Impact of Changes in Rainfall and Temperature on Production of Darjeeling Tea in India

Abstract

Globally, there has been a lot of focus on climate variability, especially variability in annual precipitation and temperatures. Depending on the area, different climate variables have different degrees of variation. Therefore, analyzing the temporal and spatial changes or dynamics of meteorological or climatic variables in light of climate change is crucial to identifying the changes induced by climate and providing workable adaptation solutions. This study examined how climate variability affects tea production in Darjeeling, West Bengal, India. It also looked at trends in temperature and rainfall between 1991 and 2023. In order to identify significant trends in these climatic factors and their relationship to tea productivity, the study used a variety of statistical tests, including the Sen’s Slope Estimator test, the Mann–Kendall’s test, and regression tests. The study revealed a positive growth trend in rainfall (Sen’s slope = 0.25, p = 0.001, R² = 0.032), maximum temperature (Sen’s slope = 1.02, p = 0.026, R² = 0.095), and minimum temperature (Sen’s slope = 4.38, p = 0.006, R² = 0.556). Even with the rise in rainfall, there has been a decline in tea productivity, as seen by the sharp decline in both the tea cultivated area and the production of tea. The results obtained from the regression analysis showed an inverse relationship between temperature anomalies and tea yield (R = −0.45, p = 0.02, R² = 0.49), indicating that the growing temperatures were not favorable for the production of tea. Rainfall anomalies, on the other hand, positively correlated with tea yield (R = 0.56, p = 0.01, R² = 0.68), demonstrating that fluctuations in rainfall have the potential to affect production but not enough to offset the detrimental effects of rising temperatures. These results underline how susceptible the tea sector in Darjeeling is to climate change adversities and the necessity of adopting adaptive methods to lessen these negative consequences. The results carry significance not only for regional stakeholders but also for the global tea industry, which encounters comparable obstacles in other areas.

1. Introduction

The Darjeeling region, located in West Bengal, India, is home to the GI-tagged Darjeeling tea (scientifically known as Camellia sinensis L.), which is world famous due to its distinct flavor and rich taste [1]. Global climate change has caused serious challenges for agriculture, such as inducing rapid changes in farming systems, crop productivity, and resource availability. Crop and livestock productivity have been disrupted by the increased frequency of extreme events like droughts and floods brought on by rising temperatures and unpredictable precipitation. Climate change has led to reduced yields of rainfed crops in Kenya [2]. In Brazil, climate change has resulted in decreased viable cultivation areas and increased susceptibility to pests, which may potentially reduce traditional coffee-growing regions by half over the next 30 years [3]. Due to increased temperatures and lowered chilling hours, apple orchards in the Indian Himalayas have moved to higher elevations [4]. Wheat yields in the Indo-Gangetic Plains have been reduced by about 8%, while the total wheat production registered reductions of up to 36% [5]. Negative impacts of climate change have also been registered in the production of cotton [6], oilseeds [7], fruits, nuts, and seeds [8]. The effects of climate change are being felt all around the world and the Darjeeling region is also facing these challenges in terms of variability in rainfall and temperature [9], which have huge impacts on the production as well as productivity of the tea in the studied area [10]. The specific blend of temperature, rainfall, and humidity conditions and seasonal patterns that contribute to the taste of Darjeeling tea is becoming increasingly vulnerable to climate shifts. This issue has implications for the livelihoods of thousands of tea plantation workers, as well as the tea industry’s sustainability and the preservation of the unique agricultural ecosystem in this region [11,12].

The global tea industry is currently facing mounting challenges due to climate variations. Worldwide changes in weather patterns, extreme climatic events, and fluctuations in temperature as well as rainfall are causing more and more disruptions [13,14,15,16,17]. These shifts have far-reaching effects on tea cultivation as they impact factors such as growth, yield, and the overall tea leaf quality [18,19,20,21]. Besides rainfall and temperature, the tea quality is also affected by geography, water stress, light factors, altitude, and the herbivores and microbes in the region [22]. These variabilities have implications on tea growth stages such as higher temperatures initially speeding up the development of tea shoots but then shortening the plucking period and increasing the susceptibility to pests, which results in lower production [23,24]. In regions like Darjeeling in West Bengal, India, where tea has a major impact on the nation’s economy, these effects are especially noteworthy. Situated amidst the Indian Himalayas, this region benefits from characteristics such as high-altitude plantations, cool temperatures, and distinct seasonal variations. It is these factors that contribute to the exceptional quality and distinctive flavor profile of Darjeeling tea, earning it the well-deserved title of “Champagne of Teas” [25].

