Determination of Grid-Wise Monsoon Onset and Its Spatial Analysis for India (1901–2019)

Abstract

Monsoon onset in India has always been a topic of interest for the research fraternity and various stakeholders. This study aimed to determine the monsoon onset date at the grid point scale, to obtain the trend of monsoon onset, and to unravel the spatial distribution of monsoon onset during the period 1901–2019 (especially in different climate modes). Based on observed cumulative rainfall, the piecewise linear regression model (PLRM), which employs least-squares principles, finds changepoints that signify the beginning of the monsoon season with the onset of monsoon. In this study, monsoon onset is examined with respect to several climate modes to evaluate their impact on monsoon onset. Monsoon onset is delayed in El Niño and drought years due to strong negative anomalies that are revealed by a spatial examination of monsoon onset. However, because of local atmospheric circulation impacts, there are outliers. The study also reveals areas with notable monotonic tendencies in monsoon onset, suggesting future changes in onset dates. These areas need more sophisticated frameworks for developing mitigation strategies since they should be viewed as susceptible. The comparison of the PLRM outcomes with objective methods reveals a strong correlation, confirming the accuracy of the PLRM method. Overall, the PLRM has been shown to be a useful tool for predicting the start of the monsoon on fine spatial scales and may be used in conjunction with regional climate models to anticipate the start of the monsoon in various regions of India. The results of this study could have a significant impact on regional planning and policy initiatives for sustainable development.

1. Introduction

Monsoon onset is generally viewed with the isochrones passing through different regions, and it has been observed that those isochrones are not dense enough to provide the monsoon onset date at a finer spatial scale [1]. Various methodologies for the determination have been introduced, and in most cases, each one of them uses different criteria [2]. However, few methodologies are found available to provide a monsoon onset date in pan-India with a finer spatial scale [3].

The availability of the onset date of monsoon with fine spatial scale provides information that is highly relevant for the policy and plan-making exercise for the development of the area [4]. Determination of monsoon onset at a fine scale and calculation of the normal onset by creating the climatology of the calculated monsoon onset never provides a guarantee to the accurate prediction or estimate of the monsoon onset for meteorological purposes [5]. Due to the role of many factors responsible for influencing the monsoon onset and their highly dynamic nature with the local level atmospheric circulation, the piecewise linear regression model (hereafter PLRM) was used in this study.

PLRM was used earlier for identifying the change in the structure of climate variables, such as the annual trend of wind, rainfall, and some other important variables [6]. Liu et al. (2010) have also clearly highlighted in their study that the model is very efficient and can also be used in many other cases for identifying the trend and changepoint lying within the series [6]. The PLRM is based on the principle of least squares. In the present study, only one changepoint was the requirement to identify the onset of monsoon; however, the number of change points can also be increased by modifying the programming code. Tome (2004) used the FORTRAN 90 program to execute the model and advocated the efficiency of the model in climate change studies to secure non-deceptive results [7]. PLRM has been used in many studies to detect the shift in the phase due to the interaction between two or more variables. Deng et al. (2020) identified the shift in the phases of tree growth based on the change in response of leaf traits due to changes in the environment [8].

Walker and Bordoni (2016) determined the monsoon onset for each grid using the moisture flux and followed the method for determination provided by Cook and Buckley (2009) [9,10]. However, the limitation observed with the determination of monsoon onset in the methodology adopted by Walker and Bordoni (2016) is the non-availability of the high-resolution, long-term, reanalyzed dataset for the calculation of moisture flux. Moreover, Pai et al. (2020) have discussed a new method to obtain the grid-wise monsoon onset for a different group of grids with a certain threshold limit for identifying monsoon onset [11].

However, the objective of the present study was (a) to determine the onset of the monsoon for each grid using the observed dataset and (b) to perform the spatial analysis of the trend of monsoon onset.

2. Materials and Methods

2.1. Study Area

The study area lies with the coordinate extent 68.22° E–97.17° E to 8.07° N–32.88° N. The onset date calculation is performed for each grid lying between the mentioned geographical coordinate extent.

