.057	0.311	0.855
Altitude	0.002	0.001	0.124
Farming purpose	1.042	0.972	0.283
Contact with an extension agent	0.273	0.189	0.148
Exposure to media	−0.061	0.185	0.741
Household size	−0.208	0.269	0.440
Engagement in non-agricultural activity	−0.383	0.352	0.277
Accessibility of transport	1.786	0.739	0.016
Awareness of climate change	4.917	1.727	0.004
Monthly income	0.817	0.602	0.175
Distance from market	−2.299	0.544	0.001
Distance from home	0.313	0.363	0.389
Availability of water	1.998	1.18	0.090

reveals valuable insights into the factors that affect farmers’ decisions to adopt adaptation strategies in agricultural activity. The findings clearly indicate that out of the 16 independent variables, 6 variables (gender, age, exposure to media, household size, engagement in non-agricultural activity, and distance from market) showed a negative correlation with the adaptation decision. On the other hand, 12 variables (farmer’s education status, experience in agricultural activity, landholding size, number of livestock, altitude of the agricultural land, purpose of farming, contact with extension agent, transport accessibility, climate change awareness, farmer’s monthly income, distance from home, and water availability) had a positive influence on farmers’ adaptation decision. A more detailed analysis revealed that out of these 12 positive influencing variables, 4 variables, i.e., landholding size, transport accessibility, climate change awareness, and water availability, significantly influenced farmers’ adaptation decision at the 10% alpha level. Conversely, distance to the market was found to be a significant negative (p < 0.1) contributor to the decision to adopt adaptation measures among the 6 negative influencing variables. The negative coefficient of the variable indicates that farmers near a market have better access to information and inputs as well as market opportunities for their produce, which helps them in practicing different adaptation measures.

6. Discussion

The current research highlights that agricultural practices in the cold desert region of the Himalayas have been frequently affected by climate change-induced uncertainty in recent years. In particular, this study emphasizes the perception of climate change by local farmers in the Lahaul and Spiti regions and explores how farmers’ sociodemographic factors influence their adaptation decisions. The finding provides important insights for ensuring the sustainable agricultural practices in this fragile environment.

The observed decrease in rainfall, particularly during winter and summer months, suggests a shift in the region’s historical precipitation patterns. While annual or seasonal temperature trends were not statistically significant, the rise in October and December temperatures aligns with broader observations of rising temperatures in the Himalayas [61]. These trends underscore the potential threat of climate change to disrupt water resource availability and hamper the agricultural activities. So, the research underlines that the shifts in rainfall patterns, rising temperatures, and more frequent extreme weather events will likely affect the traditional agricultural practices in the Lahaul and Spiti regions.

In the tough highland terrains, climatic impacts can vary for the communities living in the same administrative units due to elevation differences [62]. Therefore, the local people’s perceptions of the changing climatic parameters are crucial for understanding their potential responses. While scientific research has identified significant environmental responses to climate and human impacts, there is a dearth of data on human well-being in the Himalayan region. Nevertheless, it is evident that climate change is profoundly affecting people’s livelihoods and access to natural resources in the mountainous region. The Chi-squared result suggests a strong association of education level and respondent’s age with the perception of climate change. Similar results are also found in the study of Singh et al. (2020) [63], where the observational records of unpredictable climatic events generally matched the views of local respondents in the Lahaul and Spiti regions. Due to sharp variation in the climatic variables, which are adversely affected on crop yield area in the western Himalayas [19]. There has been a shift in agricultural practices in the region from growing traditionally resilient crops to cash crops (exotic vegetables, pea, cauliflower, etc.). The increasing water demands of cash crops replacing traditional, resilient crops threaten the limited water resources of the region, as documented by Sharma et al. (2023) [64]. These crops are water intensive and put pressure on the limited water resources of the cold desert region. A more important factor is acceptance and awareness about climate change. If a farmer is well aware of the changing climatic conditions and has knowledge about its possible impact, they will be more likely to adapt to the new conditions.