Without effective research, the tea industry in Darjeeling faces a precarious future. The potential harm includes reductions in productivity [26,27,28], lower product quality [29,30,31], increased production costs [32,33] due to climate-related interventions, and economic instability. These challenges could not only jeopardize the region’s tea heritage but also the livelihoods of those who are part of this age-old industry [34,35].

Scientific research has a pivotal role in solving these impending problems. Several studies have been undertaken by researchers and scientists on the yield [36,37,38,39,40] and production of tea [41,42,43] and suitability analyses [30,44,45]. Tea yield is denoted as the quantity harvested per unit area, while the total amount of tea harvested from all cultivated areas is referred to as production. Dutta [46] projected a 2 °C temperature increase in tea-growing regions by 2050 due to climate change, which could drastically alter growing conditions. In contrast, Rahman [47] studied rainfall patterns in Bangladesh and discovered variations in tea yields across divisions. Jayasinghe and Kumar [30] suggested that temperature increases, uneven rainfall distribution, drought, high CO₂ levels, and changes in solar radiation can affect tea production. Duncan et al. [27] noted that monthly tea yields in Assam exhibited sensitivity to rising average temperatures. This sensitivity highlights how susceptible tea crops are to rising temperatures, which could result in lower yields. In their paper, Mallik and Ghosh [48] addressed the impact of temperature and drought on tea productivity as well as the anticipated shift in the period of peak production due to climate change issues.

A study on the temperature variations in tea-growing regions found that the annual temperature range was the most influential factor (51%) for the tea distribution in China, followed by the rainfall of the warmest quarter (14%) and minimum temperature of the coldest month (9.7%) [49]. Wijeratne et al. [50] identified an optimum temperature of around 22 °C for tea production in Sri Lanka. Beyond this threshold, rising mean air temperatures can have detrimental effects on tea cultivation. Rigden et al. [51] took a different approach by associating tea yield variations with rising temperatures, predicting potential declines in tea yields in Kenya. However, they also suggested that changing soil moisture patterns could mitigate some of these adverse effects. Li et al. [40] in their study in Yunnan, China, noted that increasing precipitation, consecutive dry days (CDDs), and heavy rainfall events (R20) can lead to decreased tea yield and prices. The productivity of tea and tea plantations are significantly impacted by shifting temperature patterns. Mila et al. [52] in their study in Bangladesh found that temperature, rainfall, and carbon dioxide levels have a major impact on tea yields over the short and long term. Their effects include tea plants becoming stressed by rising temperatures and excessive rainfall, consequently lowering their output, while sufficient rainfall can have a favorable impact on yields. Ahmed et al. [53] opined that climate change and variability modifies humidity which has obstructive impacts on tea yield, quality, and regional suitability, rendering tea systems and the related economies vulnerable. According to Lou et al. [43], rising temperatures shorten plucking periods, thereby reducing yields and quality. Jayasinghe et al. [54] stated that altered annual mean temperatures and rainfall patterns reduce suitable habitats for tea, leading to decreased tea production. These patterns necessitate undertaking studies on the impacts of climate change and variability on tea.

Therefore, the prime focus of this study was to conduct an investigation on how variations in rainfall, temperature, and the tea-growing region affect tea yield and production in the Darjeeling region of the Indian state of West Bengal. In-depth knowledge of the region’s rainfall and temperature variabilities in relation to variations in tea yield was obtained through an interannual investigation of the trends in all these parameters. Therefore, the research efforts explored the effects of climate-related variability on Darjeeling tea production. This study holds significance not only for the local tea community but also for the broader global tea industry.

2. Materials and Methods

2.1. Study Area

Figure 1 shows our study area, located in the Darjeeling district of West Bengal in India, which is known globally for its high-altitude tea gardens that produce the world-famous Darjeeling tea. Geographically, the area spans between 26.27° N and 27.13° N latitude and 87.59° E and 88.53° E longitude. It has a diversified terrain with steep hills and heights that range from approx. 600 to 2000 m above sea level. Together with the longitudinal and latitudinal breadth, this altitudinal range creates a distinct set of meteorological conditions that are perfect for tea growing. With an average annual rainfall of between 2000 and 3500 mm, mostly between June and September, Darjeeling has a unique monsoonal climate. Darjeeling’s climate fluctuates periodically, with summer temperatures of 15 °C to 25 °C, which, when combined with high humidity, encourage strong growth, and winter temperatures of 5 °C to 15 °C, which provide chilly temperatures that are favorable to dormancy in tea plants. The microclimatic conditions are influenced by the diverse topography, which includes varying slopes. For instance, tea leaves grown on north-facing slopes tend to retain moisture longer and have a milder flavor profile. On the other hand, tea leaves grown on south-facing slopes receive more direct sunlight, which accelerates their growth. The productivity and quality of tea are greatly impacted by these changes in environmental conditions as well as the geographical distribution of the tea garden regions. In order to ensure the sustainability of Darjeeling’s tea business, it is imperative that these characteristics are understood in order to manage the tea output properly and mitigate the effects of climatic variability.

