1 Introduction

Air pollution caused by industrialization is a primary concern in today’s society. It affects human health at a local level (Balakrishnan et al. 2019; World health organization 2016) and on climate at the global level (Ramanathan and Feng, 2009) amid the economic development. Though major push is given to renewable energy sources of late, it is challenging to meet the increasing energy demand and thus resorts to fossil fuel-based energy sources. There should be concentrated effort to reduce the effects caused by emissions from polluting industries through a sustainable and scientific way. There is a strong need to look for ventilation corridors or sites with high pollution dispersion potential towards mitigating these effects.

The atmospheric boundary layer (ABL) influences the vertical mixing of air pollutants at the earth’s surface, and the planetary boundary layer height (PBLH) is an essential parameter in the air pollution dispersion (Stull 1988). PBLH can be estimated using radiosonde and Lidar, which is location-specific so limited by spatial resolution. However, satellite-based retrievals such as radio occultation (GPS RO) and atmospheric sounders (AIRS, CrIs, INSAT 3D etc.) offer both spatial and temporal coverage (Seidel et al., 2010; Smith et al. 2009; Xie et al., 2012; Hareef baba shaeb et al., 2021).

Air pollutant’s concentration gets diluted away from the source in the atmosphere due to mean wind and convection, which depend on geographic location and land surface condition. ABL height and wind speed can be used in combination to study air pollution potential (Holzworth 1967). Ventilation coefficient (VC), a product of ABL height and average wind speed within the ABL, is a good indicator of air pollution dispersion potential (Goyal et al. 2006; Lu et al., 2012). Assimilative capacity, i.e., the environment’s ability to disperse the pollutants away, avoiding user exposure, can be determined using the ventilation coefficient (Manju et al., 2002). Thus studies on ventilation coefficient, i.e., ventilation corridors, are significant to assess the suitability of a particular location/zone for establishing an industry that can have a local impact as an air pollution source. Moreover, an air pollution potential index (APPI) comes as a valuable tool in governance in assessing a particular site suitable for the polluting industry establishment and planning-related mitigation measures. This tool will also help monitor potential threats caused by already existing industries in case of an accident and subsequent air dispersion modeling.

Several researchers studied pollution potential based on the ventilation coefficient at several locations in India and the world (Gassmann and Mazzeo, 2000; Goyal and Rama 2002; Krishnan and Kunhikrishnan, 2004; Lu et al., 2012; Iyer and Raj, 2013; Abiye et al., 2016). These studies mostly use regularly available morning (00 UTC) and evening (12:00 UTC) radiosonde data at sparse locations to arrive at the ventilation coefficient. However, spatial variation of ventilation coefficient and its implications to ventilation corridors and pollution potential has not been attempted. Further, the highest ventilation coefficient, i.e., low pollution potential (or high assimilative capacity), occurs at afternoon time (08:30 UTC) (Manju et al., 2002; Lu et al., 2012). Satellite-based methods have the advantage of capturing spatial variation with optimum temporal resolution compared to point-based radiosonde measurements. Towards this, we use afternoon time SNPP-Crls soundings (Han et al. 2013), representing a fully developed mixed layer (Anurose et al., 2018; Hareef baba shaeb, 2019) to arrive at the VC.

This study uses 5 years (2015–2019) of PBLH data estimated from SNPP-CrIS and mean wind speed obtained from ERA-5 reanalysis data to generate spatial maps of VC/APPI over India. We also analyze ventilation coefficient and APPI climatology over selected cities of India. The objectives of this study are as follows:

  1. 1.

    To map and study spatial variation of VC/APPI over India by analyzing monthly, seasonal, and annual averages of VC and identify the areas with high VC/low pollution potential

  2. 2.

    To understand the monthly and seasonal variation of VC/pollution potential at different cities located near coastal, desert, Indo-Gangetic Plain (IGP), and inland regions in India

With 1.38 billion populations, India is at the risk of high air pollution due to increasing anthropogenic activities, i.e., power production, industries, and vehicular emissions. These potential pollution maps and their climatology will assist in designing policies, planning, and interventions towards sustainable development.

