The Impact of Lockdown on eBC Mass Concentration over a Semi-arid Region in Southern India: Ground Observation and Model Simulations

Received: April 30, 2021
Revised: August 27, 2021
Accepted: August 30, 2021

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Cite this article:

Palle, S.N.R., Tandule, C.R., Kalluri, R.O.R., Gugamsetty, B., Kotalo, R.G., Thotli, L.R., Akkiraju, B., Lingala, S.S.R. (2021). The Impact of Lockdown on eBC Mass Concentration over a Semi-arid Region in Southern India: Ground Observation and Model Simulations. Aerosol Air Qual. Res. 21, 210101. https://doi.org/10.4209/aaqr.210101

HIGHLIGHTS

• The contribution of eBC_FF is –29.79% and eBC_BB is 56.41% during lockdown.
• The amount of energy trapped in the atmosphere decreases to 13.31% during lockdown.
• The reduction of heating rate is 13% due to composite and 25% due to eBC aerosols.
• MERRA-2 BC is well correlated with Aethalometer eBC concentration in both the years

Keywords: COVID-19, Equivalent black carbon, Fossil fuel, Radiative forcing, Semi-arid

1 INTRODUCTION

In terms of urbanization and population, India is one of the fastest-growing countries in the world. Being exceedingly vulnerable to air quality, it is a well-known global hotspot of air pollution which regularly severe. A recent study by Thomas et al. (2019), over the Indian region showed that the number of hazy days increases at ~2.6 days per year according to the biomass burning activities. In this situation, the World Health Organization (WHO) declared coronavirus disease 2019 (COVID-19) as a pandemic due to widespread global infection with its first outbreak in Wuhan, Hubei, China. Globally, as of 5^th July 2021, there have been 183.37 Million confirmed cases of COVID-19, including 3.97 Million deaths, resulting in lockdown in many countries worldwide. In India, the first COVID-19 positive confirmed case was reported on 30^thJanuary 2020 in India's Southern state, Kerala (Prasad et al., 2020). Henceforth, the number of confirmed COVID-19 positive cases closed at 500; the first nationwide lockdown (named Janata Curfew) for fourteen hours was on 22^nd March 2020. This curfew was imposed suspending passenger travel both by air and on road, except essential sectors in the entire country. The rules framed by the government resulted in the reduction of human activities during the COVID-19 pandemic which impacted the air quality and helped in improving/understanding the air quality. Therefore, it provided an excellent opportunity for the government to implement a strategic plan focused on emphasizing multi-pollutant emission reductions and overall air pollution-related risk.

Investigations on air quality over Indian regions during this unprecedented lockdown showed a significant improvement in air quality (Chauhan and Singh, 2020; Jain and Sharma, 2020; Sarfraz et al., 2020; Kalluri et al., 2021). The parameters used by the investigators in the above-said investigations in quantifying the air quality are particulate matter (PM), nitrogen oxides (NO₂), sulfur dioxide (SO₂), ozone (O₃), carbon monoxide (CO), and meteorological parameters. For example, Sharma et al., 2020 reported a reduction of 43% in PM₁₀, 31% in PM_2.5, 18% in NO₂, and 10% in CO, during the lockdown period compared to the previous years. On the other hand, negligible changes in SO₂ and growth in O₃ by 17% are seen over India. Hence, there was a unique prospect to retrieve the anthropogenic interference on air quality aspects from the local to the global scales during the lockdown situation.

One of the essential constituents of particulate matter is equivalent black carbon (eBC). The incomplete combustion of fossil fuels and biomass burnings are the primary sources of eBC aerosols in the atmosphere (Li et al., 2020; Petzold et al., 2013). eBC is a graphitic form of carbon particles with unique physical properties and can impact regional air quality (Kaur et al., 2020). Due to its absorbing nature in short-wave and long-wave radiation regions, it has tremendous climate implications (IPCC, 2013), especially over south Asia where absorbing aerosols abundance is high (Fan et al., 2021; Shalini et al., 2020; Kalluri et al., 2020b; Reddy et al., 2019). Hence, eBC has considered the second most potent climate forcing agent between carbon dioxide and methane (Gogoi et al., 2019; Jacobson, 2001). Absorbing aerosols like eBC aerosols can significantly heat the atmosphere and cool the surface (Kalluri et al., 2016, 2020a) also heat the surface (for example, eBC on snow (Kang et al., 2020)), resulting in a net positive forcing on Earth-atmospheric system (Gogoi et al., 2017). Keeping this in view the eBC aerosols and radiative forcing impact due to the COVID-19 lockdown from 25^th March to 20^th June 2020 and the similar data from the corresponding dates of the years from 2016 to 2019 over a semi-arid location, Anantapur has been reported and also presented in Table 1.

