Investigating the spatial and temporal variation of Aerosols and Cloud Parameters over South Asia, using remote sensing

The most important component in determining the spatiotemporal distribution of aerosol at local and regional levels is aerosol optical depth (AOD). In this study, data has been obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Aqua satellite to examine spatiotemporal variations in AOD and their effects on the Angstrom Exponent (AE), and clouds parameters, namely cloud fraction, cloud optical thickness, cloud top pressure, cloud top temperature, ice cloud water path, liquid cloud water path, ice cloud effective radius, and liquid cloud effective radius over South Asia from July 2002 to July 2021. The highest values of AOD (0.53–0.7) were observed in the Indo-Gangetic basin (IGB) region over South Asia. The value of AOD of 0.7 is observed in the IGB region during summer. The 0.2 AOD value is observed in winter. The highest mean AOD (0.63 ± 0.09) observed in Bangladesh is due to the noteworthy increase in agricultural activities. The negative correlation between AOD and AE was noticed in Karachi (−0.24), Lahore (−0.04), Rawalpindi (−0.08), Mumbai (−0.03), Kathmandu (−0.49), Colombo (−0.19), and in Kabul (−0.51). A positive correlation is observed in Delhi (0.21), Bangalore (0.09), and Dhaka (0.17).


Introduction
Aerosols assemble from a combination of various natural and artificial sources, each aerosol particle is a composite of various chemical elements that determine a material's refractive index (Alam et al., 2010a;Ranjan et al., 2007). Aerosols affect the earth directly and indirectly due to the absorption and emission of solar radiation. Remote sensing, and radiation transfer techniques are used to determine the optical properties of aerosols and the spatial and temporal effects of aerosols on climate change (Tripathi et al., 2005). Spatial and temporal variations are close to nature seasonally and geographically (Y. J. . Satellite datasets have been used to track the spatiotemporal distribution of aerosols at local and global scales to solve this spatial limitation issue more effectively (Y. J. . Anthropogenic aerosols have been found to alter clouds and their optical characteristics in a variety of measurements and they indirectly affect the cloud size, shape, and density, and change cloud development processes like cloud's lifetime, precipitation, and cloud formation (Spracklen et al., 2011;Vitousek et al., 2004). A significant portion of the MODIS-Aqua satellite provides a useful data source for comprehending how AOD affects clouds and the Angstrom exponent (AE) (Prasad et al., 2004). AE is the parameter that determines the size of the aerosol particles. The light as it travels through the atmosphere because of cloud droplet scattering and absorption its attenuation is measured by the cloud optical thickness (COT). It has numerous uses in radiative transfer, climate change, and consequently in calculating the Earth's radiation budget (Prasad et al., 2004). Cloud Fraction (CF) is the percentage of pixel size that covers the cloud. We selected the combined data for our analysis because MODIS provides CF data for daytime and nighttime separately or jointly. Cloud top pressure (CTP) is determined as the highest altitude and clear portion of the cloud, and is equivalent to cloud top height above sea level. If correlated with surface properties and atmosphere, CTP needs to be analyzed (Balakrishnaiah et al., 2012). Cloud top temperature (CTT), measures the radiation of specific wavelengths which is emitted from the top of the cloud. Cloud effective radius (CER) types Ice Cloud Effective Radius (ICER), and Liquid Cloud effective radius (LCER). The CER is the cloud droplet's mean value and size distribution in the atmosphere. The CWP has two types of parameters that were used in this study, namely ICWP and LCWP.
In recent years decline in the quality of the air in Asian countries due to aerosol emissions from both human and natural activities (Bilal et al., 2021;Mhawish et al., 2020). The South Asian nations of Pakistan, India, Nepal, Sri Lanka, Afghanistan, Bangladesh, Maldives, and Bhutan make up 40% of the Asian continent's population. In the south and east of the Asian continent, several investigations on aerosol-cloud interactions have been carried out to address the geographical and temporal variability of aerosols and clouds (Alam et al., 2010a(Alam et al., , 2014Balakrishnaiah et al., 2012;Kang et al., 2015). Since the deployment of the MODIS satellite, numerous analyses of the AOD data have been carried out to examine the pollution in South Asia ( Dahutia et al., 2019;J. Huang et al., 2020).
In the past few years, increasing industrialization has affected the air quality so it is important to investigate the properties of aerosols and cloud parameters and check their effects on climate change (Dahutia et al., 2019;Gopal et al., 2016;J. Huang et al., 2020;Tang et al., 2014).
Analysis of the aerosol properties over Karachi (Tariq & Ul-Haq, 2018) found that the highest AOD occurred in July due to a higher concentration of coarse mode particles. Due to the predominance of small aerosol particles, Mhawish et al. (2021a, b) discovered previously unidentified aerosol hotspots across the eastern Indian coastline states of Odisha, West Bengal, and Bihar using high spatial resolution data from Multi-Angle Imaging Spectro Radiometer