2 Study Area, Data, and Methods

2.1 Study Area

In India, a variety of climatic regions are present, namely tropical (south India), temperate and alpine (Himalayan North), and continental (north Indian landmass), while coastal regions of the country experience unvarying warmth and frequent rains (Attri and Tyagi, 2010). Census data (www.censusindia.gov.in/2011) reveals that 67% households use firewood/crop residue, cow dung cake/coal, etc. (rural, 87%; urban, 26%), while 29% households use LPG/PNG/electricity/biogas (rural, 12%; urban, 66%), and the remaining 3% households use kerosene (rural, 1%; urban, 8%). Population density (no. of persons per sq. km) is an important factor in understanding the risk of air pollution and anthropogenic contribution. Population density tends to increase in the industrial areas and cities as the movement of human populations is more towards these areas, creating air pollution hotspots. Figure 1a shows population density mapped using 2011 census data (http://populationcommission.nic.in). Among the union territories (UT), Delhi has the maximum population density (11,297 persons/km2), and among the states, Bihar is the most densely populated state with 1102 persons/km2. The IGP region is continuing to be the most populated in the country.

Figure 1
figure 1

a India map showing population density and CEPI scores for industrial cluster/area; b spatial distribution of Cities with elevation data for India (https://lta.cr.usgs.gov/GTOPO30)

Full size image

Comprehensive environmental pollution index (CEPI) for different industrial clusters/areas is assigned by Central Pollution Control Board (CPCB), India. A sub-index score of more than 60 shows a critical level of air pollution, whereas a score in between 50 and 60 shows a severe level of pollution. A score below 50 is normal. Out of 88 industrial areas, only 20 fall in normal category. Figure 1a shows air pollution industries locations and levels measured by CPCB (www.cpcb.nic.in/divisionsofheadoffice/ess/NewItem_152_Final-Book_2.pdf). Figure 1b shows the spatial distribution of cities considered for the monthly and seasonal analysis of a ventilation coefficient.

2.2 Data and Methodology

2.2.1 Estimated PBLH Data from CrIS on Soumi NPP Satellite

The Cross-track Infrared Sounder (CrIS) onboard the Suomi National Polar-Orbiting Operational Environmental Satellite System Preparatory Project (SNPP) computes the vertical profiles of temperature and relative humidity with an excellent vertical resolution (Smith Sr et al., 2009; Han et al., 2013). The retrieval biases for temperature are reported to be within 0.5 K from the surface up to the lower stratosphere. For water vapor, the biases are within 20% up to 400 hPa (Sun et al., 2017), which indicates high accuracy of retrieved profiles (Zhou et al., 2019). Estimation of the planetary boundary layer height (PBLH) was done by the National Remote Sensing Centre (NRSC), Indian Space Research Organisation (ISRO), Hyderabad, India, using these vertical profiles from SNPP CrIS (Prijith et al., 2016). The daily, 7 days, and monthly estimated PBLH data are available at the NRSC, ISRO website (https://bhuvan-app3.nrsc.gov.in/data/download/index.php) with a horizontal grid resolution of 0.25° × 0.25°. Prijith et al. (2016) used a combination of the vertical gradient method (Hennemuth and Lammert, 2006; Basha and Ratnam, 2009; Seidel et al., 2010; Wang and Wang, 2014, Hareef baba shaeb, 2019) of meteorological variables to produce the spatiotemporal PBLH data from CrIS vertical profiles over the Indian landmass. In the present study, the daily PBLH data have been used to calculate the atmospheric ventilation coefficient for 2015 to 2019.

2.2.2 Determination of Average Wind Speed Within the Mixing Layer from ERA5 Reanalysis Data

The fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) atmospheric reanalysis of the global climate (ERA5) provides many atmospheric variables produced at the surface and pressure levels. Besides, the ERA5 data are globally available in the Climate Data Store (https://cds.climate.copernicus.eu/cdsapp#!/home) at 0.25° × 0.25° grids of latitude–longitude with different time scales. ERA-5 winds were validated to be offering the best estimate of mean wind speeds compared to other reanalysis products (Ramon et al., 2019).

The hourly surface pressure, air temperature, mixing layer height, and orography data were obtained from ERA5 single level reanalysis data from 2015 to 2019 (Copernicus Climate Change Service, 2017). Firstly, the pressure at the mixing layer height was calculated using the hypsometric equation (Eq. 1) from single level ERA5 reanalysis data (Wallace and Hobbs, 2006).

$${p}_{{z}_{i}}={p}_{s}{e}^{-\frac{g\left({z}_{i}+{z}_{s}\right)}{{R}_{d}T}}$$
(1)

where \({p}_{{z}_{i}}\) is the pressure value at the mixing layer height (hPa), \({p}_{s}\) is the surface pressure (hPa), \(g\) is the gravitational acceleration constant (\(g=\) 9.80665 m·s−2), \({z}_{i}\) is the mixing layer height above ground level (m), \({z}_{s}\) is the orography or elevation of the grid point above the mean sea level (m), \({R}_{d}\) is the gas constant for dry air (\({R}_{d}=\) 287.05 J·kg−1·K−1), and \(T\) is the absolute air temperature (K).