Climate change poses different agricultural problems in mountainous regions, like Lahaul and Spiti. During field surveys, the sampled farmers were requested to point out the different climate-related problems in their agricultural practices. Based on the ground data, the agricultural problems related to climate change were summarized. The major problem perceived by the farmers was water scarcity for irrigation purposes. The availability of water for irrigation is very important for any kind of agricultural activity, so farmers who have enough water resources available are more likely to adapt to changed conditions. The agricultural practice of this region entirely depends on the glacial melt water. However, with the impact of climate change, decreasing snowfall and melting glaciers at a faster rate leads to a vulnerable situation for the agricultural community of this region. Farmers are faced with challenges to maintain water resources effectively for irrigation and other agricultural needs. Shashni and Sharma (2022) [34] also examined the sustainable agricultural practices in the Lahaul and Spiti regions and found a close agreement between water resources and agricultural sustainability. Secondly, the majority of the respondents also reported that climate change due to erratic snowfall/rainfall in this region leads to landslide/avalanche phenomena. These phenomena have also intensified the blockage of roads and loss of road connectivity. The harvested crops could not reach the markets at the proper time especially during landslides or heavy snowfall. This increases the risk of crop damage and poses a significant financial burden to the farmers and affects their livelihood. Thirdly, most of the farmers also perceived that soil moisture is another important issue in the Lahaul and Spiti regions. Therefore, the demand for fertilizer for maintaining soil productivity has also increased immensely in the last several years. The findings are almost accordant to the results of Bhalerao et al. (2022) [65], who also reported that declining water availability and soil fertility were significant barriers for sustainable agriculture across the northeastern regions of India. Based on the current knowledge, it is evident that climate change impacts may not be reversed in the short run, and the most feasible solution is adaptation [66].

Adaptation and mitigation are essential elements to address the impacts of climate change. According to Bryant et al. (2000) [67], adaptation is a complex process that incorporates the characteristics of climatic attributes and agricultural systems. It also takes into account the socioeconomic and demographic factors as an influencing factor. This study also aimed to assess the influencing factors affecting farmers’ adaptation decisions by considering the socioeconomic and demographic characteristics of the farmers. The binary logistics regression (BLR) model effectively determined the crucial factors that influenced farmers’ selection of adaptation strategies for coping with climate change. The findings revealed that factors like landholding size, transport accessibility, climate change awareness, water availability, and distance from market played a vital role in shaping farmers adaptation decisions. Farmers with larger landholding are likely to adapt to changing climatic conditions as they have more resources [18]. Shorter distances to markets with better connectivity to road networks increases the likelihood of adaptation, such as easier access to resources or selling and buying the product. Rana et al. (2021) [68] reported the similar results for the Kullu district in Himachal Pradesh, where farmers had higher literacy, closer markets, and better transport accessibility and were found to be less vulnerable to climate change impacts. The model also hints at the potential positive influence of extension agents and income and negative impact of altitude; however, their p-values are not significant at a satisfactory level. A similar finding has been reported among local farmers in the Kashmir Himalaya, and the study revealed that sociodemographic characteristics played a crucial role in the farmer’s decision-making process [36]. To the best of our knowledge, no prior research has thoroughly examined the factors influencing adaptation decisions in cold desert regions, such as Lahaul and Spiti. Therefore, this research work is considered as novel and helpful in promoting wider adoption of agricultural practices in the Himalayan region. The Lahaul and Spiti regions have gone through significant socioeconomic changes due to improved road connectivity, access to markets, higher literacy rates, and the emergence of green peas as a cash crop [69,70]. Figure 7 shows the agricultural activity of the Lahaul and Spiti regions with the help of some field-taken photographs. Farmers’ perceptions coupled with the observed climatic changes necessitates promoting climate-smart agriculture (CSA) practices in this region. Climate-smart agriculture (CSA) is an approach aimed at improving farming practices in light of climate change mitigation and adaptation [71]. It can be defined as a strategy that reduces greenhouse gas emissions, enhances farmers’ ability to adapt to changing conditions, and sustainably increases agricultural production to improve livelihoods [72]. It also offers an opportunity to combat climate change-related food insecurity [73]. But their widespread adoption appears to be stagnating, especially in developing nations [74]. Therefore, it is essential for people to take prompt action to encourage the adoption of CSA technology. From the field survey, it is evident that the government plays a crucial role in funding climate-smart agriculture (CSA) technology by establishing an enabling environment. Government policies and programs that offer subsidies, transfers, or incentives can help farmers obtain access to CSA technology. Subsidies, in particular, have proven to be an important tool for promoting the adoption of CSA technology. Polyhouse is one example of a successful CSA tool that has played a critical role in this harsh climatic region. Polyhouses are an effective solution for plant growth, which also helps to extending the cultivation period. The government provides financial assistance to the farmers through various schemes. Under the scheme, farmers can apply for subsidies to set up the polyhouses. According to the farmers, the subsidies cover a certain percentage of the total cost, which can significantly reduce the financial burden on them. Government agencies collaborate with agricultural universities and research institutions to organize different agricultural training programs. These agricultural training programs aim to raise awareness about climate change and promote strategies for long-term sustainability in agriculture within this region. Aryal et al. (2020) [75] highlighted that the participation of the farmers in agricultural training programs significantly enhance their ability to adopt adaptation strategies. There is a need to educate farmers regarding the long-term benefits of CSA practices as some practices need a long time to get benefits. Empowering farmers with this knowledge is key to building resilience and sustainability in agricultural systems amidst a changing climate.