 
2 SITE DESCRIPTION AND METHODOLOGY

 
2.1 Site Description

This study analyses the impact of social and travel lockdown on aerosol properties over Anantapur, semi-arid, rain shadow Rayalaseema region in the peninsular Indian state of Andhra Pradesh. The observational site (Department of Physics, Sri Krishnadevaraya University) in Anantapur district is situated on the southern edge of the Anantapur town, as shown in Fig. 1. The observational site is approximately 10 km away from the city's centre, situated beside the national highway 42 (Anantapur–Chennai) and just 7 km away from the national highway 44 (Hyderabad–Bangalore). Additionally, the observational site is close to the railway route connecting Bangalore and other north Indian megacities. This observation site experiences frequent droughts and the second-lowest rainfall in India. The southwest monsoon contributes > 60–70% of the total rainfall, and the average rainfall is only ~150 mm during the northeast monsoon period (Raja Obul Reddy et al., 2016). The possible aerosol sources are present at this site during the summer (March–May) including agricultural burning activites, fossil fuel burning activities, and transported/windblown dust from north-western winds (Kalluri et al., 2017; Gopal et al., 2014, 2015, 2016). Domestic burning is the main aerosol source in winter (December–February) (Gopal et al., 2017; Lingaswamy et al., 2017; Kalluri et al., 2016).

Fig. 1. Map of India showing various observational site (green circles) under ISRO-ARFI Project in which the present study location of Anantapur, Andhra Pradesh, and Google map showing the Department of Physics, Sri Krishnadevaraya University campus (Measurement Location) is highlighted.

The government of India imposed lockdown for more than two months with several relaxations in different phases. The timeline of the lockdown with restrictions and relaxations is given in Table 2. These phases are namely before lockdown (01^st March–24^th March), henceforth termed as BLD (24 days); lockdown 1.0 (LDP1; 21 days) from 25^th March to 14^th April; lockdown 2.0 (LDP2; 19 days) from 15^th April to 03^rd May; lockdown 3.0 (LDP3; 14 days) from 04^th May to 17^th May; lockdown 4.0 (LDP4; 14 days) from 18^th May to 31^st May; lockdown 5.0 (LDP5; 30 days) from 01^st June to 30^th June; the entire lockdown period (LD; 98 days) from 25^th March to 30^th June (LD considered as the mean for all the lockdown phases).

 
2.2 Meteorology

Meteorological conditions play a crucial role while discussing variations of aerosols at a particular site. The surface meteorological parameters [viz., temperature (T in °C), relative humidity (RH in %), wind speed (WS in m s^–1), and wind direction (WD in °)] have been measured on a day-to-day basis at the observational site by employing Automatic Weather Station (AWS) (Campbell Scientific, made in Canada). Fig. 2 shows the day-to-day variation of air temperature, relative humidity, wind speed, and wind direction during the study period. The daily mean surface air temperature varied in the range of 28–38°C (29–37°C), whereas RH varied in the range of 21–73% (18–65%) during 2020 (2019). The winds are mostly from the southwest, with a maximum wind speed of ~7 m s^–1 during both years. Thus, the meteorological conditions during both years almost resemble the same except in magnitude.

Fig. 2. Variation in daily average temperature (°C) and relative humidity (%) along with wind rose plot showing local wind direction and distribution of wind speed (m s^–1) over the study location during the months of March–June for the years 2019 and 2020.

 
2.3 Instrumentation

Simultaneous eBC mass concentration measurements and spectral aerosol optical depths (AODs) have been carried out regularly by deploying a seven-channel Aethalometer and five-channel MICROTOPS-II Sunphotometer, respectively. In the present study, quality-controlled data sets have been used for the respective lockdown phases.