2.2. Database and Methodology

The meteorological data for Darjeeling for the present study were obtained using the index number 42693, as recorded by the IMD in Pune. The obtained data were analyzed using various statistical methods. The meteorological station had complete datasets and there were no missing data. The historical climatic data, including rainfall and temperature records for the study area, and tea production data for a period of 23 years, from 1991 to 2023, were available.

Table 1. Data sources.

Sl. No.	Data	Type	Level	Time Period	Source
1	Rainfall	Secondary	District	1991–2023	IMD, Pune
2	Temperature	Secondary	District	1991–2023	IMD, Pune
3	Tea production	Secondary	District	1991–2023	Tea Board of India, Kolkata

provides a comprehensive overview of the sources of the data.

2.3. Trend Analysis

Webber and Hawkins [55] defined the term “trend” as the overall change in a data series over an extended time period, or alternatively, as the alteration in the analyzed variable with respect to time. The relationship between temperature, precipitation, and their temporal resolution showed a trend. Calculating the coefficient of determination and regression analyses are two statistical tests. To investigate the significant pattern of temperature and rainfall, the R² value was calculated. The Least-Squares Method was implemented to determine the regression line slope, and the Mann–Kendall test was employed to identify and evaluate the trend. The mean, standard deviation (SD), and coefficient of variation (CV) of precipitation and temperature were calculated in order to find any correlations or relationships. In addition, the regression model and the coefficient of determination (denoted by R²) were also used to ascertain the significance of yield, production, and acreage changes. These specific tests were chosen because of their ability to reliably identify patterns and connections in yield and climate data. These techniques take unpredictability into consideration while ensuring reliable results.

2.4. Mann–Kendall’s Test

When analyzing trends in climatological data, the Mann–Kendall’s test is an important non-parametric statistical tool [56]. The rationale behind using this test is to statistically assess whether there is an upward or downward trend in the studied variable over time. Mann developed the test in 1945, and Panda and Sahu [57] stated that it has been applied extensively to time series data related to hydrology and climate [58,59]. Two benefits come from using this test. First of all, it is a non-parametric test so the data do not have to be normally distributed. Additionally, because the time series is homogeneous, the test is not sensitive to sudden discontinuities. The data are independent and ordered randomly, and thus, H₀ (i.e., the null hypothesis) says that there is no trend in the dataset. H₁ (i.e., the alternative hypothesis) presupposes a trend, which is what H₀ is tested against.

2.5. Sen’s Slope Estimator Test

Another non-parametric technique called Sen’s Slope estimator is predominantly employed to ascertain the trend’s slope within a time series [60]. It is mainly used to discover trends in univariate time series. The main advantage in using this test is that it is fairly resistant to outliers. A trend’s real slope, or the amount that the variable changes annually, may be found using Sen’s non-parametric technique. The program used to conduct this test is called XLSTAT-2017. While a negative value signifies a declining trend, a positive number denotes a rising trend. The time series’ direction can be determined using Sen’s slope.

2.6. Regression Models

Basically, regression model analysis identifies the influence of each independent variable on the dependent variable without affecting the other variables involved in the study. In other words, it can be said that regression models help prevent spurious correlations by controlling the confounders. Regression analysis models and the coefficient of determination (R²) are the two statistical techniques used to assess the importance of area, production, and yield trends [61]. Data can be expressed as a linear function of time using linear trend estimation. Finding significant variations in a group of data connected by a categorical element can also be achieved using a linear trend. Y = aX + B is the linear trend equation, where “a” denotes the slope and “b” is the Y-intercept.

The best-fit curved line that is most helpful when the data’s rate of change spikes or falls suddenly before leveling out is the logarithmic trend line. Both positive and negative values can be used in a logarithmic trend line. Thiscan be expressed as an equation:

lnYi = lnB1 + B2lnXi + ui

Here, B1 and B2 are linear and can be used in Ordinary Least Squares (OLS) regression as well as the logarithms of the variables X and Y. These models are also known as log-log, double-log, or log linear models because of their linearity. The primary selling point of this widely used model is its ability to quantify Y’s elasticity with regard to X using the slope coefficient B2 or “the percentage change in Y for a given percentage change in X”.