Wind speeds below the obtained pressure level were averaged at each latitude–longitude grid point of 0.25° for a given time from the ERA5 pressure levels data over the Indian region. The average wind speed within the mixing layer is an essential variable in calculating the atmospheric ventilation coefficient (Saha et al., 2019).

2.2.3 Calculation of Ventilation Coefficient

The ventilation coefficient (VC) is the product of mixing layer height and average wind speed within the mixing layer (Saha et al., 2019). The VC is given by Eq. 2.

$$VC={z}_{i}\bullet U$$
(2)

where \(VC\) is atmospheric ventilation coefficient (m2·s−1), \({z}_{i}\) is the mixing layer height (m), and \(U\) is the average wind speed within the mixing layer (m·s−1).

This has been calculated using Eq. 2 considering the daily PBLH derived using Crls soundings and ERA-5 wind speed up to mixed layer height. Later monthly averages and seasonal averages for all the years from 2015–2019 are calculated. ArcGIS software is used for the interpolation (kriging) and subsequent generation of VC/APPI maps. To study the seasonal variation of VC, four distinct seasons are considered, i.e., pre-monsoon (PMS) (March, April, May), summer monsoon season (SMS) (June, July, August, September), post-monsoon (PoMS) (October, November), and winter (December, January, February).

3 Results and Discussion

3.1 Spatial Variation of VC and APPI Zones

Figures 2, 3, 4, and 5 show the spatial variation of monthly mean ventilation coefficient, PBLH, and mean wind speed obtained using the daily data from 5 years (2015–2019). The spatial variation changes through different months, indicating the monthly variation of VC. US National Weather Service (Gross 1970) and Atmospheric Environment Service Canada uses VC criteria of high (< 6000 m2/s), medium (6000–12,000 m2/s), and low pollution potential (> 12,000 m2/s). However, based on the variability of VC over the Indian region, the ventilation coefficient has been categorized into six different pollution potential zones or APPI as shown in Table 1.

Fig. 2
figure 2

Monthly mean maps of ventilation coefficient (a, d, g), PBLH (b, e, h), and wind speed (c, f, i) generated using 5-year (2015–2019) daily data for the months of December, January, and February (winter)

Full size image
Fig. 3
figure 3

Monthly mean maps of ventilation coefficient (a, d, g), PBLH (b, e, h), and wind speed (c, f, i) generated using 5-year (2015–2019) daily data for the months of March, April, and May (pre-monsoon)

Full size image
Fig. 4
figure 4

Monthly mean maps of ventilation coefficient (a, d, g, j), PBLH (b, e, h, k), and wind speed (c, f, i, l) generated using 5-year (2015–2019) daily data for the months of June, July, August, and September (summer monsoon season)

Full size image
Fig. 5
figure 5

Monthly mean maps of ventilation coefficient (a, d), PBLH (b, e), and wind speed (c, f) generated using 5-year (2015–2019) daily data for the months of October and November (post-monsoon)

Full size image
Table 1 List of air pollution potential index (APPI) categories
Full size table

During the winter months, i.e., December (Fig. 2a, b, c), the east coast of southern India (especially Andhra Pradesh and Tamilnadu) registers maximum range in PBLH (1600–2000 m). PBLH values 800–1600 m dominate the south Indian region and parts of western and eastern India. Central and northwest Indian parts are dominated by PBLH in the range 400–800 m and the northern most region’s minimum PBLH range (0–400 m). Wind speed is found to be highest (8–10 m/s) in the northeastern tip of J&K and coast of Tamilnadu (TN) (6–8 m/s). Parts of Andhra Pradesh (AP), TN, Gujarat (GJ), and Himachal Pradesh (HP) register wind speed of 4–6 m/s, and the rest of India reports wind speeds of 2–4 m/s with a minimum (0–2 m/s) occurring in the northeast. This different spatial variation in PBLH and WS gives rise to different spatial variations in VC. East coast of TN registers high (> 10,000 m2/s) VC resulting in APPI value of 6, i.e., very low pollution potential, most probable ventilation corridor, followed by the east coast of AP with APPI value of 4. Parts of Maharashtra (MH), Karnataka (KA), West Bengal (WB), and TN register APPI value of 3, and the rest of India registers APPI of 1–2, indicating very high to high pollution potential. Interestingly, though the IGP region registers moderate PBLH (800–1600 m) during this month, low wind speeds (2–4 m/s) result in high pollution potential, leading to stagnation of pollutants. We also observe in Fig. 2a that high polluting industries (with a CEPI score of 50–80) are located in this region and also in the Delhi NCR region in combination with a heavy population (population density 1100 per sq. km) making it the polluted hotspot region in Asia (Sen et al., 2017).

