Twenty locations have been selected for the analysis of TCO and are listed in Table 1.

For the comparison of total column variation, the selected locations are grouped into four categories, namely south Indian region, middle Indian region, northeast Indian region, and north Indian region. Nagarcoil, Pondicherry, Kasargod, Panaji, Bellari, Vishakapattanam, and Ramagundam are grouped under the South Indian region; Ahmednagar, Raipur Rajkot, Kharagpur, Bhopal, and Patna are grouped in the middle Indian region. Aizawl, Dibrugarh, Gangtok are clustered in the North East region, and the remaining locations Lucknow, Jaipur, Chandigarh, and Anantnag are treated as North Indian regions. Selected locations are shown in Figure 1.

Total column ozone (TCO) overpass data were obtained from two different instruments: Total Ozone Mapping Spectrometer (TOMS) and Ozone Monitoring Instrument (OMI) on Earth Probe and AURA satellites respectively. TOMS measures backscattered UV solar radiances from the Earth’s surface and from atop of clouds. OMI measures the complete spectrum from 270 nm to 500 nm at an average spectral resolution of 0.5 nm. The sensor provides global daily coverage of columnar concentration of ozone, NO₂, and formaldehyde. Online data is available from 2004 to the present. The spatial resolutions of the datasets are 1˚ latitude × 1.25˚ longitude for TOMS and 1˚ latitude × 1˚ longitude for OMI [55] [56]. The tropospheric column ozone data determined by subtracting measurements of microwave limb sounder (MLS) stratospheric column ozone from ozone monitoring instrument (OMI) measured total column ozone after adjusting the inter-calibration differences of the two instruments using the convective cloud differential (CCD) method used during the period 2005-2018 [57].

For computing the long term TCO variations, the monthly average total column O₃ time series data were derived from the daily TOMS datasets A linear regression statistical model represented by X = a ∗ t + c is applied to monthly mean TCO data for each selected location. Here, “X” denotes the monthly average TCO, “a” denotes the coefficient of the slope in the linear regression line, “t” represents the time period in months and “c” represents the intercept. More details about the statistical model are available here [58] [59] [60]. The variation trend is obtained by using the equation,

Trend = b × 12 × 10 Avg TCO × 100

Here “b” is the slope coefficient, Avg_TCO is the average value of TCO for the selected period.

Figure 2 shows the monthly average TCO over the study locations, and it shows a strong seasonal variation. All the locations show the maximum concentration in summer months and the minimum in winter months.

It has been found that in the south and middle Indian region, maximum monthly average TCO values ranging up to 290 DU during the months of April, May, and June. These high values in April-June were mostly linked to more sunlight hours, high temperatures, and low humidity. In the northeastern region, monthly averaged maximum TCO values were found in April and May, and in the northern Indian regions in February and March. Decreased TCO

concentrations were observed throughout the regions from June to November due to increased monsoon rainfall, cloud cover, large wind speed, and lower atmospheric temperatures. Further, the concentration of TCO in the whole area was found to be very low during the months of November-January, and a rising trend was observed from December to March.

Due to the favorable conditions, such as more sunlight hours, high temperatures, and low humidity, a high concentration of TCO has been observed in all the locations during the summer seasons. In the stratosphere, solar UV radiation plays an important role in the variation of O₃ concentration. Brasseur, 1993 [61] reported that the changes in ultraviolet spectral irradiance can directly affect the formation of the stratospheric O₃. A model study by Haigh, 1994 [62] revealed that a 1% increase in UV radiation at the maximum of a solar cycle will generate a 2% increase in O₃ concentrations in the stratosphere. Atmospheric O₃ is positively correlated with temperature and sunshine hours in a day, and it is negatively correlated with precipitation, humidity, wind speed, and cloud cover [63]. TCO decline was found to be statistically significant for all the months.

The statistical analysis was carried out into two phases, from 1979 to 1994, and from 2005 to 2018, and the detail is shown in Table 2. It was found that

Anantnag (−4.63%), Dibrugarh (−4.09%), and Chandigarh (−4.06%) show a significant decreasing trend; whereas Gangtok (−3.88%), Lucknow (−3.50%) Aizawl (−2.68%), and Jaipur (−2.46%) shows a moderate decreasing trend.