An exponential trend line is usually helpful when the dataset varies or progresses at a very faster rate. If there are any zero or negative values in the data, then it is not possible to construct an exponential trendline. In general, (1) absolute changes, (2) relative changes, or (3) percentage changes or growth rates are provided by the log linear trend line. One can take into consideration the following model, which is also known as the exponential regression model:

Yi = B1 × X^b2 ie^ui

2.7. Sustainability Index

The formula to compute the Yield Anomaly Index is YAI = (X-mean)/SD, where X is the crop yield for each year and SD is the standard deviation. The sustainability index of crop production was computed using the Geometric Growth Index approach [62]. The sustainability index is based on yields and cultivated land. Production is deemed sustainable when the Yield Index exceeds the Land Index, and is unsustainable if the Land Index exceeds the Yield Index. The formula to determine the geometric growth index is given below:

It = √ k (1 + p1) (1 + p2) … (1 + pk)

where It is the index of the output and input used in period t, p1 is the percentage growth rate between year t and t − 1, and k = n − 1 (where ‘n’ is the number of observations).

3. Results

The tables below (i.e.,

Table 2. Summary statistics of rainfall (cm) from 1991 to 2023.

	Minimum	Maximum	Mean	SD	Kurtosis	Skewness	CV (%)
January	0.00	5.36	1.80	1.69	−0.80	0.73	93.88
February	0.22	19.29	3.32	3.98	8.57	2.77	119.87
March	0.73	15.06	6.99	3.94	−0.90	0.19	56.36
April	3.95	33.01	15.56	6.35	1.01	0.61	40.80
May	17.04	42.97	26.65	6.79	−0.53	0.58	25.47
June	32.46	98.44	56.30	16.33	0.89	0.91	29.11
July	42.98	111.72	72.74	19.74	−0.60	0.49	27.13
August	34.43	86.22	57.10	13.65	−0.66	0.25	23.90
September	17.76	74.73	44.32	14.14	−0.83	0.23	31.90
October	2.98	39.30	14.90	9.58	−0.08	0.82	64.29
November	0.00	15.11	1.68	2.77	18.11	3.98	164.88
December	0.00	5.57	0.90	1.18	7.16	2.40	131.11

,

Table 3. Summary statistics of maximum temperature (°C) from 1991 to 2023.

	Minimum	Maximum	Mean	SD	Kurtosis	Skewness	CV (%)
January	17.46	22.05	20.01	1.09	0.01	−0.27	5.44
February	20.58	25.94	22.91	1.41	−0.58	0.08	6.17
March	24.15	28.91	26.94	1.09	0.29	−0.45	4.03
April	26.42	31.31	29.00	1.25	−0.70	0.10	4.31
May	27.82	31.77	29.24	0.93	0.17	0.60	3.19
June	28.20	31.23	29.50	0.75	−0.38	0.57	2.53
July	28.24	30.98	29.21	0.70	−0.28	0.55	2.41
August	28.64	31.05	29.66	0.64	−0.13	0.60	2.15
September	28.03	30.53	29.10	0.71	−0.86	0.50	2.45
October	26.98	30.03	28.24	0.70	−0.02	0.09	2.47
November	24.64	27.05	25.81	0.61	−0.51	−0.09	2.38
December	20.27	23.83	22.17	0.91	−0.52	−0.27	4.10

and

Table 4. Summary statistics of minimum temperature (°C) from 1991 to 2023.

	Mean	Minimum	Maximum	SD	CV (%)	Kurtosis	Skewness
January	8.13	6.09	9.75	0.87	10.65	0.14	−0.32
February	10.48	8.59	13.37	1.06	10.06	0.79	0.32
March	14.44	12.33	16.93	1.00	6.94	0.23	0.29
April	18.07	15.70	20.07	0.85	4.71	1.58	0.12
May	20.10	18.79	21.11	0.63	3.12	−0.45	−0.19
June	21.94	20.36	23.11	0.64	2.90	−0.02	−0.19
July	22.55	21.13	24.09	0.69	3.05	0.47	0.12
August	22.65	21.26	24.09	0.73	3.21	−0.53	0.29
September	21.78	20.40	23.62	0.78	3.59	−0.06	0.49
October	18.91	16.99	21.16	0.93	4.93	0.25	0.37
November	13.87	11.91	15.63	0.99	7.12	−0.19	−0.06
December	9.92	7.76	12.24	0.94	9.47	0.88	0.46

) depict a statistical summary of the rainfall, maximum temperature, and minimum temperature data, respectively. The maximum rainfall for the study area was observed in the month of July and the minimum rainfall was observed in November to January. Similarly, the maximum temperature was observed during the months of April and May and the minimum temperature was observed during December and January. SD shows the deviation or spread of the values around the mean. It indicates the variation from the average. For example, in the case of rainfall, the SD for the month of July was 19.74, which suggests a considerable amount of variability in the data points. The coefficient of variation can be defined as the ratio of the SD to the average, expressed as a percentage. It provides a measure of the relative variability.