During the January (Fig. 2d, e, f) and February (Fig. 2g, h, i) months in the winter season, there is not much change in the region of highest PBLH. However, in the other areas, PBLH increases except in the northernmost areas, and at the same time, wind speed increases considerably in the IGP region and northwestern parts. This increase leads to VC improvement and APPI in western states (Gujarat, Rajasthan) and parts of central India. One can understand these improvements in VC and APPI due to increasing solar radiation reaching the ground in February compared to December, resulting in more convection and thereby thicker PBLH.

Similar observations of higher mixing heights in northeastern and southern parts of the country and the general increase of the value of north to south were reported by Vittal Murty et al. (1980) using radiosonde ascents data available at 0000 GMT (0530 IST) and at 1200 GMT (1730 IST) at 11 locations using data from 1959–1963. VC values in the range 4000 to 6000 m2/s and uniform VC for central India were reported, which closely matches with our observation; moreover, an excellent spatial variation made possible by satellite-based observations is captured in the present study.

During PMS month, i.e., March (Fig. 3a, b, c), the highest PBLH (1600–2000 m) was observed in the east coast of Tamilnadu and considerably higher PBLH (1200–1600) in AP, WB, Bihar, Assam Gujarat (GJ), Rajasthan (RS), the eastern coast of Odisha, east Maharashtra, and central Chhattisgarh (CG). South India is dominated by wind speeds in the range of 2–4 m/s, while north India is dominated by higher wind speeds (4–6 m/s). The highest VC values of 6000–8000 m2/s (thus APPI of 4) were observed in eastern parts of TN and IGP and western parts of GJ and RS. Northernmost parts, i.e., J&K, HP, Uttaranchal (UL), and Arunachal Pradesh (ARP), register a minimum APPI of 1, indicating very high air pollution potential. In the rest of India, APPI values range from 2 to 3, indicating high to medium air pollution potential in this month was observed. In the April month (Fig. 3d, e, f), uniform PBLH (800–1200 m) is widespread except higher PBLH (1200–1600 m) observed in eastern TN, AP, Assam, Bihar, and GJ and lower values (0–800 m) in the northernmost parts. There is not much change in WS observed compared to March. From April to May (Fig. 3g, h, i), one can observe that high APPI is found in Gujarat, parts of Bihar, and Jharkhand making them suitable ventilation corridors. The general decrease of mixing heights and ventilation coefficients from north to south during April is reported by Vittal Murty et al. (1980). They attributed this to maximum surface heat and its pattern this month.

During SMS, months, i.e., June (Fig. 4a, b, c), July (Fig. 4d, e, f), and August (Fig. 4g, h, i), the southwest monsoon winds (8–10 m/s) start flowing and considerable drop in temperatures due to rainfall gives rise to different patterns in spatial distribution of PBLH and wind speed. This results in high VC value (8000 to > 10,000 m2/s) regions and high to very high APPI values 5 and 6, respectively, in western RJ, Gujarat, west MH, KA, AP, and the southern tip of Kerala and TN. In September month, the VC pattern changes owing to monsoon retreat and decreased wind speeds (Fig. 4j, k, l), and APPI values in the range 4–5 were observed in GJ, RJ, IGP, and in few scattered places. As many places are found with APPI of 5–6 in this season, the air-polluting industries (Fig. 1a) located in these places can maximize their production, causing less pollution concentration at the local level, and there is an advantage of rain washout of the pollutants also. So the policies need to be framed in such a way to maximize production in the month/season when the region registers an APPI of 5–6, leading to sustainable growth. Vittal Murty et al. (1980) observed low mixing heights in the central parts of the country compared to the southeastern parts during the August month. Similarly, they reported that the southern tip of India recorded higher values than northern parts, which they attributed to the more significant influence of monsoonal winds in the south.