A higher concentration of TCO was found at Anantnag (349 DU) whereas a lower concentration was found at Kasargod (227 DU). One important feature found in the study is that there is a marked decrease in TCO/year. Every year, on average, there was a decrease in TCO observed, which was maximum at Anantnag (−0.050%) and minimum at Vishakapattanam (−0.017%). Gangtok (−0.033), Dibrugarh (−0.034), Aizawl (−0.029), and Chandigarh (−0.028) are also experiencing a substantial decreasing trend/year. The variations in concentrations of TCO over selected locations for the periods (1979-1994 and 2005-2018) are shown in Figure 3. The decadal analysis shows a substantial declining trend of TCO over the north and northeast Indian region. The declining trend can be due to numerous factors. There is considerable aerosol loading in the northern and northeastern Indian regions, and these aerosols react with atmospheric gases and play an important role in reducing the tendency of TCO in these regions [64] [65].

Figure 4 shows the retrieved annual average variation of TCO over selected study areas. It is observed that the concentrations of TCO over the study locations are varying from 225 DU to 350 DU, and TCO gradually increases from an average concentration near the equator to a maximum at high latitude. A linear

regression trend has also been included in the figure, and the regression slope shows different for different sites.

A strong decline trend can be clearly seen in high latitude regions with TCO values ranging from 230DU to 348 DU. Further, the TCO concentrations in late 2018’s were significantly lower than the corresponding levels recorded in 1979. The figure also shows the large inter annual variability of TCO concentrations over the study period. Factors such as the dynamical structure of the atmosphere, changes in air circulation from the troposphere to the stratosphere, changes in the anthropogenic O₃ depletion substances, solar flux, variations in tropopause height, etc. were strongly controlled the annual variations of TCO [66] [67] [68].

In order to study the latitudinal variation of TCO, the average value of TCO was calculated from the monthly mean TCO for all the selected locations. Latitudinal variation of TCO for the years 1979, 1994, 2005, and 2018 is shown in Figure 5. From the figure, it is observed that latitudinal variation of TCO shows an exponential increase from Nagarcoil to Anantnag. In 2018, Anantnag (33.45˚N) recorded 32.52 DU more TCO than Nagarcoil (8.9˚N). There has been a substantial reduction of TCO (21.58 DU in Anantnag and 13.4 DU in Nagarcoil) observed from 1979 to 2018 from low to higher latitude regions. In addition to these, we found that the average TCO was increased by 2.52 DU per degree of latitude for the period of 1979-2018.

A study by Bhattacharya and Bhoumick, 2012 [69] revealed that the TCO level has been increasing by 15DU as go from Trivandrum to New Delhi. Geographical and meteorological feature of the Indian sub-continent is one of the main reasons for the spatial variability of TCO concentration observed over different locations. The western and southern part of India is surrounded by the Arabian Sea and the Indian Ocean. The northwest region of India is generally dry. The northern region is bounded by Great Himalaya and Tibetan Plateau. The north-eastern part is categorized with dense vegetation, mountains, and Indo-Gangetic Plains. This may be the reason responsible for the observed temporal and spatial variation of TCO.

Monthly mean airflow patterns at 100 mbar over the Indian subcontinent for the year 2018 were retrieved from The National Center for Environmental Prediction (NCEP), reanalysis for studying the influence of wind pattern on the TCO over the Indian subcontinent is shown in Figure 6.

At 100 m bar level, the low latitude wind was much stronger than that at higher latitude. The TCO profile shows a distinct gradual variation from south to north direction over the subcontinent for all the seasons. During the first phase of the summer season (March, April), winds were stronger and the circulation was south westerly-westerly (from ocean to land). The south westerly-westerly wind gets weakened by May and the north-easterly wind starts in June. The wind direction remains northeasterly until October when the airflow was mostly from the continent. In March and April, the concentration of TCO shows a prominent change over all the study areas. It was noted that the concentration of TCO observed to be maximum in the northern region where the strong easterly wind profile existing. As monsoon approaches, the concentration of TCO was found to decrease throughout the north Indian region during which wind direction was from east-west. In the month of June, the relatively low easterly wind was observed in the south-middle Indian region, and the westerly wind was observed in the north Indian region. During July and August, the entire wind pattern in the sub-continent has changed into the east-west direction.

