Trend and change-point analyses of meteorological variables using Mann–Kendall family tests and innovative trend assessment techniques in New Bhupania command (India)

Climate change (CC) has risen as a critical issue of concern in recent years, encompassing a wide range of fields including biodiversity conservation, food security, and environmental sustainability. As temperature rises and the water cycle experiences variations, along with an increase in sea levels and the frequency of extreme events such as floods and droughts, the effects of CC are becoming more apparent (Airon et al. 2018; Chauhan et al. 2022). The fundamental components of climate, such as temperature, precipitation, wind speed, and humidity, are among the key variables that are undergoing change (Mundo-Molina 2015; Shaloo et al. 2022). The behavior of these crucial hydro-climatic variables is not uniform across regions and exhibits unique localized patterns. According to climatic models’ simulation and empirical evidences, rising temperatures result in an increase in atmospheric water vapor, leading to more intense precipitation events and the potential for devastating floods (Arora 2019; Dwivedi et al. 2019; Aish et al. 2021; Fuglie 2021). The Intergovernmental Panel on Climate Change (IPCC) apprised that the Indian subcontinent will experience the adverse effects of increased climatic variability, rising temperatures, and a decrease in summer rainfall by the 2020s, resulting in water stress (Ahmad et al. 2020; IPCC 2022). The study of CC in India has revealed notable findings, including a warming trend of 0.68 °C per century, fluctuations in rainfall patterns across the nation, an increase in extreme rainfall events in the Northwest during the summer monsoon, a decrease in the number of rainy days along the east coast, among other impacts (Aggarwal et al. 2004).

In India, the southwest (SW) monsoon is a source of a majority of the country's precipitation, accounting for 80% of the annual total rainfall (Prasanna 2014; Bisht et al. 2018; Dwivedi et al. 2019). The changing patterns, frequency, and variability of the SW monsoon have significant implications on agricultural productivity, water resource management, and the nation's economy. The rise in temperature and short, intense rainfall events result in high evapotranspiration (ET) from crop fields. However, the increase in atmospheric CO₂ and humidity tends to reduce transpiration, countering the effects of higher temperatures on ET (Rind et al. 1990; Pan et al. 2015). The combination of changing rainfall patterns, increased temperatures, and elevated ET rates is expected to cause water stress in plants. To ensure efficient water resource management and improved readiness for natural disasters like floods and droughts, it is indispensable to understand the changing patterns of climatic variables (Mohan & Ramsundram 2014; Nistor et al. 2016; Radhakrishnan et al. 2017).