During October in the post-monsoon season (Fig. 5a, b, c), wind speeds are observed to be less (2–4 m/s) in the majority of the Indian region, giving rise to be not better than APPI value of 3 despite the good amount of areas showing up PBLH of 1200–1600 m. However, during November (Fig. 5d, e, f), wind speeds in southern India, particularly TN, AP, and KA, increase (4–6 m/s), possibly caused by increasing storm activity and associated cyclones touching the landmass during this month. In association with good PBLH (1200–2000 m), this increase in wind speeds results in APPI value four and better (5–6), indicating the potential pollutant dispersion in the east coast of AP and TN.

BLH values of 800–1700 m and VC values of 3000 to 6000 m2/s were reported by Vittal Murty et al. (1980) for October. They observed that the extreme southern tip of India registered higher values and northeast parts, the lowest. This observation closely matches our results.

3.2 Seasonal Variation of VC and APPI

Figure 6a shows that during PMS, Western GJ, SW of Rajasthan, and the junction region of UP, Bihar, CH, and JK show APPI of 4–5(below medium to low pollution potential). These values make them suitable sites for industries to reduce the air pollution effects, or already existing air-polluting industries in these regions can maximize the production in this season. The main problem is the concentration of the pollutants, not the pollutants themselves, for which APPI is the solution. Feasibility is left to policymakers, but we provide the solution. During SMS (Fig. 6b), most of western India and New Delhi region show good APPI value in the range 4–6 (below medium to very low pollution potential), making these regions possible sites for industrialization or maximum production. Figure 1a shows the good number of industries with high CEPI scores falling in these regions, so the key is to maximize the production in this season and put less air pollution burden. During the post-monsoon (Fig. 6c), the east coast of TN possesses an APPI value of 4 (below medium pollution potential), making it a good site for industrial establishment. During winter (Fig. 6d), the east coast of Tamilnadu and AP shows APPI values in the range 6–4 (very low to below medium), making them suitable sites for air-polluting industries to reduce the effects of pollution.

Fig. 6
figure 6

Seasonal mean maps of ventilation coefficient for a PMS, b SMS, c PoMS, and d winter generated using 5-year (2015–2019) daily data

Full size image

3.3 Annual Changes in VC and APPI

Figure 7a–e show annual averages from 2015 to 2019 and a 5-year average of VC and APPI. East coastal Tamilnadu and western Gujarat consistently show APPI value of 4 and above, implying low pollution potential, i.e., ventilation corridors favorable for polluting industries. These sites offer the highest dispersion of pollutants. Five-year annual average maps suggest that Tamilnadu, Andhra Pradesh, southern Kerala, Karnataka, western Maharashtra, western Telangana, Gujarat, western Rajasthan, UP, and Bihar possess low pollution potential, i.e., effectively disperse the pollutants compared to the rest of the states/regions. However, look at annual VC/APPI distribution (Fig. 7f) and comparison with Fig. 1a suggesting that many high CEPI score industries are placed in very high pollution potential zones. Policy changes and future planning should consider moving the highly polluting industries to APPI zones of 4 or greater or maximize production in different seasons, making less air pollution.

Fig. 7
figure 7

Annual mean maps of ventilation coefficient for the year a 2015, b 2016, c 2017, d 2018, e 2019, and f 2015–2019 generated using daily data in the respective year

Full size image

3.4 VC and APPI Analysis over Multiple Cities in India

The analysis of ventilation coefficients over cities is essential as the population density is more due to migration of people from other places to these cities for want of jobs, livelihood, etc. This high population leads to increased pollution from vehicular traffic in addition to polluting industries. Here, major metropolitan cities (population > 4 million) and few other cities of geographical importance (desert, IGP, etc.) are considered for the VC and APPI analysis. Policymakers can use this information to plan effective strategies to reduce the impact of pollution on a large population. Moreover, VC is very useful in atmospheric dispersion modeling for metropolitan areas (Holzworth, 1967).