MODIS
MODIS operates onboard two satellites namely Aqua and Terra and is called Earth Observing System (EOS). These two satellites allow scientists to study aerosols in space with accuracy (Xiong et al., 2009). MODIS instrument has radiometric sensitivity of 12 bits and has 36 bands of wavelength from 0.41μm to 14.4μm (Kang et al., 2015). MODIS data is useful to check the effect of aerosols on the formation of clouds and their parameters (Prasad et al., 2004). In this study, we have used deep blue (DB) Aqua-MODIS (MYD08_M3 v6.1) monthly product at 1°×1° spatial resolution from the period July 2002-July 2021. AOD at 1° × 1°, AE at 1° × 1°, and Aqua monthly cloud parameters (CF), (COT), (CTT), (CTP), (ICWP), (LCWP), (ICER), (LCER)at 1° × 1° have been obtained from http://giovanni.gsfc.nasa.gov.To understand the association of AOD with AE and cloud parameters.  were observed in the north of Pakistan, the southeastern region of India, Afghanistan, Nepal, and Bhutan. The highest value of AOD ˃ 0.6 exists over south-eastern Pakistan, Gujranwala, Sialkot, Faisalabad, Jacobabad, Larkana, Hyderabad, and Nawabshah (Tariq et al., 2022). Due to the growing industry, crop residue burning, high population density, burning of biomass, heavy urbanization, burning of coal, and weather conditions the high AOD (>0.7) observed in the IGB region over South Asia (Sharif et al., 2015). The annual mean AOD over Bangladesh is ˃ 0.6 due to the emissions from brick kilns, transportation of sea salts from the Arabian Sea, and desert dust particles from the Indian region (Mainul et al., 2015). Figure 2b shows AE value is small when coarse-mode particles exist and high where it is decreasing particles (fine-mode) exist.       (Ali et al., 2016). Increased temperature conditions and reduced precipitation, and pollution coming from India may all be contributing factors to the higher AOD (>0.23) over the low-altitude parts of southern Malawi and Lake Malawi and its surroundings (Nyasulu et al., 2020). Table 2 shows the mean values of AOD high in Bangladesh (0.63 ±0.09). (Tariq & ul-Haq, 2020) AOD is observed to be maximum in July (1.00 ± 0.34) and has an associated AE value of 0.85 ± 0.29 over Lahore and Karachi. Balakrishnaiah et al., (2012) high  Temporal variation of AOD, AE, and cloud parameters    Figure 4a shows the correlation map of AOD and AE. The negative correlation -0.4 to -0.13 was observed over the major area of Afghanistan, Pakistan, Sir Lanka, Nepal, and India's south region. The highest positive values (0.13 to 0.4) were noticed over east India, Bhutan, and northwestern Bangladesh. The AOD was observed to be higher in eastern Pakistan due to the presence of bigger dust particles, and AE is reported to be lesser in the southwestern area of Pakistan (Tariq et al., 2021).

Relationship between AOD and AE
The correlation (r) value for the selected cities of South Asia is shown in Table recorded a correlation of (0.93) obtained among MISR and Aqua AOD with a slope of 0.95 and a y-intercept is 0.003. rise, cloud cover rises as well, changing the cloud properties. This is because areas with low air pressure have a greater propensity to build up aerosol particles, which are necessary for cloud formation (Rosenfeld et al., 2001). Figure 5b exhibits the comparison of Aqua AOD and CF and the value of correlation is 0.691. Alam et al., (2014) recorded in Karachi, Lahore, and DG Khan, the association between AOD and CF was more than 0.6, 0.42, and 0.37 in Peshawar and Swat areas, and no clear relationships between AOD and CF were discovered. Kang et al., (2015) estimated that R² values were discovered to be lowest for BJ (0.0004) and greatest for KM (0.176). The correlation was discovered to be negative for the regions BJ (-0.02), HH (-0.03), CD (-0.07), and LH (-0.08) and positive for the remaining regions (Kang et al. 2015). In the region, KM had the highest positive r value (0.42) during the period (2003 -2013) over China (Kang et al., (2015).