Trend analysis is a crucial aspect of hydro-meteorological research that helps in identifying patterns in data over time. It refers to the general trend or progression of a set of data over a long period, characterized by a persistent change in the value of the dependent variable (Hawkins & Weber 1980). To assess the potential repercussions of shifts in temperature and precipitation patterns, researchers have been meticulously analyzing trends and change points within climatic variability across local, regional, and global scales. Notably, the focus on time-series modeling and change-point (CP) identification has gained substantial prominence owing to the rapid evolution of environmental circumstances. Within this context, the utilization of parametric or non-parametric methods under a statistical framework has proven invaluable in ascertaining whether a given dataset adheres to a certain distribution or exhibits a trend at a predefined level of significance. Diverse non-parametric tests, such as the Mann–Kendall (MK), Modified Mann–Kendall test (MMK), Innovative Trend Analysis (ITA), and Innovative Polygon Trend Analysis (IPTA),and homogeneity tests such as Pettitt's test, Standard Normal Homogeneity Test (SNHT), Buishand Range Test, and Von Neumann Ratio Test (VNRT) have emerged as widely adopted tools for detecting trends and pinpointing change points within historical sequences of climatic and hydrological variables. These tests hold the capability to illuminate trends and abrupt changes that may signify important shifts in these crucial environmental parameters. Kumar et al. (2010) analyzed the monthly, seasonal, and annual trends of rainfall using monthly data series during the period 1871–2005 for 30 sub-regions in India and found significant decrease in the SW monsoon and increase in the post-monsoon season in Kerala state. Regional trends in the monsoons were observed in the past century by Kumar et al. 1994; Guhathakurta & Rajeevan 2008; and Krishnakumar et al. 2009. The daily maximum temperature (Tmax) and minimum temperature (Tmin)-related studies revealed that the trends in temperature analysis are similar to what have been reported worldwide. Mamta et al. (2020) found that there was an annual average day time difference in the state of −0.78 °C and also the trend of rainfall indices indicated that there was an increase in the rainfall intensity in the Bawal district, Haryana. Several studies in India have documented an increasing trend in Tmax and Tmin over time, as reported by Malik & Singh 2019; Akhoury & Avishek 2020; and Srivastava et al. 2021. Studies have found that there has been a noticeable increase in mean annual temperature, with a rise of 0.05°C per year from 1971 to 2003 (Kothawale et al. 2016). Prabhakar et al. (2017) used MMK test for assessing the long-term rainfall variability over Champua watershed, Odisha. The results indicated that the MMK test was also more efficient in detecting trends in the study region. Lornezhad et al. (2023) used MMK test to conduct a case study in Lorestan province, Iran, for analysis of precipitation and drought trends. The results of research on an annual scale showed that all stations have a significant negative trend at the level of 5%. Singh et al. (2021) employed the ITA methodology to examine rainfall patterns across India spanning from 1901 to 2019. Their findings revealed notable trends in both seasonal and yearly rainfall across nearly all Indian sub-divisions. However, decline in precipitation was identified in the central northeast region of the nation. Additionally, the analysis indicated diminishing winter rainfall trends in a majority of the country's sub-divisions. Sharma & Ghosh (2022) conducted a study in the Haridwar district of Haryana by considering average temperature (monthly) and rainfall data (monthly) for 40 years (1981–2020). In the analysis, IPTA method was used and the results revealed that Tmin and precipitation vary by years, and Tmax data have a homogenous behavior.

Salarijazi et al. (2012) investigated streamflow data from the Ahvaz hydrometric station, representing the Karun watershed. They utilized Pettitt's test for CP detection and the MK test for trend analysis. While Pettitt's test revealed no change point, the MK test indicated a clear increasing trend in the data. Furthermore, research conducted in various parts of Haryana showed increasing trends in both monthly maximum and annual rainfall. Additionally, the results of other studies also indicated a significant spatial variability (Malik & Singh 2019; Nain & Hooda 2019; Verma & Ghosh 2019; Singh et al. 2020). The increased frequency and severity of heavy precipitation during the Kharif season is primarily responsible for the rise in monthly precipitation. The homogeneity tests (Pettitts test, SNHT, Buishand range test, and VNRT) were used to analysis of rainfall data. The results showed that the maximum daily rainfall data of the Kano gauging station is homogenous and classified as useful (Mohammad et al. 2021). Similarly, homogeneity test was used for ground water level change detection point for various locations in Haryana. It was found that maximum change points occurred between 2000 and 2006 (Singh et al. 2020). The future CCs are being predicted by analyzing the variability of climatic parameters using statistical trend analysis techniques. These predictions, based on historical patterns of hydro-climatic variables, will help in developing measures to mitigate the negative effects of CC on water resources, from drought, soil moisture deficit, reduced crop yields, and water logging.

Predictions for future CC and associated variables can assist in preparing more effectively for future irrigation and ground water needs. Strategies, including flood management, in situ soil moisture management, drought resistance, managing waterlogged lands, and water-saving crops, can help mitigate the impacts of CC on water resources. Similarly, studying the Bhupania minor canal command, Dulhera distributary, and Western Yamuna Canal Command (WYCC) in India is crucial, particularly in the context of CC. These canal systems are essential for irrigation, providing water to agricultural lands and communities. With CC altering precipitation patterns and increasing the frequency of extreme weather events, understanding the dynamics of these canal systems becomes even more important. By conducting studies on these canal commands, researchers can assess how CC affects water availability, distribution efficiency, and crop yields. Furthermore, understanding these canal systems can help identify adaptive strategies to cope with changing climatic conditions, ensuring resilience in agricultural practices and addressing water scarcity challenges in the region. Keeping in view the significance of the variability of climatic factors and their impacts on the crop yield, the present study was undertaken. The purpose of this research is to identify trends and break-points in climatic variables in the Bhupania region of WYCC in India, utilizing climatic data from 1990 to 2020.