3.4.1 Coastal Region

Figure 8 shows 5-year monthly and seasonal mean variation of ventilation coefficient over three metropolitan cities located on the west coast (Mumbai) and east coast (Chennai and Kolkata). Maximum VC (10,466 ± 7234 m2/s) in Mumbai is observed during the monsoon season (maximum in July—15,785 ± 5353 m2/s) and minimum VC in PoMS (4390 ± 2870 m2/s) followed by winter (minimum in January—3761 ± 2128 m2/s). This implies that Mumbai offers very low pollution potential during the SMS (with APPI—6) and medium pollution potential (APPI—3) during the PoMS. In contrast to the west coast, in the east coast, maximum in VC (8626 ± 4109 m2/s for Chennai and 5512 ± 2962 m2/s for Kolkata) occurs in the winter season (maximum in December for Chennai—10,261 ± 5088 m2/s and in January for Kolkata—6156 ± 2920 m2/s), and minimum VC (2789 ± 3343 m2/s for Chennai and 3545 ± 4321 m2/s for Kolkata) occurs during SMS (minimum in June for Chennai—1582 ± 1942 m2/s and in July for Kolkata—2148 ± 541 m2/s). It is observed from Fig. 2 that during winter, these stations have high PBLH (1200–2000 m), which in association with the wind speed (4–6 m/s), result in high VC values. Low and medium pollution potential (APPI—5 and 3) respectively was observed for Chennai and Kolkata during the winter season, and high pollution potential (APPI—2) occurs for both cities during the SMS. Annual mean VC values for Mumbai, Chennai, and Kolkata are observed to be 5739 ± 3467 m2/s, 5639 ± 4039 m2/s, and 4568 ± 3010 m2/s, respectively.

Fig. 8
figure 8

Monthly and seasonal variation of VC for Mumbai (a, d), Kolkata (b, e), and Chennai (c, f) generated using 5-year (2015–2019) daily data

Full size image

Vittal Murthy et al. (1980) observed the afternoon (12:00 UTC) ventilation coefficients for Mumbai, maximum VC was observed during the SMS (i.e., August), and the minimum was observed during winter (i.e., January). They attributed maximum observed values during this season to the strong winds of the SMS. However, the minimum for Chennai and Kolkata reported occurring during PoMS (i.e., October) at this time (12:00 UTC). Iyer and Ernst Raj (2013), using the radiosonde data (00 UTC) for 30 years (1971–2000), reported that the Mumbai, Chennai, and Kolkata metros have relatively low pollution potential due to their proximity to the coast, which is influenced by the processes such as land/sea breeze circulations.

3.4.2 Inland Region

Figure 9 shows the 5-year monthly and seasonal mean variation of ventilation coefficients over three cities, Bangalore, Hyderabad, and Nagpur, located inland with increasing latitude. Maximum in VC (8669 ± 10,164 m2/s for Bangalore and 8765 ± 10,209 m2/s for Hyderabad) occurs in the SMS (maximum in July for Bangalore—12,870 ± 13,128 m2/s and Hyderabad—10,129 ± 10,618 m2/s), and minimum VC (2370 ± 2040 m2/s for Bangalore and 3134 ± 1860 m2/s for Hyderabad) occurs during PMS (minimum in April for Bangalore—1150 ± 1370 m2/s and Hyderabad—1933 ± 2242 m2/s).

Fig. 9
figure 9

Monthly and seasonal variation of VC for Bangalore (a, d), Hyderabad (b, e), and Nagpur (c, f) generated using 5-year (2015–2019) daily data

Full size image

From Fig. 3, one can observe that low wind speeds (2–4 m/s) in PMS are responsible for an observed minimum in VC though PBLH range from 800 to 1600 m during this season. Low pollution potential (APPI—5) was observed for Bangalore and Hyderabad during the SMS, and high pollution potential (APPI—2) occurs for both cities during the PMS. However, in Nagpur, maximum VC (3964 ± 1168 m2/s) is observed during the PMS (maximum in July—5522 ± 7940 m2/s) and minimum VC (2722 ± 2886 m2/s) in winter (Minimum in December- 2027 ± 2339 m2/s). This observation implies that Nagpur offers medium pollution potential during the PMS (with APPI—3) and high pollution potential (APPI—2) during the winter. Annual mean VC values for Bangalore, Hyderabad, and Nagpur are observed to be 5510 ± 7351 m2/s, 5195 ± 5736 m2/s, and 2722 ± 2886 m2/s, respectively.

Kumar (2019), using radiosonde data at 06:00 UTC during 2012–13, observed similar seasonal variation in VC at Hyderabad with maximum value occurring during SMS (July and August) and minimum during winter followed by PMS. This variability is attributed to higher wind speeds during the SMS compared to PMS and winter. Kompalli et al. (2014), using the radiosonde measurements at 9:00 UTC (14:00 LT) during 2011–2012, observed that the maximum VC occurred at Nagpur during the PMS and minimum during the SMS and followed by the winter season. Maximum in VC during the PMS was attributed to higher PBLH owing to thermal convection driven by high surface temperature and during winter to the calm winds and shallow PBLH.