The employed MICROTOPS-II Sunphotometer measures the intensity of direct solar irradiance at five narrow-band spectral channels centred at 380, 500, 870, 936, and 1020 nm through optical collimators with a full field view of 2.5°. The spectral AOD and total columnar water vapour content are the salient parameters in the sunphotometer determined based on the Bougher-Lambert-Beer law. Further details of the instrument, like working principle, theory, data analysis, and calibration techniques, have been published elsewhere (Ichoku et al., 2002; Hussain et al., 2018). Sunphotometer data is not accessible after 01^st May 2020 due to unfavourable sky conditions.

Measurements of eBC aerosol mass concentration have been carried out using a seven-channel (370, 470, 520, 590, 660, 880, and 950 nm) Aethalometer (Model AE-42 of Magee Scientific, California, USA) by operating at 4 LPM of flow rate with the recording time base of 3 min interval. Before considering the analysis data, correction of the loading effect for the raw data was applied, as described in Virkkula et al. (2007). As a result, Aethalometer manufacturers have reported ±5% uncertainty (Allen et al., 1999; Hansen, 2005). However, further details about measurement principles, data analysis, correction schemes were described in several earlier studies by Arnott et al. (2005); Virkkula et al. (2015, 2007); Weingartner et al. (2003); and hence not discussed more here.

2.4 MERRA-2 BC

Gridded BC data (0.625° × 0.5° longitude-by-latitude resolution) was retrieved from NASA's Modern Era Retrospective Analysis for Research and Applications, Version-2 (MERRA-2) re-analysis data for the measurement period (https://disc.gsfc.nasa.gov/datasets/). More detailed information about MERRA-2 is available elsewhere (Lasko et al., 2018; Qin et al., 2019). The gridded value geographically closest to a study region is selected as the comparison object for observation in this site. The black carbon aerosol transport is examined in the study using reanalysis wind data from MERRA, (Rienecker et al., 2011) reanalysis products at a grid resolution of 0.25° by 0.25° (http://mirador.gsfc.nasa.gov/).

In addition to a standard meteorological analysis, MERRA-2 includes an aerosol analysis, as described in Randles et al. (2017). Aerosols in MERRA-2 are simulated with a radiatively coupled version of the Goddard Chemistry, Aerosol, Radiation, and Transport model (GOCART; Chin et al., 2002; Colarco et al., 2010). GOCART treats the sources, sinks, and chemistry of 15 externally mixed aerosol mass mixing ratio tracers: dust (five noninteracting size bins), sea salt (five noninteracting size bins), hydrophobic and hydrophilic black and organic carbon (eBC and OC, respectively; four tracers), and sulfate (SO₄^2–). Both dust and sea salt have wind speed- dependent emissions (Tandule et al., 2020). Primary sulfate and carbonaceous aerosol species are emitted principally from fossil fuel combustion, biomass burning, and biofuel consumption, with additional biogenic sources of organic carbon. Secondary sources of sulfate include chemical oxidation of sulfur dioxide gas (SO₂) and dimethyl sulfide (DMS), and we include a database of volcanic SO₂ emissions and injection heights.

The hourly mean Boundary Layer Height (BLH), were obtained from ERA-5 (https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels?tab=form).

 
2.5 Estimating the Fossil Fuel and the Biomass Burning Component of Equivalent Black Carbon Mass Concentrations

After applying the loading effect correction algorithm, the corrected eBC mass concentration data has been apportioned to identify the fossil fuel and biomass burning contribution to the total eBC measured. The power-law aerosol absorption coefficients (σ_abs) is proportional to λ^–α, where α is the absorption Ångström exponent, and λ is the wavelength which describes the spectral dependence of the absorption coefficient (Ångström, 1964). Consequently, for a wavelength pair, it can be derived using the following relation:

In general, coal-burning, or fossil fuel emissions, contain more eBC than other light-absorbing compounds, thus (α_FF ~1) is also applicable to combustion of fossil fuels (Kirchstetter et al., 2004; Zotter et al., 2017). The aerosols produced by biomass combustion, on the other hand, are composed of light-absorbing organic substances and eBC, therefore α_BB ~2 is relavant to biomass combustion (Sandradewi et al., 2008a; Zotter et al., 2017). They show strong light absorption in the near UV region of the spectrum and have a minimal contribution in the near IR region of the spectrum. With this background, Sandradewi et al. (2008a) developed a two-component Aethalometer model to obtain a qualitative measure of the relative presence of fossil fuel burning and biomass burning sources in the measured eBC. This model utilizes the light absorption by fossil fuel and wood-burning sources with the basic assumption that fossil fuel and wood burning are the dominating sources of carbonaceous aerosols. Hence, the corrected σ_abs obtained from the Aethalometer is the sum of  due to fossil fuel burning and  due to biomass burning:

Using Beer-Lambert’s Law, the relation between the corrected σ_abs, λ, and α for biomass burning (BB) and fossil fuel burning (FF) is:

For a given value of α_FF and α_BB, the values for , , , and  have computed using Eqs. (2)–(4). According to this study, α_FF and α_BB values are 1, 2 respectively, which were comparable to those in the literature (Kaur et al., 2020; Kumar et al., 2020; Rajesh and Ramachandran, 2017; Vaishya et al., 2017). The contributions of fossil fuel burning (eBC_FF @ 880 nm) and biomass burning (eBC_BB @ 880 nm) to total eBC @ 880 nm are then derived via the corresponding mass absorption cross section (MAC) values (MAC_BB and MAC_FF, respectively):

where MAC_FF (880 nm) and MAC_BB (880 nm) is 16.6 m²g^–1 taken from the manufacturer (Hansen, 2005). Zotter et al. (2017) reported that there is no significant difference between MAC_FF to MAC_BB at 880 nm. Hence the same MAC value was taken for both MAC_FF and MAC_BB.

The limitations of the deconvolution of eBC to eBC_FF and eBC_BB using this two-component Aethalometer model is that assuming the fossil fuel and wood burning are the dominating sources of black carbon aerosols. It is also known that wind-blown dust is one of the light-absorbing aerosols, especially absorbing UV regime (e.g., 370 nm in Aethalometer). This absorption by dust particles in shorter wavelengths may be difficult to identify the absorption of eBC_BB. One more limitation is that assumption of appropriate values for α_FF and α_BB. This model is sensitive on the selected wavelength range and the prior assumed α_FF and α_BB values, as some studies used the 370–880 nm (Dumka et al., 2018; Kaur et al., 2020; Vaishya et al., 2017), others the 470–950 nm (Kumar et al., 2020; Sandradewi et al., 2008a, b; Zotter et al., 2017) and 370–950 nm (Rajesh and Ramachandran, 2017). Assumed α_FF ranged between 0.9–1.1 and α_BB between ~1.7–3.5 values by (Schnaiter et al., 2003, 2005; Kirchstetter et al., 2004; Sandradewi et al., 2008a, b; Stockwell et al., 2016; Diapouli et al., 2017; Rajesh and Ramachandran, 2017; Vaishya et al., 2017; Zotter et al., 2017; Dumka et al., 2018; Kaur et al., 2020; Kumar et al., 2020).

 
2.6 Estimation of the Short-wave Direct Aerosol Radiative Forcing

The short-wave direct aerosol radiative forcing (SDARF) estimated by using the standard atmospheric model, Santa Barbara Discrete Ordinate Radiative Transfer (SBDART), which uses the discrete ordinates radiative transfer (DISORT) integration of the radiative transfer equations. Several investigators reported that this model is well suited to estimate the earth-atmosphere system's radiation budget (Aruna et al., 2016; Babu et al., 2002; Kant et al., 2020; Moorthy et al., 2009). The necessary and significant inputs for computing SDARF are spectral AOD, spectral single scattering albedo (SSA), and the scattering phase function. The AOD and eBC mass concentration data sets have been obtained from MICROTOPS sun photometer and Aethalometer during the entire phases. The SSA, Phase function, Angstrom exponent data sets were obtained from the optical properties of aerosols and clouds (OPAC) model (Hess et al., 1998).Additional inputs like surface albedo, atmospheric profiles of temperature, pressure, columnar ozone, and water vapor content are necessary to estimate shortwave radiative forcing in SBDART model. Surface reflection collected by the 8-Day, Level 3 Moderate Resolution Imaging Spectroradiometer (MODIS) onboard Terra (MOD09A1) and Aqua (MYD09A1) satellites at different wavelengths (0.469, 0.555, 0.645, 0.859, 1.24, 1.64, and 2.13 m) were used to calculate the surface spectral albedo. The columnar ozone and water vapour content datasets from the MICROTOPS Ozonometer have also been obtained.