2.2. Data

Rainfall data for this study were obtained from the India Meteorological Department (hereafter IMD) [12]. The default grid dataset was obtained with 0.25° × 0.25° spatial resolution. However, to remove the possible influence of the stochastic event and highly uncertain circumstances at the local level at 0.25° × 0.25°, the grid dataset’s resolution was converted into 1° × 1° spatial resolution. This scheme of monsoon was also adopted by Walker and Bordoni (2016) [9] as it is a well-known conventional approach.

A list of El Niño, La Niña, flood, and drought (1901–2019) is obtained from the Indian Institute of Tropical Meteorology (hereafter IITM) (

Table 1. Year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 1901–2016. Positive and negative IOD years are based on the period 1960–2016.

Climate Mode	Years
El Niño	1902, 1904, 1911, 1913, 1918, 1925, 1930, 1935, 1944, 1951, 1957, 1965, 1968, 1972, 1976, 1982, 1987, 1991, 1997, 2002, 2006, 2009, 2015
La Niña	1908, 1909, 1910, 1916, 1933, 1948, 1949, 1954, 1955, 1970, 1973, 1975, 1988, 1999, 2000, 2007, 2010, 2011
Drought	1901, 1904, 1905, 1911, 1918, 1920, 1941, 1951, 1965, 1966, 1968, 1972, 1974, 1979, 1982, 1985, 1986, 1987, 2002, 2004, 2009, 2014, 2015
Flood	1910, 1916, 1917, 1933, 1942, 1947, 1956, 1959, 1961, 1970, 1975, 1983, 1988, 1994
Negative IOD	1960, 1964, 1974, 1981, 1989, 1992, 1996, 1998, 2010, 2014, 2016
Positive IOD	1961, 1963, 1972, 1982, 1983, 1994, 1997, 2003, 2012, 2015

). The list of negative Indian Ocean dipole (hereafter IOD) and positive IOD (1960–2016) is obtained from the Bureau of Meteorology, Australia, and also used in the study on spatio-temporal onset characteristics of Indian summer monsoon rainfall by Saini et al. (2022) [3] (Table 1). We detected exceptional occurrences of La Niñas, El Niños, drought years, flood years, and IODs. The occurrences of these phenomena together were disregarded. Not all onsets can be fully attributable to ENSO and IOD-related SSTs, despite the importance of SST-related changes. Because early/late onsets of floods and droughts cause them independently of ENSO and IOD, we regarded them as separate groups for analysis. The onset dates of some floods and droughts, however, may be impacted.

2.3. Methods

We used the PLRM method, taking advantage of its capacity to identify changepoints via least-squares ideas related to slope variations [13]. This technique successfully determined the start of the monsoon. We calculated the onset date exactly by computing the cumulative rainfall. The monsoon season, which greatly influences India’s yearly rainfall [14,15], is notable for causing a dramatic increase in rainfall to signal the beginning of the season. Many regression models are available to fulfill the requirement for many different purposes. PLRM is the linear model that inhibits the quality to accommodate multiple linear models and get them fit for different changepoints, i.e., ‘c’ in 0 1 (Figure 1). In terms of understanding the complicated dynamics of this complex climatic event, the PLRM represents a tremendous advance. The flexibility of PLRM allows it to capture nonlinear patterns and abrupt changes in monsoon onset, which improves modeling and analysis precision. This section explores the practical advantages of PLRM in comprehending monsoon onset. A more accurate interpretation, localized trend analysis, forecasting, the capacity to identify vulnerable phases, and data-driven insights are all made possible by PLRM’s versatility in capturing nonlinearity and identifying abrupt transitions. Here, the changepoint does not need to be known before the execution of the analysis. Regression gets broken at the changepoint location, and therefore, the function is written and performed in such a way that the continuity is maintained after the changepoint [16].