3.4.3 Desert Region

Figure 10 shows the 5-year monthly and seasonal mean variation of ventilation coefficient over two cities, Jodhpur and Jaisalmer, located near the Thar desert. Maximum in VC occurs in the PMS (7062 ± 5478 m2/s) for Jodhpur and SMS (6081 ± 4454 m2/s) for Jaisalmer. For both cities, maximum VC occurs during March (9473 ± 9524 m2/s for Jodhpur and 7540 ± 5165 m2/s for Jaisalmer). Minimum VC (2237 ± 2012 m2/s for Jodhpur and 2229 ± 2045 m2/s for Jaisalmer) occurs during PoMS (minimum in November for Jodhpur—1443 ± 1802 m2/s and Jaisalmer—1950 ± 1224 m2/s). Figures 3, 4, and 5 suggest that moderate to high wind speeds (4–8 m/s) in association with moderate PBLH (800–1200 m) result in high VC during PMS and SMS, while low wind speeds (2–4 m/s) and PBLH (400–800 m) result in observed low VC values during the PoMS. This observation implies that the below medium pollution potential (APPI—4) is observed during the PMS and SMS, respectively, for Jodhpur and Jaisalmer. High pollution potential (APPI—2) is observed during PoMS for both cities. Annual mean VC values for Jodhpur and Jaisalmer are observed to be 4934 ± 4761 m2/s and 5000 ± 4545 m2/s, respectively.

Fig. 10
figure 10

Monthly and seasonal variation of VC for Jodhpur (a, c) and Jaisalmer (b, d) generated using 5-year (2015–2019) daily data

Full size image

Vittal Murty et al. (1980) reported that the maximum in VC occurs in April and the minimum during January for Jodhpur. However, this is during the afternoon hours at 12:00 UTC in contrast to the present study, which uses the data at 09:00 UTC corresponding to a fully developed mixed layer.

3.4.4 In and Around IGP Region

Figure 11 shows the 5-year monthly mean and seasonal variation of ventilation coefficients over six cities, New Delhi, Lucknow, Kanpur, Varanasi, Patna, and Rourkela, located in and around IGP. Maximum in VC (4202 ± 3869 m2/s at New Delhi; 4814 ± 3114 m2/s at Lucknow; 5267 ± 2822 m2/s at Kanpur; 6251 ± 4297 m2/s at Patna; 5801 ± 3612 m2/s at Varanasi; 4184 ± 2477 m2/s at Rourkela) occurs during PMS. It can be observed from Fig. 3 that in this season, PBLH in the range 800–1600 m caused by high surface temperatures and wind speeds in the range 4–6 m/s are observed, which lead to high VC during this season. Maximum VC occurs in September for New Delhi (5717 ± 5549 m2/s), Lucknow (7421 ± 5394 m2/s), Kanpur (5633 ± 3656 m2/s) and in March, July, and May for Patna (7388 ± 3946 m2/s), Varanasi (7441 ± 7425 m2/s), and Rourkela (5006 ± 3456 m2/s), respectively. Minimum VC occurs during PoMS (2048 ± 2362 m2/s at New Delhi; 2818 ± 2813 m2/s at Lucknow; 2251 ± 2183 m2/s at Kanpur; 4008 ± 3038 m2/s at Patna; 3062 ± 2592 m2/s at Varanasi) except for Rourkela, where the minimum (1694 ± 2278 m2/s at Rourkela) occurs during SMS. Figure 5 shows that the minimum in VC during PoMS is caused by low PBLH (400–800 m) and wind speeds (2–4 m/s). Minimum VC occurs in July for Lucknow (2618 ± 2798 m2/s) and Kanpur (753 ± 199 m2/s), in November for Patna (2524 ± 1204 m2/s) and Varanasi (2015 ± 1216 m2/s), and in October and August for New Delhi (876 ± 684 m2/s) and Rourkela (962 ± 676 m2/s), respectively. The result indicates that the cities in and around IGP possess medium (APPI—3; New Delhi, Lucknow, Kanpur, Varanasi, and Rourkela) to below medium pollution potential (APPI—4; Patna) during PMS. High (APPI—2; New Delhi, Lucknow, Kanpur, Varanasi) to medium (APPI:3; Patna) pollution potential is observed during PoMS, while at Rourkela, very high pollution potential (APPI—1) is observed. There is an urgent need to frame policies and mitigation measures to combat air pollution effects, which worsen due to the high pollution potential observed in IGP, and large population density (Fig. 1a) is going to be affected by this.