The trend analysis involves examination of the long-term changes in the variable over an extended period, and capturing the gradual shifts that occur in the data series over time. The study was conducted using monthly precipitation, Tmax, and Tmin gridded data procured from Jhajjar district from the Indian Meteorological Department (IMD) for the period 1990–2020. In this study, trend analysis was performed on historical meteorological data using non-parametric methods such as the MK test, MMK, ITA, IPTA, Sen's slope estimator, and Homogeneity tests. In this investigation, our focus lies in discerning trends within meteorological data of the New Bhupania minor command. In 1990, the Food and Agricultural Organization (FAO), along with the International Commission on Irrigation and Drainage, and the World Meteorological Organization, collaborated to develop a novel method for estimating Reference Evapotranspiration (ET₀) known as the FAO Penman–Monteith Equation. Detailed guidelines for this estimation were outlined by Allen et al. in 1998, which were then utilized for calculating ET₀ in our study and subsequently employed for trend analysis. The null hypothesis (H₀) posits the absence of trends, while the alternate hypothesis (H₁) contemplates upward or downward trends. Our examination entails diverse trend tests applied to precipitation, Tmax and Tmin, and evapotranspiration data. These analyses are conducted at a significance level of 5%, aiming to unveil significant trends in this specific region of Haryana, India.

The MK test is a widely used non-parametric statistical test for finding trends in climatological and hydrological time-series data. This test, proposed by Mann in 1945 (Mann 1945; Kendall 1975), detects the presence of a monotonic trend in the mean or median of a time-series variable, regardless of the distribution of the data. The use of this test offers two advantages: first, its non-parametric nature means that it does not require data to follow a normal distribution, and second, its low sensitivity to abrupt interruptions makes it suitable for use with inhomogeneous time-series data. In the test, each data value is compared with all subsequent values, and a test statistic value (S) is calculated. Initially “S” is assumed to be zero. If a subsequent value is higher than the previous one, S is increased by 1, and if it's lower, S is decreased by 1. The final value of S is determined by the sum of all these operations. (Patle & Libang 2014; Pathak & Dodamani 2019; Anand et al. 2020). The null hypothesis assumed that there is no trend in time series (data are independent and random in nature). While, alternative hypothesis assumes that there is a trend in data. The detailed analysis of this method could be retrieved from Mann (1945).

Hamed and Rao introduced the MMK test as a method to discern trends within autocorrelated time-series data. Autocorrelation has a notable impact on the credibility of outcomes yielded by the MK test. The MMK test, designed to address limitations inherent in the MK test, employs variance correction approach. The autocorrelation of ranks is computed under significance level of 5%, thereby generating comprehensive experimental significance level. Notably, the utilization of the corrected variance offers notable advantages, as it harmoniously accounts for both normalized data and the autocorrelation function in the calculation of the corrected variance. The detailed procedure for MMK test can be found in Yue & Wang (2004), and Hamed & Rao (1998). In both MK and MMK tests, according to the specified significance level of 5%, if the computed p-value is less than or equal to the chosen α level of 0.05, the alternative hypothesis (H1) is accepted. This acceptance indicates the presence of a discernible trend within the data. Conversely, if the p-value exceeds α (0.05), the null hypothesis (H₀) is favored, suggesting the absence of any significant trend in the dataset. Based on the p-value, significant trend was assessed.

An ITA technique was introduced by Şen (2012). This method has found wide application in numerous studies to discern patterns within hydro-meteorological observations, and its accuracy has been assessed by comparing its outcomes with those derived from the MK method. In this approach, the dataset is bifurcated into two distinct classes, followed by independent arrangement of data points in ascending order. Subsequently, these two halves are plotted on a coordinate system. If the scatter plot of the time-series data aligns along the 1:1 (45°) straight line, this signifies an absence of trend. Conversely, data points situated above the trend line indicate an increasing trend, while those below it indicate the decreasing trend. Conversely, when data points cluster closely around the 1:1 linear relationship, this suggests the absence of a statistically significant trend. The magnitude of the trend within the data series can be ascertained by determining the mean difference between Xi and Xj. Notably, the initial observed data point was excluded from this classification procedure owing to the total count of 31 observed data points spanning 1991–2021. This form of graphical representation expedites a rapid visual assessment of time-series trends. The first half of the time series is employed to identify changes in trend. A detailed description of the ITA test can be found in Sen (2012), Sen (2017), and Shahfahad Talukdar et al. (2022).