In the case of the one change point, the regression is written as (Equations (1) and (2)) shown below.

$$y=a_1+b_{1}x for x\le c$$

(1)

$$y=a_2+b_{2}x for x\le c$$

(2)

Whereas, to maintain the continuity after the present changepoint, Equations (1) and (2) must be equal to each other (when $x=c$) (Equation (3)).

$$a_1+b_{1}c=a_2+b_{2}c$$

(3)

As a part of the next step, the solution of one of the many parameters needs to be calculated (Equation (4)).

$$a_2=a_1+c(b_1-b_2)$$

(4)

In the end, replacing the value of $a_2$ using the equations provided above results in the continuous PLRM (where $x=c$).

$$y={\{a}_1+{c(b}_1-b_2\left)\right\}+b_{2}x for x\le c$$

(5)

In the study, the definition of the monsoon onset period includes the period from 1 March to 13 August. This spans the 225th day of the year (13 August) and the 60th day of the year (1 March). The 1 March to 13 August period was selected due to its significant accuracy after a thorough investigation was conducted to determine the most precise onset day for each grid. The monsoon onset day anomalies, estimated in relation to the base period extending from 1901 to 2019, are the focus of the spatial study. Anomaly magnitudes were divided into three levels, 0 to +4/−4 as low, +4/−4 to +16/−16 as moderate, and exceeding +16/−16 as high magnitude in their respective directions, to make interpretation easier.

Several statistical techniques were used in our inquiry, including the Sen’s slope test, the Mann–Kendall test (MKT), the innovative trend analysis (hereafter ITA), and the linear regression model (hereafter LRM) [14,17]. The MKT proved crucial in identifying areas of India with a monotonic monsoon onset trend. The size of observed trends in monsoon onset days was evaluated by Sen’s slope approach. ITA made it easier to find monotonic trends for various monsoon onset day values over all of India. LRM was also used to identify monotonic trends and variations in the value of the onset day. The monotonic trend in statistics generally means the consistently increasing or decreasing trend through time and has been used in climate studies also [14,17,18]

Additionally, LRM was essential in comparing grid-wise monsoon beginning dates determined by PLRM with dates acquired using the objective technique described and used by Saini et al. (2022) [3,19]. Following the SR boundaries defined by Singh and Ranade (2010) [19], the grid-wise monsoon dates were derived for each sub-region (SR) for this research using the field mean approach. This comparative analysis attempts to show how the objective approach and the monsoon onset values produced using PLRM are in agreement.

3. Results

The research article’s Results section provides an analysis of monsoon onset anomalies for various periods, including 1901–1925, 1926–1950, 1951–75, 1976–2000, and 2001–2019. During these times, the study looks into the connections between monsoon anomalies, El Niño, La Niña, drought, flood and IOD. The part also investigates the existence of monotonic trends in the monsoon start for the entirety of India and investigates spatial distribution using a variety of analytical techniques.

Grid-wise calculation of monsoon onset was completed, and the difference from the calculated normal onset (Figure 2) was also made as part of the spatial analysis of monsoon onset. Comparison between grid-wise monsoon onset obtained in the present study and monsoon onset using the objective method of IITM at the SR level shows a suitable association between both the monsoon onset and it is significant at the 95% level (Figure 3). Here, the only exception found was SR 15 (Figure 3s), and the reason might be the high interannual variation in the monsoon onset date in the SR and the scarcity of rainfall. To obtain the monsoon onset date using PLRM, the period from 60 to 225 Nth day was selected as the results obtained using this period are similar to the ones obtained using the objective method.

Pentad anomaly-based analysis of the anomalous monsoon onset was performed (Figure 4), and except for a few grid points, low to moderate magnitude was found at the grid points in the whole 1916–1920 (Figure 4d) pentad. Similarly, the analysis was performed for each event year (Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9) so that the pentad, like 1916–1920 (Figure 4d), could be understood in detail.

3.1. Events (1901–1925)

Research conducted between 1901 and 1925 has made important contribution about the complex relationship between monsoon onset anomalies and other climate modes, with a particular emphasis on the role of El Niño and drought. Years marked by the development of El Niño occurrences and pervasive drought conditions make the clear presence of unfavourable anomalies apparent. Because of the simultaneous convergence of El Niño and drought conditions, these unfavourable anomalies not only manifest extensively but also with increased intensity. This is supported by a perceptive spatial analysis, which is graphically represented in Figure 5 and vividly illustrates the cases of monsoon onset abnormalities within this particular era.

Table 2. The year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 1901–1925.

Climate Mode	Years
El Niño	1902, 1904, 1911, 1913, 1918, 1925
La Niña	1908, 1909, 1910, 1916
Drought	1901, 1904, 1905, 1911, 1918, 1920
Flood	1910, 1916, 1917

precisely delineates the years linked with several climate modes: El Niño, La Niña, flood, and drought, providing a comprehensive summary.