The CV of the month of November, for instance, was 164.88% (Table 2), which suggests rather significant variability in relation to the mean. The distribution of values’ asymmetry can be measured using the skewness test. A data distribution that is symmetrical has a skewness value of 0. In contrast to negative skewness, which denotes a skewness to the left, a positive skewness indicates that the distribution is skewed to the right (tail is to the right of the peak). Kurtosis calculates the distribution’s “tailedness”. The distribution is normal if the kurtosis value is 3. The heavier the tails (more outliers), the larger the value; the lighter the tails, the smaller the value. In our results, the kurtosis for the maximum and minimum temperatures showed a normal distribution, but the existence of outliers was noted in February, December, and November.

Figure 2 depicts a slightly rising trend in the district’s rainfall, with an R² value of roughly 0.032 and a positive slope value (a = 0.065) in the linear regression equation. This linear regression model provides an explanation of the variation in the seasonal rainfall in the area, or R², which is the coefficient of determination.

Figure 3 and Figure 4 illustrate the regular growing trend in the seasonal maximum (max) temperature and seasonal minimum (min) temperature, respectively. The linear regression equation in Figure 3 shows a positive (+ve) slope value (a = 0.013) and the R² value is around 0.095. The linear regression equation in Figure 4 shows a positive (+ve) slope value (a = 0.042) and the R² value is around 0.556, explaining that 55% of the variability in the seasonal minimum temperature was determined using the linear regression test.

A box plot, sometimes referred to as a whisker plot (Figure 5), shows the distribution of a dataset graphically. It provides a thorough visual representation of a dataset’s distribution, including its skewness, central tendency, dispersion, and the presence of outliers. The dataset’s median, or the middle value when the data are sorted into increasing order, is represented by the line inside the box. Thus, exactly half of the observations are below the line while the other half of the data is below the line. Furthermore, none of the dataset’s three climatic parameters had an evenly distributed median line.

The whole box (Figure 6 and Figure 7) itself represents the spread of the mid 50% or exactly half of the examined dataset, which is calculated by subtracting the first quartile from the third quartile. A larger IQR indicates greater variability in the dataset. In our study, it seems that, in case of rainfall, a larger amount of data laid in the IQR compared to that of temperature. For a normally distributed dataset, around 50% of the dataset should lie within the box or the IQR (Inter-Quartile Range); this is important for displaying the variability. If the size of the box or the IQR is large, then it suggests that more than the 50% of the dataset lies within the box and vice versa. Whiskers indicate the skewness in the data distribution. If the whiskers are unequal in length or if the median is not centered within the box, it suggests skewness. If the whisker of the lower quartile is longer than that of the upper quartile, it indicates negative skewness and vice versa for positive skewness. For all three climatic parameters, the datasets were positively skewed. The width of the box and the length of the whiskers provide information about the spread and variability of the dataset. A wide box and a longer whisker indicate greater variability, while a narrow box and shorter whiskers indicate less variability.

An understanding of the relationships and patterns seen in the seasonal rainfall, maximum temperature, and minimum temperature is given in

Table 5. Mann–Kendall trend analysis of climatic parameters (1991–2023).

	Rainfall	Maximum Temperature	Minimum Temperature
Kendall’s Tau	0.48	0.59	0.63
Sen’s Slope	0.25	1.02	4.38
S	48	56	350
p-Value	0.001	0.026	0.006
Alpha	0.05	0.05	0.05

, which also shows the noteworthy trends and correlations over time. Kendall’s tau calculates the correlation between two variables. A value of 0.48 for seasonal rainfall is a high positive correlation value, which means that when one variable increased, the other one also tended to increase and vice versa. The seasonal maximum temperature had a somewhat positive connection (p = 0.59), meaning that when one variable rose, the other also tended to rise, but not significantly. The Kendall’s tau value of 0.63 for the seasonal lowest temperature likewise points to a positive correlation. The Sen’s slope is another technique to assess the strength and direction of trends in time series data. The seasonal rainfall showed a notable upward tendency over time, as seen by its positive slope of 0.25. The positive slope of 1.02 for maximum temperature indicates a seasonal maximum temperature. The seasonal minimum temperature likewise had a slightly rising trend, with a positive slope of 4.38. The computed Sen’s slope is represented by the S value. It adheres to Sen’s slope’s comparable trend. This indicates that the S value will be negative if the Sen’s slope for a given variable displays a downward trend and vice versa. The table displayed above demonstrates this. The chance of witnessing the data under the null hypothesis is indicated by the p-value. The observed link is thought to be significant if the p value is low, usually lower than the selected significance level, alpha, or 0.05. Since all of the p-values here were less than 0.05, the relationships that were identified are statistically significant.