Shortwave direct radiative forcing at surface, atmosphere, and top of the atmosphere and their associated heating rates (K day^–1) within the atmosphere are estimated using this model. The details are furnished elsewhere (Liou, 2002). Due to the lack of sunphotometer AOD values after LDP3, the aerosol radiative forcing was estimated only for BLD, LDP1, LDP2, and LDP3. Therefore, the value reported for LD for forcing related parameters is the mean values from LDP1 to LDP3 only.

 
3 RESULTS AND DISCUSSION

 
3.1 Impact of Lockdown on eBC Mass Concentration

Fig. 3 shows the temporal change in the diurnal concentrations of eBC, eBC_FF, and eBC_BB before and during the lockdown overlaid by atmospheric boundary layer height. It has been noticed from the figure that, irrespective of phase in lockdown, the concentration of eBC and eBC_FF is low during the noon hours. Two prominent peaks are noticed, primary peak during the morning hours around 06:00–07:00 h and another less prominent peak (also known as nocturnal peak) during the evening hours around 19:00–20:00 h for all the periods. These diurnal variations result from the various eBC local emission sources and their strengths and atmospheric boundary layer height. In general, the height of atmospheric boundary layer (ABL) starts increasing from 06:00–07:00 h and reaches the maximum at 14:00–15:00 h. Later on, it starts decreasing till 19:00 h and tends to be invariable from 20:00 to 06:00 h (Fig. 3). Chan et al. (2012) found a strong correlation between ABL and incoming solar radiation. After the sunrise, a gradual increase in the radiation onset breaking the aerosols trapped in the residual layer (caused due to the night-time inversions) leading to a sharp increase in near-surface pollutants concentration (eBC: primary peak) popularly known as fumigation effect. As the day advances, due to the impact of thermal buoyancy, ABL increases rapidly and redistributes aerosols vertically in deep ABL, making the concentration of eBC minimum at the surface level. The solar radiation tends to diminish as the day progresses, especially in the evening. After sunset, the turbulent mixing weakens, and thus ABL deforms to a shallow stable ABL near the earth's surface (Siva Kumar Reddy et al., 2021). The thermal inversions separate residual layer downwards and night-time traffic emissions leading to the nocturnal peak can be observed in the diurnal eBC cycle. The impact of ABL on diurnal eBC is commonly seen in all the phases, irrespective of the years. But the magnitude of the eBC concentrations varies from phase to phase in the year 2020 compared with 2019. During the year 2019, from Figs. 3(a)–3(f), the magnitude of the eBC concentrations gradually decreased due to the seasonal effect; the Indian summer monsoon. The magnitudes of eBC concentration during 2020 also follow the same trend as 2019 but with remarkable reductions. These remarkable reductions in eBC concentrations in the year 2020 might be attributed to the movement restrictions on all public and private vehicles imposed to control the spread of the COVID-19 pandemic, as shown in Table 1. It is interesting to note that when compared with 2019, the concentrations of eBC_FF in the year 2020 decreased from before LD to LDP4 due to vehicular movement restrictions and an increase in the eBC_BB. People who live near the sampling site often cook with wood fuel and participate in a variety of farming activities, which may led to the rise in eBC_BB. Since the two-component Aethalometer model assumes the possible source of eBC is from fossil fuel combustion and biomass burning, the decrease in vehicular emission increases the share of eBC_BB. After LDP4, relaxations have allowed the public to resume their daily activities with minimal restrictions; hence, eBC_FF starts dominating over eBC_BB in Anantapur. To have a better understanding of the impact of LD on the surface eBC, eBC_FF, and eBC_BB concentrations, we herewith illustrate the mean eBC, eBC_FF, and eBC_BB concentrations for each LD phase (derived from daily mean values) along with those relative changes between 2019 and 2020 as shown in Fig. 4. The concentration of eBC_FF in (Fig. 4(a)) 2020 shows exponential decrease from BLD (1.4 ± 0.4 µg m^–3) to LDP5 (0.4 ± 0.1 µg m^–3), whereas eBC_BB (Fig. 4(b)) showed increase in LDP1and attained a peak value of 0.6 ± 0.1 µg m^–3 during LDP2 and started decreasing from LDP3. While considering the mean variation between BLD and DLD during 2020, it has witnessed a decrease of 33.95% in eBC and 47.19% in eBC_FF with a 23.49% increase in eBC_BB. When compared the relative change between 2020 and 2019, eBC_FF always showed decreasing tendency. However, eBC_BB showed an increasing tendency from LDP1 to LDP4 with relative change as high as 161% during LDP2. Moreover, to have a much clearer understanding of the impact of the unprecedented cessation of human activities during the lockdown on the regional aerosol concentration, the percentage of contribution of eBC_FF and eBC_BB to the total eBC for the respective phase is shown over each bar in Fig. 4. One can observe that in the year 2020, eBC_FF decreased from LDP1 (where nearly all services and factories were suspended, except emergency services) to LDP4. When the travel restrictions are resumed (i.e., during LDP5), eBC_FF increased from 58% in LDP4 to 79% in LDP5. This pattern was not observed in the year 2019.