Here, ‘s’ represents the trend slope, while |AB| signifies the trend length. The variables ‘x₁’ and ‘x₂’ denote successive points in the first part horizontally, and ‘y₁’ and ‘y₂’ do the same for the second part. When data lie above the trend line (1:1) it indicates the positive (Increasing) trend and vice versa. The polygon shape shows how the time series behaves over a year. Lines between months show how things change month-to-month. If the lines are similar, monthly changes have a small effect on the overall average change. Polygons get more complex with more dynamic weather events.

The detection of change points, which mark a sudden shift in a time series, is critical for determining the time interval during which the change occurred. In this study, several CP detection techniques were utilized, such as Pettitt's test, SNHT, Buishand range test, and VNRT.

The test statistic K can be evaluated against standard values at various confidence levels to identify change points in a time series.

In the above formula, and represent the mean and standard deviation of the time-series data, respectively. The year t can be considered as a change point or breaking point where the value of Tt reaches its maximum value. The null hypothesis can be rejected when the statistic value is higher than the critical value, which will depend on the sample size (n).

The calculated N value is compared with critical N values provided by Wijngaard et al. 2003. Null hypothesis of homogeneity is accepted when the calculated N value is greater than the critical N value.

The results of homogeneity tests, coupled with deciding whether to accept or reject the null hypothesis regarding Tmin, Tmax, Precipitation, and ET₀. The approach outlined by Wijngaard et al. (2003) was employed to identify change points for specific parameters.

• (a) No change point or homogeneous (HG): A series can be deemed homogeneous if, at a 5% significance level, either no test or just one out of four tests fails to reject the null hypothesis.

• (b) Doubtful series (DF): A series may be considered inhomogeneous and critically evaluated before further analysis if two out of four tests reject the null hypothesis at 5% significant level.

• (c) Change point or inhomogeneous (CP): A series could indicate the presence of a change point or demonstrate inhomogeneity if the null hypothesis is rejected by more than two tests at a significance level of 5%.

A flow chart of developed methodology to accomplish the objectives is presented in Figure 2. First, the study area was meticulously selected, and climatic variables were derived from the IMD dataset. Subsequently, preprocessing procedures were applied to eliminate missing data from the dataset. Once preprocessing was completed, the data were subjected to trend analysis utilizing the MK and ITA techniques. The identification of trends was conducted to ascertain the presence, magnitude, and direction of any discernible trends. The Sen's slope method was employed for further characterization of the identified trends.

The study analyzed the impacts of CC on trends, shifts, and variabilities in the data of Tmin and Tmax, precipitation, and evapotranspiration over a 31-year period for the New Bhupania minor command, WYCC. Non-parametric MK, MMK, ITA, IPTA statistical tests were used to analyze the climatic variables. Further, Pettitt's, SNHT, Buishand, and VNRT were used to detect the change point, along with Sen's slope estimator. The results of the trend analysis were considered significant at a 5% significance level.

The investigation pertained to the temporal evolution of Reference Evapotranspiration (ET₀) spanning the years 1990–2020. The resultant findings are systematically presented in Table 3. ET₀ computations were executed utilizing the FAO Penman–Monteith model, encompassing assessments for individual calendar months, annual aggregates, and seasons within the annual cycle. Notably, a statistically significant negative trend in ET₀ of −0.27 was ascertained in January, indicative of a declining trend. Conversely, the implementation of the MMK test revealed a significant positive deviation for April, registering an ET₀ upsurge of 0.49, suggesting an ascending trajectory in this specific month at 5% significance level in both methods.

To conduct CP detection analysis, we evaluated the mean monthly, annual, and seasonal (summer, rainy, and winter) datasets encompassing Tmin, Tmax, Precipitation, and evapotranspiration information. Statistical tests including Pettitt's test, SNHT, Buishand range test, and VNRT were performed.