Relationship between AOD and CTP
The correlation map between AOD and CTP is shown in Figure 4d.   Kang et al., (2015); Yoram J. Kaufman, Koren, et al., (2005) analyzed the regions KM and NJ, and the correlation coefficient was found to be extremely positive at 0.48 and 0.47, respectively, and strongly negative r (-0.40) for the region LH. Moreover, across the Atlantic Ocean, it was noted that CER was trending downward while AOD was rising.  negative values are also seen over Sylhet (Ali et al., 2016)

Relationship between AOD and CWP
The Figure 4h shows the map of south Asia's positive correlation between AOD and ICWP 0.13 -0.4 noticed over Quetta, Lahore cities of Pakistan, Mumbai, Karnataka cities of India, and the lowest negative correlation values observed in Sir Lanka, and north Pakistan. Myhre et al., (2007) reported a positive link between AOD and CER. They believed that as CTP falls, AOD and CER rise, reducing the positive association between AOD and CER when CTP is high (Myhre et al., 2007). The highest correlation of 0.05 is shown in Mumbai. The highest value of R² (0.01) was in Rawalpindi (Pakistan) and Kathmandu (India). Figure 5h exhibits the correlation between AOD and ICWP and the value of r is positive over south Asia (0.518).
In Figure 4i the map of AOD and LCWP the positive correlation value (0.13-0.4) observe in middle India, and the negative correlation is shown in Pakistan, Bangladesh, and Sir Lanka. The highest value of r is 0.4 is shown in Mumbai. The relationship between AOD and LCWP is shown in Figure 5(i) the correlation r between both AOD and LCWP is 0.485.

Seasonal variation of AE
The seasonal variation of AE increased during the summer season. Figure 7a shows the map of DJF with the highest values at (1.36-1.7) observe in the area of India, Nepal, Bhutan, and Bangladesh. The lowest values of (0.7-1.03) were observed in the southern area of Pakistan and Afghanistan during the winter season. Aerosol layer production in nearby Pakistani areas is mostly caused by the burning of agricultural leftovers in parts of northwestern India (Vadrevu et al., 2011). Figure 7b shows MAM the highest ranging from (1.36 to 1.7) covering the area of southeastern India, Bhutan, Sir Lanka, and Bangladesh. A moderate value of 1.19 was observed in the middle region of India, and Afghanistan's north side. The lowest value of (0.7-1.03) was observed in the region of Pakistan, and also observe in Afghanistan's south. In the winter, AE reveals rising trends of 1.04%, 1.06%, and 2.65% over Lahore, Faisalabad, and Multan, respectively, whereas AOD displays growing trends of 1.46%, 2.05%, and 0.86% over the same cities (Babu et al., 2013;Dey & Di Girolamo, 2010). Figure        Khyber Pakhtunkhwa, Punjab, and Sindh Provinces of Pakistan and India. Kang et al., (2015) for all the places where AOD and CTT were negatively associated, the CTT was found to be high in late fall and low in the summer. The lowest values notice in the north of Pakistan, and the South region of India. Figure 12b shows a map of MAM and the highest values observed in Afghanistan, Punjab, and the Sindh Provinces of Pakistan, India region, and also observed in Bangladesh. Figure 12c show JJA season variation with the lowest values (28-28.9) observed in the major area of Pakistan, India, and Sir Lanka. Figure 12d SON seas highest values observed in Afghanistan, Pakistan, Bangladesh, Nepal, and the northwestern region of India. Kang et al., (2015) believed that as CTP decrease AOD and CER rise, and positive association between AOD and CER when CTP is high.

Conclusion
The relationships between AOD, Angstrom exponent, and cloud parameters discuss in this study from July 2002 to July 2021. The highest AOD is observed in these areas where the economical and agricultural activities are high and AOD is low in these areas where population density is less and limited agricultural activities. The mean value of AOD is highest in Bangladesh (0.63). In