Fig. 11
figure 11

Monthly and seasonal variation of VC for New Delhi (a, b), Kanpur (c, d), Varanasi (e, f), Patna (g, h), and Rourkela (i, j) generated using 5-year (2015–2019) daily data

Full size image

Annual mean VC values for New Delhi, Lucknow, Kanpur, Varanasi, Patna, and Rourkela are observed to be 3568 ± 3471 m2/s, 4240 ± 3210 m2/s, 4335 ± 2839 m2/s, 4773 ± 3649 m2/s, 5193 ± 3943 m2/s, and 2497 ± 2613 m2/s, respectively. Vittal Murty et al. (1980) observed that the maximum VC occurred in PMS (April month) and minimum during SMS (August month) at New Delhi. Iyer and Raj (2013) reported high pollution potential in Delhi during all seasons. However, these studies are done using radiosonde data collected at 00 and 12 UTC compared to our results derived using the satellite data at 09 UTC.

4 Conclusions

We prepared 5-year (2015–2019) mean monthly, seasonal, and annual maps of VC and APPI for the first time for India using the inputs from daily data of PBLH derived using S-NPP vertical profiles and wind speed from ERA-5. Climatology of VC and APPI over 14 cities in India is analyzed, which aids in dispersion modeling for the pollutant concentrations, site planning, and mitigating industrial air pollution. The conclusions are presented below.

1. VC/APPI exhibits spatial and temporal variation, as dynamic variables modulate this; i.e., PBLH and mean wind speed vary with location. The following regions are identified as sites that allow maximum dispersion and are thus suitable for setting up air-polluting industries to reduce their effects on human health. Moreover, already existing air-polluting industries in these regions can maximize the production in respective seasons, and policymakers can move in the air-polluting industries (high CEPI score) in other regions to these sites.

  1. i.

    During the winter season, the east coast of Tamilnadu and Andhra Pradesh shows high VC (6000 to > 10,000 m2/s) and below medium to very low pollution potential (APPI values in the range 4–6).

  2. ii.

    During PMS, Western Gujarat, Southwest of Rajasthan, and the junction region of Uttar Pradesh, Bihar, Chhattisgarh, and Jharkhand show high VC (6000–10,000 m2/s) and APPI of 4–5 (below medium to low pollution potential).

  3. iii.

    During SMS, most of western India and Delhi NCR show good to high VC (6000 to > 10,000 m2/s) and APPI value in the range 4–6 (below medium to very low pollution potential), making these regions possible sites for industrialization or maximum production.

  4. iv.

    During PoMS, the east coast of TN possesses an APPI value of 4 (below medium pollution potential), making it a good site for Industrial establishment.

  5. v.

    Annual averages of VC and APPI suggest that east coastal Tamilnadu and western Gujarat consistently show below medium to very low pollution potential (APPI ≥ 4).

2. Maximum VC is observed at New Delhi, Lucknow, Kanpur, Jodhpur, Patna, Varanasi, Rourkela, and Nagpur during PMS; at Jaisalmer, Mumbai, Hyderabad, and Bangalore during SMS; and at Kolkata and Chennai during the winter season. Minimum VC is observed at New Delhi, Jaisalmer, Lucknow, Kanpur, Jodhpur, Patna, Varanasi, and Mumbai during PoMS; at Kolkata, Rourkela, and Chennai during SMS; and at Hyderabad and Bangalore during the PMS. This information can be used by authorities in planning the vehicular and industrial pollution mitigation measures, thus reducing the effects on human health.

3. Using the annual average VC, the cities with increasing pollution potential are identified as Mumbai, Chennai, Bangalore, Hyderabad, Patna, Jaisalmer, Jodhpur, Varanasi, Kolkata, Kanpur, Lucknow, New Delhi, Nagpur, and Rourkela. This knowledge enables the regulatory authorities to prioritize the air pollution mitigation measures in different cities.

The present study has more significant implications for performing a similar analysis for the globe, prioritizing the locations for the industrial development in various countries. One can adopt a comparable method to create global maps of VC/APPI. International regulatory authorities can monitor the setting up of the polluting industries by different countries in the highly ventilated corridors (APPI—6) to reduce air pollution-induced climate change effects.