A trend analysis was also conducted for both the area and production of tea and three types of trend lines (Figure 8) were drawn: linear, logarithmic, and exponential trend lines.

The trend analysis of the tea cultivated area and production, as shown in Figure 8 and Figure 9, respectively, showed that the trend in each analysis (linear, logarithmic, and exponential trends) followed a negative trend as the slope value was negative in each case. That means after one unit of increase in X, i.e., one year, there was a decrease in Y, which is production or area. In the case of a linear trend line, the slope value was −176.9 (Figure 9), which means that one unit increase in time caused a −176.9 decrease in production. The same trend was observed in the logarithmic and exponential trend analyses of tea production. After that, the value of R² in each of the three trend lines showed significant variation (0.661 in the linear, 0.630 in the logarithmic, and 0.544 in the exponential line (Figure 1)), which indicates that a significant proportion of variable Y (production) can be explained by variable X (time). For the area used for tea cultivation, a similar trend was seen (Figure 8). In the case of the logarithmic trend line for the tea area, the slope value was −393.0 (Figure 8), which means that one unit of increase in time caused a −393.0 decrease in area. The same trend was observed in the linear and exponential trend analyses of the tea area.

Table 6. Descriptive statistics of tea area and production.

	Maximum	Minimum	Mean	SD	Kurtosis	Skewness	CV (%)
Area	20,085	17,228	18,130.33	649.72	1.51	1.28	3.58
Production	13,932	3210	9352.58	2103.30	1.42	−0.41	22.49

discusses the descriptive statistics of the tea production and area, which include the average, SD, CV, kurtosis, and skewness. These statistics were calculated for 1991 to 2023. The computed table indicates a clear left-tailed distribution and the highest relative variability was observed in tea production. SD displays the variance or distribution of values around the mean. It displays the departure from the mean. For instance, the standard deviation for the tea area was 649.72, indicating a significant degree of variability in the datasets. The ratio of the standard deviation (SD) to the mean, shown as a percentage, is termed the coefficient of variation (CV). It serves as a measurement of relative variability.

Table 6 shows the tea production coefficient of variation, which was 22.49%, which is comparatively high in relation to the mean. The asymmetry in the value distribution is measured by skewness. A symmetric distribution is suggested by a skewness value of 0. When the distribution is positively skewed (tail is to the right of the peak), it is skewed to the right; when it is negatively skewed, it is skewed to the left. Kurtosis calculates the “tailedness”. A normal distribution is indicated by a kurtosis value of 3. More outliers or heavier tails are indicated by values larger than 3, whilst lighter tails are indicated by values less than 3. In our results, a normal distribution was shown by the kurtosis for both the tea area and production.

The impacts of climate-related fluctuations and changes on vulnerable socioeconomic sectors like agriculture need to be assessed, as these issues have become the most pressing global concerns. Furthermore, by comprehending the current regional climate patterns, we can better prepare for consequences of future climate change on agricultural outputs. The most important factors affecting agricultural conditions out of all the variables that affect climatic variability are temperature and rainfall. This also applies to the district under investigation. The association between temperature anomalies and rainfall anomalies and tea yield anomalies is depicted in Figure 10. Furthermore, the figure demonstrates that both temperature and tea yield normally have a direct correlation with rainfall; that is, if one rises (that is, the independent variable (rainfall)), the other increases too (that is, the dependent variable (tea yield)), and vice versa. Years with temperature deviations of more than 10% of the mean and years with rainfall deviations were computed and compared to years with an insufficient tea output.

Table 7. Years with yield deficits (>10%), and rainfall and temperature deviations.