Particular years stand out among those in this chronological range because they were marked by the existence of significant and pervasive unfavourable abnormalities. The years 1911, 1913, and 1918, in particular, stand out for exhibiting such abnormalities. However, each of these remarkable years also happens to have El Niño conditions, a correlation that has been described in Table 2. A thorough study has also shown the occurrence of years that show a mixed anomalous pattern. The years 1904, 1909, 1910, 1916, 1917, 1919, 1920, and 1925 are notable examples of years where major positive anomalies coexist with partial negative anomalies. This detailed pattern shows the complex dynamics of monsoon behavior during El Niño and drought occurrences.

Importantly, the study highlights the persistent predominance of positive anomalies over large areas of India during years characterized by either El Niño or drought, emphasizing the significant regional effects of these meteorological occurrences. On the other hand, the study discovers cases of strong negative anomalies that point to the monsoon arriving earlier than expected. These events may be seen to have occurred in the years 1901, 1902, 1903, 1907, and 1912. In these situations, a strong negative anomaly of roughly −20 days denotes the monsoon’s early arrival over a sizable portion of the planet.

The period from 1901 to 1925 provides fascinating insight into the complex interplay between anomalous monsoon onset and the cascade consequences of El Niño and drought episodes. The data come together to highlight the ubiquitous and strong nature of positive anomalies during such climatic events, exposing the fascinating interaction between several climate modes and the complex dynamics of monsoon phenomena. This thorough understanding makes a significant contribution to our understanding of monsoon behaviour and forms a key building block for developing prediction models and efficient coping mechanisms for the effects of future climate change.

3.2. Events (1926–1950)

The investigation of the monsoon onset anomalies during the period 1926–1950 in the research article reveals noteworthy patterns that shed light on the interplay between climate modes and the monsoon phenomenon. The analysis pinpoints specific years within this period that exhibited distinct characteristics. In particular, years such as 1933, 1936, 1938, and 1941 display widespread positive anomalies of high magnitude, as illustrated in Figure 6. On the other hand, only a few years, including 1926, 1935, and 1944, stand out for their notable widespread negative anomalies of high magnitude. In the remaining years, the anomalies are generally of moderate or low magnitude.

One intriguing finding is the contrast between El Niño years and the magnitude of anomalies observed in different regions of India, as documented in

Table 3. The year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 1926–1950.

Climate Mode	Years
El Niño	1930, 1935, 1944
La Niña	1933, 1948, 1949
Drought	1941
Flood	1933, 1942, 1947

. Interestingly, despite El Niño conditions being associated with global climate patterns, the larger part of India does not exhibit high-magnitude anomalies during these years. The study also highlights a unique case in 1933, which is marked as a La Niña year. This year presents a positive anomaly of monsoon onset, indicating a potentially interesting interplay between La Niña and monsoon behaviour. Another notable observation emerges from the solitary drought year in this period, 1941. This year shows a high-magnitude monsoon onset anomaly. This finding adds to the growing understanding of the complex relationships between climatic factors and monsoon dynamics.

The research’s exploration of the period 1926–1950 brought to light distinctive patterns of monsoon onset anomalies. The identification of years with high-magnitude anomalies, the interactions between climate modes and anomalies, and special cases like the La Niña-associated anomaly and the drought year’s anomaly contribute to a deeper comprehension of the intricacies of monsoon behaviour. These insights have the potential to refine climate predictions and inform strategies for mitigating the impacts of monsoon variability.

3.3. Events (1951–1975)

The complex domain of monsoon onset anomalies throughout the period from 1951 to 1975 was examined with a particular focus on their complicated interaction with meteorological events such as El Niño and drought. This period is notable for the significant intricacy that defines the connection between climatic factors and the start of the monsoon period. It is important to emphasise that years characterised by the concurrent occurrence of El Niño and drought conditions demonstrate significant and severe departures from the typical initiation patterns of the monsoon.

In the year 1951, an event of El Niño occurred along with drought conditions in India (

Table 4. The year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 1951–1975. Positive and negative IOD is based on the period 1960–2016 and here for the period 1951–1975.