Climatic factors such as temperature, precipitation, and ET₀ constitute pivotal variables that exert a profound influence on the well-being of individuals, researchers, and policymakers, especially those reliant on agricultural and allied domains. The dynamics of climate fluctuations possess the potential to exacerbate existing economic and social infrastructures, thereby underscoring the urgency of investigating prevailing trends and detecting pivotal change points. In this context, the present research delves into a meticulous analysis of climate variability and its ramifications on agriculture and allied activities in the New Bhupania minor command area. The methodology employed encompasses a comprehensive array of trend analyses techniques, including the MK, MMK, ITA, and IPTA. The examination of selected meteorological variables is rooted in the quest to identify discernible patterns and deviations. The application of the MK and MMK tests to the Tmax dataset revealed trends of varying significance, with the MMK test showcasing significance in September and July, aligning with climatic peculiarities.

However, the ITA and IPTA tests emerged as the most discerning and effective approaches, capturing significant trends across all months of the year. This affirms the superiority of the ITA and IPTA methods in detecting temperature-related trends, surpassing the other two methods employed. The resultant insights underline a prevailing trend of escalating annual average temperatures, as evidenced across all adopted trend analysis methodologies. Notably, the MK test was able to detect a trend in September, while the MMK test exhibited a similar acumen, pinpointing a shift in July. Impressively, the IPTA test yielded congruent results, reinforcing the robustness of the ITA methodology. Further scrutiny into average annual Tmin elucidated a consistent increasing trajectory, in consonance with all employed trend analysis methodologies. Augmenting this narrative, the MK test underscored increments in August and September, while the MMK approach depicted deviations in the months of March, July, August, and September.

Our findings align with previous trend analysis studies, demonstrating consistency with regard to the annual and seasonal mean temperature trends. Arora et al. (2005) conducted a similar investigation, observing an upward trend in 53 stations and downward trend in 17 stations in India. They reported an overall rise of approximately 0.42°C in annual mean temperature over a century. Dash et al. (2007) evaluated CC impacts in India using temperature data and noted temperature increases of 1°C in winter and 1.1°C in the post-monsoon period. Similarly, Attri & Tyagi (2010) concluded that temperatures had risen by 0.56°C from 1901 to 2009. This trend of increasing annual mean temperature, particularly beyond the 1990 baseline (1960–1990), was also noted in our study. Remarkably, our results indicate positive changes, contrasting with the earlier studies that reported no trends in Tmax and Tmin temperature changes. Variations in temperature exert multifaceted effects: an elevation in summer's maximum temperature contributes to increasing evapotranspiration (ET) rates and intensifies heatwaves. Conversely, temperature reductions during the winter season can induce chilling injury responses and hinder germination processes. Notably, fluctuations in Tmax, whether increasing or downward, exert detrimental repercussions on both agricultural crops and human well-being. Our findings are similar to that of various researchers across the globe, including Chattopadhyay & Edwards 2016; Kothawale et al. 2016; and Airon et al. 2018. This consensus strengthens the robustness of our findings and supports the notion of a broader trend toward increasing temperatures observed in various geographical contexts. By contrast, Rathore et al. (2013) explored annual and seasonal temperature trends by state level over the period 1951–2010, finding increasing trends in annual mean temperature for 21 states, ranging from 0.01 to 0.05°C/year, with some variations. While Jammu and Kashmir, Punjab, and Uttarakhand exhibited negative tendencies of −0.01°C/year, Punjab displayed statistical significance at the 5% level. Notably, Chhattisgarh, Haryana, Meghalaya, Orissa, Uttar Pradesh, and West Bengal showed no significant trends. This diverges from our findings, wherein positive trends were identified.