Years with Rainfall Deviation > 10%		Years with Temperature Deviation > 10%		Years with Yield Deficiency > 10%
1992	−22.26	1992	11.86	1992	−34.51
1994	−25.37	1996	15.67	1994	−20.54
1996	−12.94	2006	12.35	1996	−13.45
1997	−16.56	2013	14.56	2000	−12.87
2000	−10.31	2017	15.37	2006	−13.56
2005	−12.05	2018	19.32	2013	−23.45
2006	−12.40	2019	21.87	2016	−14.90
2013	−13.65	2020	14.56	2017	−16.92
2014	−13.90	2021	18.63	2018	−18.32
2015	−11.39	2022	43.93	2019	−15.45
2016	−15.34	2023	30.87	2020	−28.83
2021	−11.09			2021	−25.54
2022	−10.86			2022	−26.39
2023	−10.87			2023	−36.16

clearly shows the years with rainfall deficits and years with surpluses in temperature above 10% of the mean, together with years with deficiencies in agricultural productivity. It was discovered that the years with rainfall deviations had lower tea yields (Table 7). Similarly, temperature increases of more than 10% of the mean caused a decrease in tea yields. The difference in rainfall had a greater effect than the temperature variation, even when the amounts varied. For instance, Table 7 illustrates that in 2023, the temperature deviation was 30.87% and the rainfall deviation was −10.87%. These factors combined to produce a decrease in tea production of −36.16 percent for the year.

The tea yield anomaly correlations with temperature and rainfall anomalies indicated that the crop had a negative relationship with temperature anomalies and a positive relationship with rainfall, i.e., a rainfall of 0.56 and temperature of −0.45 (

Table 8. Linear correlation and regression analysis between climatic parameter and crop yield anomalies.

	Rainfall Anomalies	Temperature Anomalies
R	0.56	−0.45
R2	0.68	0.49
p value	0.01	0.02
Significance	Significant	Significant

). Significant variations caused by the independent variable are determined using the coefficient of determination or R². In our results, the rainfall anomalies accounted for 68% of the variation in the tea yield anomalies, while the temperature anomalies accounted for 49%. Additionally, both outcomes exhibited statistical significance.

Table 9. Crop sustainability.

	Area (ha)	Total Production (Thousand kg)	Yield	YAI	Land Index	Sustainability Index
1991	20,085.00	13,932.00	0.69	1.61	3.01	Unsustainable
1992	19,309.00	12,355.00	0.64	1.13	1.81	Unsustainable
1993	19,324.00	13,026.00	0.67	1.44	1.84	Unsustainable
1994	19,280.00	11,092.00	0.58	0.54	1.77	Unsustainable
1995	18,932.00	11,298.00	0.60	0.74	1.23	Unsustainable
1996	17,551.00	10,614.00	0.60	0.81	−0.89	Sustainable
1997	17,760.00	10,054.00	0.57	0.46	−0.57	Sustainable
1998	17,830.00	10,253.00	0.58	0.54	−0.46	Sustainable
1999	17,604.00	8653.00	0.49	−0.22	−0.81	Sustainable
2000	17,228.00	9281.00	0.54	0.21	−1.39	Sustainable
2001	17,453.00	9841.00	0.56	0.44	−1.04	Sustainable
2002	17,463.00	9180.00	0.53	0.09	−1.03	Sustainable
2003	17,580.00	9582.00	0.55	0.27	−0.85	Sustainable
2004	17,522.00	10,065.00	0.57	0.53	−0.94	Sustainable
2005	17,539.00	11,312.00	0.64	1.17	−0.91	Sustainable
2006	17,542.00	10,854.00	0.62	0.93	−0.91	Sustainable
2007	17,818.00	10,007.00	0.56	0.42	−0.48	Sustainable
2008	17,818.00	11,586.00	0.65	1.22	−0.48	Sustainable
2009	17,961.00	8910.00	0.50	−0.18	−0.26	Sustainable
2010	17,961.00	8630.00	0.48	−0.32	−0.26	Unsustainable
2011	17,948.00	9140.00	0.51	−0.06	−0.28	Sustainable
2012	17,948.00	8930.00	0.50	−0.16	−0.28	Sustainable
2013	18,033.00	9130.00	0.51	−0.09	−0.15	Sustainable
2014	18,033.00	8510.00	0.47	−0.40	−0.15	Unsustainable
2015	18,033.00	8760.00	0.49	−0.27	−0.15	Unsustainable
2016	18,378.00	8130.00	0.44	−0.66	0.38	Unsustainable
2017	18,820.00	3210.00	0.17	−3.13	1.06	Unsustainable
2018	18,258.00	7690.00	0.42	−0.86	0.20	Unsustainable
2019	18,258.00	7960.00	0.44	−0.72	0.20	Unsustainable
2020	18,258.00	6700.00	0.37	−1.35	0.20	Unsustainable
2021	18,258.00	7010.00	0.38	−1.19	0.20	Unsustainable
2022	18,258.00	6930.00	0.38	−1.23	0.20	Unsustainable
2023	18,258.00	6010.00	0.33	−1.69	0.20	Unsustainable

showing the sustainability index for tea in the district highlights the agricultural dynamics and the transition towards unsustainability over the years. In the early 1990s, the crop yield indices consistently fell below that of the base year, indicating low yields. Despite some fluctuations, the land use indices remained relatively stable, implying consistent land utilization. However, the sustainability indices consistently portrayed sustainability during the late 1990s and early 2000s, indicative of challenges in maintaining sustainable agricultural practices. The transition period (2010–2015) saw mixed trends, with slight improvements in crop yields but sustainability had still not been attained. From 2014 onwards, a remarkable shift occurred, with negative crop yield indices and a transition to unsustainable practices, evidenced by negative sustainability indices.