Climate Mode	Years
El Niño	1951, 1957, 1965, 1968, 1972
La Niña	1954, 1955, 1970, 1973, 1975
Drought	1951, 1965, 1966, 1968, 1972, 1974
Flood	1956, 1959, 1961, 1970, 1975
Negative IOD	1960, 1964, 1974
Positive IOD	1961, 1963, 1972

). The specific regions got affected including the peninsular region, northeastern parts, and western Uttar Pradesh. The effect of this event was notably adverse, as indicated by a negative anomaly, as seen in Figure 7. In contrast, the years 1965 and 1968 were designated as El Niño and drought years; nonetheless, they did not result in substantial consequences.

The time period from 1951 to 1975 was characterised by a limited occurrence of positive anomalies. This may be primarily attributable to the existence of a positive IOD in the northern areas of India, namely inside the Indo-Gangetic plain. Significantly, the years 1970 and 1971 had considerable positive anomalies, with 1970 being distinguished by the presence of La Niña conditions and experiencing a year of flooding. In light of the notable fluctuations seen throughout this particular period, it is crucial to undertake thorough examinations of the unusual years within this era, with particular emphasis on the regional atmospheric circulation patterns and their corresponding determinants. These findings play a crucial role in advancing the creation of reliable forecasting models and the establishment of impactful policies aimed at addressing the consequences of imminent climate change.

3.4. Events (1976–2000)

The study set out on a thorough investigation of the complex interdependencies that weave through climate modes, anomalies, and the complex dynamics underpinning the monsoon system throughout the period spanning 1976 to 2000. This thorough inquiry produced revelations that highlight the complex behaviour that defined this particular period.

From this historical perspective, the years 1982 and 1987 stand out as being crucial because of their unique dual-characteristic profiles that were shaped by the coexistence of El Niño and drought circumstances (

Table 5. The year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 1976–2000.

Climate Mode	Years
El Niño	1976, 1982, 1987, 1991, 1997
La Niña	1988, 1999, 2000
Drought	1979, 1982, 1985, 1986, 1987
Flood	1983, 1988, 1994
Negative IOD	1981, 1989, 1992, 1996, 1998
Positive IOD	1982, 1983, 1994, 1997

). Vivid depiction in Figure 8, a detailed spatial investigation of monsoon beginning anomalies, for instance, reveals an intriguing pattern for 1982 that is characterized by a dearth of extensive abnormalities that are primarily restricted to the peninsular region. The year 1987, in comparison, offered a quite different picture, one in which a clear positive anomaly dominated the area of western India and the Western Ghats. An important observation is the resonant spatial similarity between 1987 and the drought-affected year 1986, despite subtle differences in India’s far western regions. These findings expand our understanding of the complex web that the monsoonal dynamics weaves by highlighting the probable persistence of unique regional patterns in monsoon onset anomalies under particular climatic settings.

The year 1988, which is marked by the coexistence of La Niña and flood conditions, is revealed through careful analysis. With the significant exception of West Bengal, the adverse anomaly displayed this year indicates an early start to the monsoon. The La Niña events of 1999 and 2000, which defied expectations, have provided more intriguing findings. These years are different from the predicted pattern of early onset associated with La Niña occurrences; they manifested as negative anomalies in some parts, which is noticeable in the western and certain northern territories. This discrepancy casts doubt on well-held beliefs that underlie the behavior of the monsoon during La Niña conditions.

Furthermore, the peninsular region of India received particular attention for the perceptible impression of positive anomalies linked to late-onset conditions. When considered in the context of the positive IOD events that could be observed in 1982, 1983, and 1994, this phenomenon took on greater relevance. The years 1976 and 2000 reveal the complicated association connecting climate modes, anomalies, and dynamics unique to the monsoon system. The research findings portray the fluidity and dynamism inherent in these interactions.

3.5. Events (2001–2019)

The period from 2001 to 2019 provided crucial new information about the dynamics of monsoon onset anomalies and their intricate relationships with other climate modes. Particularly in the years 2002 and 2004, which are highlighted in

Table 6. The year with different climate modes. El Niño, La Niña, flood, and drought event years are based on the period 2001–2019.