Along with the positive aspect of the work, it is important to address some limitations for the future scope of the study. The climatic data used in this study has coarser resolution gridded data, and because of that, the spatial map provided a more general overview. Another limitation is that the survey only included 22 villages, but expanding the sample villages and gathering more responses from the farmers could make the result more realistic. The involvement of local communities and the integration of traditional knowledge can be important in developing effective adaptation strategies. Therefore, future research should focus more on these aspects and examine the barriers that farmers face in their adaptation planning. This study also touches on the importance of climate-smart agricultural (CSA) practices but does not elaborate on their effects. Thus, another promising area for future research will be assessing the effects of climate-smart agriculture (CSA) practices on sustainable agriculture in the Lahaul and Spiti regions. The critical point is identifying which form or technologies of CSA are suitable in these cold desert regions. Finally, addressing climate-induced agricultural problems, promoting sustainable practices, and facilitating adaptation strategies, this research could be instrumental to policymakers. However, given the increasing impact of climate change and growing pressure on agricultural food production, more efforts must be focused on advancing research in this field.

7. Conclusions

The current research aimed to investigate how farmers in the Lahaul and Spiti regions perceive climate change and how they adapt to its effects on agriculture. The research suggests that most farmers have noticed the changes in climatic variables over the past few decades and have implemented various adaptation measures in their agricultural practices. Based on the results of the historical climatic data (1971–2023), it could be inferred that there are decreasing rainfall trends in the Lahaul and Spiti regions, particularly during pre-monsoon and winter months. The tremendous and consistent increase in temperatures showed that climate change was occurring and posing various challenges for local farmers in the cold desert regions. The study also found a strong relationship between sociodemographic variables (such as education status and age) and farmers’ perceptions of climate change in the region. Additionally, some agricultural issues, like drying water resources, loss of soil moisture, and crop damage, were the most concerning climate change impacts reported by the local farmers. Despite these obstacles, farmers in the region consistently innovate and adopt techniques to enhance their returns on investment, showcasing their resilience in navigating the complexities of agricultural practices. Based on the weighted average mean result, the most common adaptation strategy was irrigation/water resource-related adaptation, followed by crop and variety adaptation, agricultural land-related adaptation, and so on. This information provides insights into the issues and needs of the farmers, indicating a potential for government interventions to enhance their resilience. Furthermore, the study identified significant determinants influencing farmers’ decisions to adopt some climate change adaptation strategies. The results of the binary logistic regression (BLR) model revealed that landholding size, accessibility of transport, awareness of climate change, distance to market and availability of water significantly influenced farmers’ decisions. This type of study will contribute to the development of an appropriate strategy for designing climate-resilient agricultural systems in the Himalayan cold desert region.