Fig. 3. Effects of lockdown on diurnal variation of equivalent black carbon (eBC-blue) mass concentrations, equivalent black carbon concentrations from fossil fuel (eBC_FF-red), and biomass burning (eBC_BB-black) in compared with 2019 (a–g) & 2020 (h–n) scaled to the left side. The Atmospheric boundary layer height is represented with values overlaid on line plot (Green) scaled to the right side. a and h represent before lockdown, b and i–lockdown period 1, c and j–lockdown period 2, d and k–lockdown period 3, e and l–lockdown period 4, f and m–lockdown period 5, g and n–mean for lockdown period 1–5. Complete details and dates for the lockdowns are provided in the text.

Fig. 4. Phases wise mean eBC concentration values contribution of fossil fuel combustion (a) and biomass burning (b), during in 2019 (orange) and 2020 (green) along with absolute change in respective values between 2020 and 2019 (violet). Values illustrates on top of the orange and green bars are the percentage share of with respect to the total eBC during 2019 and 2020 respectively, whereas values on violet bar are the relative change in percentage between 2020 and 2019.

As shown in Table 2, the mean eBC concentration in the year 2020 during BLD, LDP1, LDP2, LDP3, LDP4, and LDP5 is much lower when compared with the same periods during 2016–2019. eBC emissions declined from 1.66 ± 0.5 to 1.07 ± 0.49 µg m^–3 (35 %) during the lockdowns period over Anantapur, as compared to the same period in the previous 4 years (2016–2019). eBC emissions during BLD was decreased from 2.23 ± 0.11 to 1.74 ± 0.37 µg m^–3 (22%), in LDP1 from 2.40 ± 0.23 to 1.54 ± 0.36 µg m^–3 (36%), in LDP2 from 2.05 ± 0.13 to 1.59 ± 0.20 µg m^–3 (22%), in LDP3 from 1.48 ± 0.14 to 1.07 ± 0.19 µg m^–3 (28%), in LDP4 from 1.36 ± 0.21 to 0.73 ± 0.11 µg m^–3 (46%) and in LDP5 from 1.03 ± 0.20 to 0.47± 0.07 µg m^–3 (54%) as compared to the same period in the last 4 years. It was observed that eBC concentration, when compared with BLD 2016–19, there is also a noticeable decrease in the BLD phase of 2020. The government of India has imposed specific emission control measures over various industrial sectors, which have played a significant role in keeping the particulate emission in control and including that of eBC. Among those control measures, the nationwide implementation of Bharat Stage (BS) emission standards in 2000 were followed more rigorous criteria with Bharat stage IV (sulfur content of 50 ppm) in 2010 and Bharat Stage VI norms (sulfur content of just 10 ppm) in 2020 along with steps to improve the fuel quality. BS-VI norms significantly reduce contents like sulfur, lead, benzene, and much more, improves the octane/cetane number for petrol/diesel. A recent investigation on long-term variation from 2007 to 2015 of surface eBC aerosol mass concentration over India showed a significant decreasing trend with 0.07 µg m^–3 y^–1 over peninsular India and 0.242 µg m^–3 y^–1 overall the Indian region (Manoj et al., 2019). Similar decreasing trend of 0.02 µg m^–3 month^–1 over the eastern Himalayan region between 2009 to 2015 was reported by Sarkar et al. (2019). Ravi Kiran et al. (2018) have reported 0.27 µg m^–3 decrease per year over tropical site Gadanki during 2008 to 2017. There was a dramatic drop in eBC levels during the lockdown days in the year 2020 due to numerous restrictions and curfews imposed to combat the COVID-19 pandemic, which reduced anthropogenic emissions over Anantapur.