Precipitation trends were assessed using trend analysis tests. Most months and seasons exhibited insignificant precipitation trends according to the MK and MMK tests, with a few exceptions. However, the ITA and IPTA tests proved more adept at discerning trends. The four methods consistently depicted an increasing trend in mean annual precipitation, potentially leading to a future maximum. Elevated land and ocean surface temperatures, catalyzing evaporation, might be causally linked to increased rainfall. Goswami et al. (2006) identified a noteworthy rise in extreme rain frequency, while our study indicated an augmented rainfall during the Zaid season. Furthermore, Goswami et al. (2006) recorded significant negative rainfall in central India from 1951 to 2000, paralleling our post-monsoon findings. Esit (2023) utilized an ITA approach to contrast with the classical trend method regarding monthly and annual hydro-meteorological variables. It was determined that ITA, accompanied by significance testing, and the IPTA exhibit greater sensitivity compared with the MK test, particularly evident in precipitation series. Our findings corroborate with the existing literature, wherein ITA and IPTA were more adept at trend detection compared with MK tests. Buyukyildiz (2023) applied both ITA and statistical trend approaches to analyze annual precipitation in the Euphrates–Tigris River Basin (ETRB) of Turkey. The results indicated that while the classical MK test identified a statistically significant monotonic trend in precipitation, the ITA method detected a decreasing trend in annual total precipitation within the ETRB for the examined period. Hence, our trend analysis employing ITA aligns with this study's observation of a negative precipitation trend. Kulkarni (2012) outlined contrasting rainfall trends—increasing from 1901 to 1975 and decreasing from 1976 to 2004—diverging from our region's decreasing trend. Average rainfall exhibited a consistent increase across all trend analysis methodologies throughout the year. Chaturvedi et al. (2012), employing representative concentration pathways, projected a 4–5% rise by 2030 and a 6–14% increase by 2080 compared with the 1961–1990 baseline. Similar findings were reported in multiple other studies, including Patle & Libang 2014; Pingale et al. 2016; Chandniha et al. 2017; Meshram et al. 2017; Srivastava et al. 2021; Paramaguru et al. 2023; and Rajput et al. 2023. This convergence underscores the robustness and widespread implications of our rainfall trend findings.

The trend analysis of ET₀ values revealed a negative (decreasing) trend in January based on the MK, MMK, ITA, and IPTA methods. This primarily stems from decreases in Tmin and Tmax, which significantly influence ET₀ values. ET₀ hinges on atmospheric demand, primarily influenced by five key variables: Tmin and Tmax, relative humidity, wind speed, and sunshine hours. Our study identified a decline in Tmin during the post-monsoon months, resulting in decreased ET₀. A similar pattern was noted for the Rabi season, indicating reduced ET₀ values. However, attributing the decline in January's ET₀ solely to temperature is overly simplistic; a comprehensive analysis of wind speed and sunshine hours is essential to ascertain the full scope of ET₀ reduction. Interestingly, the pre-monsoon season also registered a decline in ET₀, possibly attributed to increased irrigation, cropping intensity, and advancements in crop management practices within the arid and semi-arid regions of India (Milesi et al. 2010). The reduction in ET₀ during the Rabi season could potentially lead to diminished primary productivity, a preference for sensible heat over latent heat, intensified surface heating, and heightened land–atmosphere interactions.

Additionally, Esit et al. (2023) conducted a similar investigation and discovered that MK and ITA with significance testing revealed an increasing trend in annual mean temperature. Our results diverge from Esit's findings concerning annual evapotranspiration, indicating a contradictory trend, where annual total evapotranspiration is seen as increasing. To better comprehend the ramifications of changing ET on hydrological and carbon cycles, as well as energy partitioning, future investigations must consider climate feedbacks. Similar outcomes have been highlighted in various studies, including those by Bandyopadhyay et al. (2009), Choudhary et al. (2022), Patle & Singh (2015), Pingale et al. (2016), and Zhang et al. (2021). These investigations have collectively demonstrated the impact of CC on evapotranspiration, strengthening the veracity of our findings and emphasizing the need for a comprehensive understanding of ET₀ dynamics.