Climate Mode	Years
El Niño	2002, 2006, 2009, 2015
La Niña	2007, 2010, 2011
Drought	2002, 2004, 2009, 2014, 2015
Flood	No Event
Negative IOD	2010, 2014, 2016
Positive IOD	2003, 2012, 2015

, years with drought events, unique patterns in monsoon onset anomalies emerged. Notably, throughout these years, a sizable area of India saw a discernible positive anomaly suggestive of a delayed monsoon commencement. Figure 9 uses spatial analysis to visually explain this remarkable event. The analysis identified noteworthy trends between 2001 and 2019 that deserve consideration. To begin with, there were no instances of flood events throughout this time. Furthermore, the distribution of high-magnitude, early-onset anomalies was not uniform across the entire country. Instead, a small number of these high-magnitude early-onset anomalies that affected the years 2003, 2006, 2012, 2014, 2015, and 2019 selectively materialized.

Understanding the complex relationships between climatic events and monsoonal dynamics is made possible by the classification of years according to several climate modes. Monsoon onset anomalies showed distinct patterns in El Niño years like 2002, 2006, 2009, and 2015. Similar to how the effects of drought years (2002, 2004, 2009, 2014, and 2015) and La Niña years (2007, 2010, and 2011) rippled across the complex web of monsoonal behaviour. Significantly, the IOD’s alternating positive and negative phases had an impact on the patterns of monsoon onset anomalies throughout this time. Positive IOD years (2003, 2012, and 2015) and negative IOD years (2010, 2014, and 2016) revealed distinct patterns in the range of monsoon onset anomalies, furthering our understanding of the complex climatic factors influencing monsoon dynamics.

3.6. Monotonic Trend in India

Considering India as a one-unit, monotonic trend using LRM represents an increasing trend (late onset). However, the trend result is not significant at the 90% level (Figure 10). Therefore, the ITA method was used to identify the trend for different sizes of monsoon onset day anomaly (Figure 11). ITA also shows no presence of a monotonic trend for India. Events lying between 0 and 5 in ITA show the decreasing trend of late monsoon onset, and events between 0 and −5 anomalous onset days are located in the positive trend region, indicating the increase. However, the detected trend for India as one unit is neither monotonic nor significant.

After obtaining such a result from the LRM and ITA, MKT was adopted to identify the spatial distribution of the significant monotonic trend of monsoon onset in India.

Figure 12a,b show the direction and magnitude of the trend. Punjab, Haryana, northern parts of Rajasthan, and parts of the Western Ghats region show a significant monotonic trend of early monsoon onset. Some patches in the Indo-Gangetic plain show the monotonic significantly decreasing trend. The positive trend of the significant grid points shows an increase of more than 1 day per decade.

Besides this, the area in Maharashtra located between the Satpura and Ajanta ranges and parts of the Maharashtra plateau shows a significant monotonic increasing trend. A grid point in the northeast, representing Manipur state, also shows the decreasing trend of the monsoon onset.

Thus, these highlighted regions with the monotonic trend will face a different onset date after a certain period. Consequently, mitigation measures to avoid huge losses are required.

4. Discussion

The findings examined in this study provide a thorough investigation of monsoon onset anomalies throughout multiple periods, illuminating complex relationships with diverse climate modes. Five key periods are covered by the study: 1901–1925, 1926–1950, 1951–75, 1976–2000, and 2001–2019. The inquiry explores the interactions between monsoon anomalies, El Niño, La Niña, drought, and flood throughout these times, offering complex insights into the dynamics of the monsoon system.

El Niño and drought coexisting between 1901 and 1925 became a major topic, leading to severe positive anomalies of monsoon onset. Numerous unfavourable anomalies are visible in years like 1911, 1913, and 1918, which were both El Niño and drought-stricken. The unexpected early monsoon arrivals in 1901, 1902, 1903, 1907, and 1912 cast doubt on preexisting theories and highlighted the need for a more thorough investigation of climatic interconnections.

Interesting relationships between climatic modes and monsoon onset anomalies are revealed for the years 1926 to 1950. Numerous positive anomalies can be seen in years like 1933, 1936, 1938, and 1941, while some years exhibit substantial negative abnormalities. Regional differences exist in the relationship between El Niño years and anomalies. Notably, a positive anomaly that occurred in 1933, a La Niña year, highlights the complex relationship between La Niña events and monsoon dynamics. The drought year of 1941 adds additional complexity.