4. Discussion

This study highlighted the substantial influence of climatic variability on Darjeeling tea by highlighting the relationship between rainfall patterns and rising temperatures. Regarding the effect of climatic parameters on tea production, high temperatures during the summer and monsoon seasons affect the tea yield and in contrast to this, higher temperatures during the winter and high rainfall during summer and winter months are beneficial for tea yields [48]. In our study, it was found that the maximum rainfall occurred during the monsoon months only, and the minimum temperatures occurred during winter months, which proved to be detrimental to the tea yield growth in our study area. Although the results showed that rainfall was increasing, there was seasonal variation in the rainfall. The same was true for temperature. Both climatic parameters affected the tea yield in a negative way because the required rainfall and temperature were not occurring in the required months in the study area, and hence, the tea yield declined over the years and unsustainable yields were obtained in the study area.

While tea production normally benefits from rainfall, the potential gains are negated by its unpredictable nature and rising temperatures. Even though rainfall anomalies might be advantageous in some situations, they provide yields that are not stable. The unique taste of tea is compromised by excessive rain at undesirable times.

The increasing temperatures and changing rainfall patterns due to climate change have decreased the appropriate planting zones for coffee in Brazil, resulting in lower production [3]. Similar patterns have been noted in Bangladesh and Ethiopia, where changes in seasonal rainfall have affected the quality of the leaves and the amount of tea produced [22]. Also, globally, several crops have suffered from the higher temperatures and erratic rainfall, which have reduced yields and degraded quality [4,24,63,64,65]. According to the results of our study, more rainfall is not enough to offset the negative consequences of warming temperatures. This demonstrates the urgent need for adaptive treatments that are specific to the Darjeeling region’s difficulties. The study highlights the issues plaguing agricultural systems due to climate change and variability [66]. Therefore, the creation of tea varieties that are climate robust becomes an essential measure to lessen the negative consequences of these changes in the environment. For production to be maintained, breeding strategies emphasizing heat and drought resilience should be stressed. Although rainfall patterns are irregular, yields can be stabilized by better resource management practices including rainwater gathering and irrigation system optimization. Additionally, a route to resilience can be found by fusing contemporary scientific methods with traditional farming expertise.

5. Conclusions

As rising temperatures and altered rainfall patterns have a negative impact on tea yields, this study concludes by highlighting the crucial influence that climatic variability has on the yield and overall production of tea in the Darjeeling region. According to the findings, this region’s tea business is more vulnerable to climate change and climate variability, since the rise in temperatures has caused a reduction in tea output despite an overall rise in rainfall. As tea production in Darjeeling supports thousands of workers’ lives and the area’s economy, this is especially problematic from a socioeconomic standpoint.

According to the findings, stakeholders including local farmers, legislators, and the international tea industry must move quickly to create and put into practice adaptive methods that might lessen the adverse consequences of climate change. In order to find more hardy tea cultivars, improve water management strategies, and investigate cutting-edge farming methods that may support tea production in the light of shifting climatic conditions, research and development expenditures are essential. This study also emphasizes the need for a holistic strategy for climate adaptation, one that supports the Darjeeling tea community via social and economic policies in addition to agricultural techniques. In order to increase tea industry sustainability, future efforts should assimilate traditional knowledge and ecological practices with scientific insights. Also, for robust and varied growth, emphasis should be placed on artisan techniques, natural farming, tree protection, and community-driven approaches. Stakeholders must also collaborate to guarantee the sustainability of the local tea industry and protect the lives of those who depend on this industry. Future studies should focus on integrating local and scientific knowledge to heighten rural resilience, thereby levelling ecological sustainability with economic growth. Tea biodiversity, scalable environmentally friendly techniques, and the influence of consumer trends and community networks are important issues.

These results demonstrate that the problems caused by climatic variability are not unique to Darjeeling but rather represent a worldwide concern that calls for concerted action on the local, national, and international scales. The viability of global tea production depends on addressing these issues, and it is the responsibility of all parties involved to act now to protect this essential sector from the impending effects of climate change.