The studies aim to identify change points through CP analysis, employing homogeneity tests for verification. Notably, a substantial number of change points were observed for Tmax, particularly in April 1997. Moreover, a significant cluster of change points emerged between 1994 and 1997. Similarly, for rainfall, prominent change points manifested in November, average annual rainfall, and during the Zaid season. The evapotranspiration (ET₀) series exhibited notable change points in January and May. These pivotal change points, detected from 1990 to 2013, are likely linked to the burgeoning industrial and commercial activities in the studied region. Jaiswal et al. (2015) obtained similar results, pinpointing a cluster of change points between 1900 and 2000. The driving forces behind this shift in climatic patterns encompass rapid urbanization resulting from population growth, anthropogenic activities, pollution, industrialization, greenhouse gas emissions, and global warming. These factors, when compared with the preindustrial era (1850–1900), have led to significant changes in climatic variables, yielding substantial impacts on livelihoods and global food security.

Upon comprehensive analysis, discernible increasing trends in both Tmax and Tmin were observed, contributing to an imbalance in the hydrological cycle. This imbalance necessitates proactive management to avert adverse consequences for water resources, food security, and the well-being of local populations and livestock. Notably, the study region may experience reduced post-monsoon rainfall, potentially affecting rice harvesting timing. This circumstance could challenge paddy growers as well as field preparation and other agricultural operations in meeting water demands during the post-monsoon season. Somehow, it will impact ground water, leading to reduction in the ground water level and its quality. Counteracting this, areas with decreasing rainfall trends could employ water harvesting techniques to mitigate moisture stress during the post-monsoon period and offer supplementary irrigation for Rabi crops. The subsequent strategies delineate further measures to address the adverse effects of climatic variables on the agricultural landscape of the study region.

The present study employed comprehensive trend analysis techniques, including the MK, MMK, ITA, and IPTA, to investigate hydro-meteorological parameters in the New Bhupania minor in the WYCC at both monthly and annual scales. Additionally, CP detection methods such as Pettitt's, SNHT, Buishand, VNRT, and Sen's slope estimator were utilized. The MK test successfully identified positive trends for maximum temperature in September and for minimum temperature in August and September. Similarly, a positive trend was observed in precipitation during the Zaid season. Furthermore, the MMK test detected significant differences in months compared with the MK test, such as July for Tmax and August and September for Tmin. Positive trends were identified in precipitation for March and April, along with negative trends in February. While MK and MMK failed to detect significant trends in the majority of variables, ITA and IPTA revealed distinct trends. These findings align with previous studies in the literature (Caloiero et al. 2018; Ceribasi & Ceyhunlu 2021; Hırca et al. 2022; Esit 2023). For all climatic variables examined in this study, the ITA and IPTA methods of trend detection consistently identified the highest number of trends compared with the MK and MMK methods. Specifically, the positive trends detected by the MK, MMK, and ITA methods were found to be 7.69, 7.69, and 84.61%, respectively, for Tmax, and 23.07, 38.46, and 76.92%, respectively, for Tmin. Similarly, for precipitation, the trends were 6.25, 25, and 56.25%, and for evapotranspiration, they were 6.25, 12.5, and 68.75%, respectively. Upon considering all three approaches, Tmin exhibited a significantly increasing trend for the months of August, September, and the average annual temperature. Precipitation during the Zaid season also displayed an increasing trend. However, a negative trend in evapotranspiration was detected in January, which could potentially impact plant biomass growth if it falls below critical levels. These findings underscore the importance of utilizing robust trend detection techniques such as ITA and IPTA for accurate assessment and monitoring of climatic trends, aiding in informed decision-making for various sectors including agriculture, water resource management, and infrastructure planning. CP analysis revealed shifts in climate variables, such as Tmin in April 1997, precipitation from 2012 onwards, and reference evapotranspiration in May 2009. The study underscores the importance of adaptive measures to mitigate CC impacts, improve infrastructure resilience, and assess risks to ecosystems and biodiversity. Future research should explore interactions and feedback loops between climatic variables to better understand future trends and their implications. Such endeavors will contribute to more effective decision-making and sustainable management of natural resources in a changing climate.

The authors are greatly indebted to the Division of Agricultural Engineering, Indian Agricultural Research Institute, New Delhi, and the Water Technology Center, New Delhi, for facilitating the conduction of the experiment.

Not applicable.

Not applicable.

Writing—original draft preparation, VG. AS, JR and M.K, writing—review and editing, DKS, JR, NLK and MK; help and discussion, VG, JR, NLK. All authors have read and agreed to the published version of the paper.

This research received no external funding.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.