The investigation examines how the monsoon and several climate modes interacted between 1951 and 1975. Within this range, particular years show specific traits. In India, distinct El Niño years are associated with varied anomaly magnitudes. We examined the connections between climate modes, anomalies, and monsoon dynamics throughout the years 1976 to 2000. Years like 1982 and 1987, when El Niño and drought coincided, display unique patterns. In 1988, a distinct early-onset pattern that is characterized by both a flood and a La Niña developed. Positive anomalies emerge when years like 1999 and 2000 break from typical La Niña tendencies. Understanding is deepened by the existence of negative anomalies during years of positive Indian Ocean dipole.

Monsoon onset anomalies and their interactions with climatic modes were studied from 2001 to 2019. Widespread positive anomalies in the drought years of 2002 and 2004 show a delayed beginning. Perspectives are offered by the lack of flood occurrences and the irregular distribution of high-magnitude, early-onset anomalies. El Niño, La Niña, and IOD phase patterns are visible when years are categorized based on climate modes.

Spatial variability is revealed by monotonic patterns in the monsoon’s arrival in India. The Mann–Kendall test detects a variety of patterns, although linear regression and individual trend analysis do not produce statistically significant results. Some areas, including sections of the Western Ghats, Punjab, and Haryana, show substantial early-onset trends. The Indo-Gangetic plain, on the other hand, exhibits declination patterns.

Due to the regional diversity, region-specific adaptive techniques are essential. Due to the absence of the monotonic trend of monsoon onset for India, the spatial distribution of the monsoon onset trend was completed because it can show the significant monotonic trend for each grid present in the study area. The grid points showing the significant monotonic trend in Figure 12 may have an abnormality in many important prevailing conditions due to the early/late onset in the future. Therefore, the policy and planning-related institutions should consider these regions as vulnerable districts, and a highly refined framework must be prepared to mitigate the impacts in the long term. While investigating the intricacies of the monsoon onset, Saini et al. (2022) found its relation with the length of the monsoon season and the amount of rainfall [1]. Therefore, in the long term, changing patterns and timing of the monsoon onset can influence the regime of agriculture, disease proliferation, river regime, etc. [15,18,20,21].

The study offers profound insights into the intricate connections between climate modes, anomalies, and monsoon dynamics over time. The behaviour of the monsoon system and its ramifications in the face of shifting climatic circumstances are better understood as a result of these results.

5. Conclusions

The purpose of this study was to analyze the spatial patterns of monsoon onset in India and to determine the onset of the monsoon at a finer spatial scale using the piecewise linear regression model (PLRM). The study clarified the complicated nature of monsoon onset and its variation over various geographic and temporal scales. There were no discernible trends found in the examination into monotonic trends in monsoon onset for all of India. However, when examining particular regions, it was found that sections of Maharashtra and some parts of northern India showed substantial monotonic tendencies toward the start of the monsoon. On the other hand, Manipur and several areas of the Indo-Gangetic plain displayed notable monotonic tendencies toward delayed onset. These results imply regional differences in monsoon patterns, necessitating regional climate adaptation and mitigation measures. The findings have important repercussions for India’s regional development policies and planning efforts for agriculture that is climate resilient. To effectively address changing monsoon patterns and their potential effects on water resources, agriculture, and ecosystems, it is helpful to understand the geographical trends in monsoon start. It is critical to understand that monsoon start is a complicated phenomenon affected by a variety of elements and that its forecast is still difficult. This study emphasizes the necessity for monitoring and research in this field by providing insightful information on the complex dynamics of monsoon onset. The role of the present stochastic regional disturbance cannot be ignored. In contrast, using the PLRM for each year by considering the strip of time from the 60th to the 225th day can yield important results. It is suggested to use this method with the regional climate models or downscaled global climate models to project the onset of monsoon in different parts of India.

In conclusion, the study highlights the importance of examining monsoon onset at a finer spatial scale and offers useful data for stakeholders, policymakers, and researchers in developing mitigation strategies for the effects of changing monsoon patterns in various parts of India. Our understanding of monsoon dynamics and its consequences for climate change and sustainable development can be improved by further research based on these findings.