 
3.2 Intercomparison of Observed and Model eBC Concentrations

We have observed the impact of the lockdown restrictions on surface eBC concentrations. Now, we would like to look at the response of the model data during the lockdown restrictions and compare the observed and model data. We have used MERRA-2 surface mass concentration of carbon black (BCSMASS) model data for validation with ground-based observation data for our current analysis. The phase-wise mean observed and MERRA-2 BC concentrations along with bias values are shown in Fig. 5(a). It can be found that the MERRA-2 component has a clear variation for the lockdown phases in the year 2020. However, the bias between observed and MERRA-2 during 2020 is consistently low when compared with 2019 data. During 2019, the bias is always higher than 0.30, while it is below 0.30 for 2020, irrespective of the sign. Also, the daily mean basics observed and MERRA-2 BC are well correlated in both the years with R values of 0.71 and 0.80 for the years 2019 and 2020, respectively as show in Fig. 5(b). In the lockdown period in 2020, MERRA-2 BC shows lower values than the corresponding period in previous year (2019). MERRA-2 satellite-based BC data also showed improved air quality at Anantapur during the lockdown. Based on the lockdown phase-wise mean, the correlation between observed and MERRA-2 are very strong, 0.94 and 0.99 for the years 2019 and 2020 (Fig. 5(c)). We can also see that the MERRA-2 has underestimated the eBC concentrations in both the years, except for few days where the concentrations are below 1 µg m^–3. Therefore, it can be hypothesized that MERRA-2 BC concentrations perform better in low pollution levels than in higher pollution levels. However, Zhao et al. (2021) noticed the overestimated MERRA-2 BC concentrations in Nanjing City, where the concentration levels are much higher when compared to our study. They also found a good correlation between observed and MERRA-2 BC on monthly mean values (0.93) rather than daily mean values (0.78). Qin et al. (2019) also reported that MERRA-2 BC concentrations are overestimated (averagely 101.6% higher) than ground-based observations; however, the ground-based and MERRA-2 mass concentrations of BC were highly correlated (0.77) over Beijing.

Fig. 5. Comparison of observed eBC concentrations with MERRA-2 BC concentration for various lockdown phases (a) and the values over the bar represents the bias between observed and modelled values, red for 2019 and blue for 2020. Validation of MERRA-2 BC at (b) daily level and (c) phase wise mean with observations.

 
3.3 Shortwave Direct Aerosol Radiative Forcing

Table 3. provides SBDART input parameters and shortwave direct aerosol forcing for before and during lockdown phases. The positive sign in relative change denotes an increase, whereas the negative sign denotes a reduction in the AOD value. The AOD levels decreased some what from 0.39 (BLD) to 0.37 (DLD). The relative change of AOD between BDL and DLD is –03.60% for the 2020 phases and 13.09% for the 2019 phases. When comparing the AOD values observed between 2020 and 2019 phases before LD, AOD values decreased by only 12.41%. During DLD, it decrease by 28.95%, a comparably substantial decline in AOD during 2020 phase. Under normal circumstances, residual crop burning and continental dust particles associated with higher wind speed are responsible for high AOD values observed during the month's March–May over the observational location (Kalluri et al., 2016).

A regional synthesis of eBC, eBC_FF, eBC_BB, and AOD, having noted for various lockdown phases, attempted to estimate the radiative forcing due to composite and eBC aerosols before and during the lockdown phases. The mean picture of the phase-wise averaged SDARF due to composite aerosols at the top of the atmosphere (TOA), in the atmosphere (ATM), and at the surface (SUR) over study location is illustrated in Fig. 6. The surface and TOA forcing due to composite aerosols increased by 11.33% from –27.88 Wm^–2 (BLD) to –24.89 Wm^–2 (DLD) at the surface and only 3.62% increase from –05.47 Wm^–2 (BLD) to –5.27 Wm^–2 (DLD) at TOA. The atmospheric forcing from BLD to DLD has been reduced from 22.41 Wm^–2 to 19.61 Wm^–2, almost a 13% reduction. Interestingly, a negative trend was observed from BLD to DLD during the 2020 phases, whereas it is positive during the 2019 phases. The high surface forcing might be due to large scattering particles inhibiting the solar radiation reaching the earth's surface.