This profile can cause less concentration of TCO in the northern latitudes due to confinement of locally generated TCO [70].

Figure 7 shows the monthly average concentrations of tropospheric O₃ over selected locations of the Indian subcontinent by employing satellite data. From the figure, it is evident that tropospheric O₃ shows significant monthly variations. The maximum concentration of tropospheric O₃ was found in the summer months due to the enhanced photochemistry in the lower atmosphere. During June, July, and August, there was asubstantial reduction of tropospheric O₃

observed due to the active south-west monsoon over these locations. In the winter months, the lower boundary layer height and the dry weather offer the elevation of concentration of tropospheric O₃. A long-term satellite-based study conducted by Kalita and Bhuyan [71] on the Indian subcontinent observed the maximum concentrations of tropospheric O₃ distribution in the northeastern (NE) and the Indo Gangetic Plains (IGP). A study on the variability of tropospheric O₃ over the coastline of the Indian subcontinent by Kulkarni et al. [72], revealed that tropospheric O₃ has a strong seasonal cycle over the Indian coastline, with a large variation over the upper east coast followed by the upper western coast. Table 3 shows the annual average variation of tropospheric O₃ observed in 2005 and 2018 and the percentage of increase. From, the table it is clear that tropospheric O₃ shows an increasing trend over all the locations with the maximum trend observed at Gangtok followed by Aizawl, Raipur, Anantnag, Vishakapattanam, and Ramagundam.

The long-term trends (2005-2018) in tropospheric O₃ over the selected locations were carried out by linear regression analysis and are shown in Figure 8. From, the figure it is clear that the concentrations of tropospheric O₃ show an increasing trend over all the locations. This increase in the trend of tropospheric O₃ was mainly due to an increased anthropogenic emission of gases in and around the locations selected. Different studies show that the high concentration of NO₂ in the atmosphere is the main reason for the increase in O₃ concentration

observed over different parts of the Indian region. Sheel et al. [73] point out that NO₂ emissions in India have been increasing over the past few years due to rapid economic and industrial growth. The enhancement of NO₂ over these regions is mainly due to the industries including power plants, vehicular exhaust, mining activities, etc. [74].

Figure 9(a) and Figure 9(b) show the monthly average annual variation of the spatial distribution of tropospheric O₃ over the Indian sub-continent for the year 2005 and 2018. The enhanced O₃ was found along the highly populated north and north-western part of India. It is also found that the tropospheric O₃ concentration in all parts of the Indian subcontinent is much higher in 2018 than in 2005. The observed change in concentration of tropospheric O₃ from 2005 to 2018 is shown in Figure 9(c). The elevated concentration of tropospheric O₃ in 2018 was mainly due to the regional enhancement of precursors at low altitudes in the presence of sunlight. Nitrogen dioxide (NO₂) plays an important role in the production of tropospheric O₃.

Studies based on bottom-up approaches indicate that the NO₂ emissions have been increasing and will continue to increase over India in the coming years. The rate of change of tropospheric O₃ simulated by regression analysis is plotted in Figure 9(d). The enhanced rate of change of O₃ was observed in the IndioGangatic region and middle Indian regions. Biofuels used for cooking in rural areas, the presence of thermal power plants, steel, sugar, and other small industries

leads to strong emissions of various pollutants in the sub-continent [75] [76]. "The decadal changes in tropospheric O₃ have also been calculated over the Indian region during periods 2005 to 2018 and it was observed that during the period 2005 the average tropospheric O₃ was 30.97 DU, whereas during 2018 it was a little higher at about 33.50 DU with an increasing value of 2.53. Though the difference is not too large, it indicates an increasing tendency. Hence, it can be said that the increasing concentration of GHGs during 2005-2018 was responsible for the higher level of the tropospheric O₃ during this period over the region.