Fig. 6. The shortwave direct aerosol radiative forcing due to composite (bar) and eBC (symbol and line) aerosols, the light green bar, and blue sphere line plot representing the 2020 phases and the cyan bar with patterns and red squares plot represents 2019 phases for comparison.

The SDARF is due to eBC represents the difference between the SDARF for aerosols with and without the eBC component. To estimate the forcing due to eBC alone, optical parameters for aerosols have been deduced again, keeping BC concentration equal to zero and constructed columnar AOD for the eBC-free atmosphere from OPAC. The surface forcing due to the eBC component is –08.58 Wm^–2 during BLD and –06.43 Wm^–2 during LD, which contributes almost 25–30% contribution to the composite aerosol forcing. The eBC forcing at TOA is 2.60 before LD and 2.14 during LD, with a 40–47% contribution to the composite forcing. Interestingly, the amount of energy trapped in the atmosphere due to eBC is 11.19 before LD and 8.56 during LD. i.e., eBC contributes almost 43–50% to the composite forcing. When compared the 2020 phases with the 2019 phases, the eBC contribution to the composite forcing in the 2020 phases is reduced to 60%, 50%, and 33% at SUR, ATM, and TOA concerning the 2019 phases. The positive forcing at the TOA arises due to the absorption by black carbon in the atmosphere, which also leads to a net cooling effect (negative forcing) at the surface. During LD, the composite atmospheric heating rate was significantly lower (0.55 K day^–1) than BLD (0.63 K day^–1), i.e., 13% reduction. On the other hand, the equivalent heating rate was slightly greater (0.79 K day^–1) than BLD (0.75 K day^–1), implying a 5% increase in the 2019 phases. Due to eBC over Anantapur, the atmosphere's overall heating rateis 0.31 K day^–1 before LD and is reduced to 0.24 K day^–1 during LD. During the lockdown phases, the heating rates due to composite aerosols decreased by 13%, whereas it is 25% due to eBC aerosols. It indicates that the restrictions on vehicular emission and anthropogenic activities during lockdown phases have significantly influenced the eBC aerosols concentrations rather than composite aerosols.

The magnitude of the aerosol radiative forcing depends not only on the aerosol composition but also on the column abundance. Consequently, it is a strong function of the AOD. The aerosol forcing efficiency is generally defined as the rate at which the atmosphere is “forced” per unit of aerosol optical depth, increasing as the AOD increases. To understand the impact of lockdown on forcing efficiency over Anantapur we also analyzed the aerosol forcing efficiency in different lockdown phases. Before LD, the net TOA and SUR forcing efficiency were about –14.18 and –72.30 Wm⁻² τ⁻¹, respectively. For the DLD, the corresponding values were about –14.16 and –66.89 Wm⁻² τ⁻¹, respectively.

 
4 CONCLUSIONS

The effect of unprecedented cessation of human activities during the COVID-19 pandemic was studied using in-situ based observations of eBC aerosols and their implications to short wave radiative properties from 01^st March to 30^th June 2020 over a semi-arid station in Southern India. The salient outcomes from the present investigation are as below:

1. The mean eBC mass concentration before and during LD was found to be 1.74 µg m^–3, and 1.12 µg m^–3, i.e., decreases by more than ~35% during LD than BLD due to the reduction of anthropogenic activities and meterology.
   
   
2. Due to the restricted vehicular emission during the lockdown period, the contribution of fossil fuel burning to total eBC showed a distinct variation between BLD (81%) and DLD (64%).
   
   
3. A good correlation between observed and MERRA-2 BC concentrations is found with a correlation coefficient greater than 0.70.
   
   
4. During LD, the net forcing at the top of the atmosphere and the surface due to composite aerosols varied from –52 to –6.19 Wm^–2, and –22.91 to –29.35 Wm^–2, respectively, and net forcing in the atmosphere changes in the range 17.27 to 23.16 Wm^–2.
   
   
5. The contribution of eBC alone to the composite forcing within the atmosphere during LD (BLD) was found to be 43.64% (49.93%), respectively.
   
   
6. The restrictions on vehicular emission during the lockdown phases have significantly influenced the eBC rather than composite aerosols. The heating rate justifies this with a 13% reduction due to composite and a 25% reduction due to eBC aerosols.

 
ACKNOWLEDGEMENTS

The authors wish to thank the Indian Space Research Organization, Bangalore for the financial support provided by ISRO-GBP under the ARFI and NOBLE Project.