Observational evidence of aerosol-warm cloud interaction over two urban locations in eastern India

Abstract

Aerosol-cloud interaction remains challenging due to the large uncertainties caused by the associated meteorological effects. This study examines the aerosol-warm cloud interaction over the cities of Bhubaneswar and Rourkela. A negative cloud effective radius (CER)-cloud optical depth and CER-cloud top pressure (CTP) relationship is found in all the regimes of aerosol optical depth (AOD) over Bhubaneswar and Rourkela, excluding CER-CTP association in heavy pollution scenarios over Rourkela. However, a significant positive CER-cloud water path (CWP) correlation is observed in all the cases of AOD over both cities. This can be attributed to strong competition between aerosol and cloud droplets for water vapor association. CER is found to increase with AOD in all the regimes of relative humidity (RH) over both cities, excluding low and high cases of RH over Bhubaneswar. In this scenario, enhanced collision and coalescence efficiency and increased water vapor content encourage the merging of smaller droplets, resulting in the growth in effective radius of clouds. Negative pressure vertical velocity (PVV) over these cities infers that the upward movement of the air parcels helps in the growth of cloud droplets and aerosol particles besides enhancing the cloud cover. The shift from the Anti-Twomey effect to the Twomey effect has been noticed in both urban areas, with the Twomey effect being the major influence. Overall results indicated aerosol-induced heating may increase the response of turbulent heat flux (sensible heat over the land), leading to enhanced stability at the lower troposphere and subsequent suppression of vertical mixing. It results in moisture trapping near the surface and increases warm clouds in Rourkela. However, the predominance of a positive semi-direct effect over Bhubaneswar leads to an adverse relationship between warm cloud cover and aerosols.

Graphical abstract

Keywords

aerosol-cloud interaction / aerosol optical depth / cloud properties / stability / moisture

1 Introduction

In the context of aerosol impact on cloud microphysical properties (Gryspeerdt et al., 2014; Kant et al., 2017, 2019a, 2019b, 2019c; Liu et al., 2017; Panda et al., 2023; Zhao et al., 2023) and precipitation (Kant et al., 2019a, 2023; Liu et al., 2022), there is a need to develop a deeper understanding of the associated interactions over India and its neighborhood as this region currently contributes significantly to the aerosol loading (Kant et al., 2019b, 2023; Thomas et al., 2019). Large uncertainty is observed due to the difficulties in separating the meteorological effect from the aerosol effect on clouds and precipitation while quantifying the anthropogenic contribution (Toll et al., 2019). Studies accounting for satellite observations, and numerical modeling suggest that atmospheric aerosols significantly influence cloud microphysical properties and the associated precipitation processes (Sporre et al., 2014; Malavelle et al., 2017; Toll et al., 2017).

The increase in cloud condensation nuclei (CCN) concentration in the atmosphere is mostly due to the enhanced supersaturation and aerosol loading, which could result in smaller cloud droplets and a surge in cloud droplet concentration, leading to an increase in cloud albedo at a constant liquid path (Twomey, 1977). However, a decrease in cloud effective radius (CER) may lead to an increase in cloud lifetime (Albrecht, 1989) and, thus, the surge in liquid water path resulting rise in the depth of clouds (Wang et al., 2014). A negative CER- aerosol optical depth (AOD) indicates a decrease in CER with an increase in aerosol loading and is usually termed as the “Twomey effect” (Twomey, 1977; Sporre et al., 2014). However, a positive AOD–CER relationship could be observed over land and ocean many a time, which has been established through several observational and modeling studies (Grandey and Stier, 2010; Liu et al., 2017; Qiu et al., 2017; Kant et al., 2019a, 2019b). Tang et al. (2014) and Wang et al. (2015) have shown different behaviors of CER with AOD at low to the high regime of AOD. However, Feingold et al. (2001) suggested that there are three kinds of CER effects with aerosols: (i) saturation of the CER with high AOD, (ii) decrease in CER with increasing AOD due to suppression of water vapor super-saturation caused by significant giant cloud particle, and (iii) an increase in CER with AOD results enhanced cloud surface area, leading to increase in CER due to higher number of smaller droplets.

Whether the AOD-cloud optical depth (COD) relationship is positive or negative depends on the balance between the simultaneous increase in CER and decrease in cloud water path (CWP; Costantino and Bréon, 2013). A study by Koren et al. (2008) suggested an increase in cloud cover with increasing atmospheric aerosol concentration. The chemical composition of aerosols and the general atmospheric thermodynamic and dynamical conditions also play a vital role in the AOD-CER relationship (Yuan et al., 2008) and aerosol-cloud interaction (ACI) over a region (Koren et al., 2010; Stathopoulos et al., 2017). ACI has consequences concerning cloud cover, cloud droplet size, and precipitation patterns. The strength and nature of these interactions can be significantly influenced by different meteorological conditions including temperature inversion, stability, and humidity. Assessing ACI under various meteorological conditions contributes to a better understanding of the intricate relationship. The dynamics of ACI can be altered by changing meteorological conditions, which has further consequences regarding regional and global climatic trends. Thus, the relationship between aerosol and cloud parameters depends upon factors such as prevalent meteorological conditions, cloud regime, and aerosol types, size distribution, amount, and chemical composition (Koren et al., 2010; Kant et al., 2017, 2019b; Liu et al., 2017).

ACI, from the observational point of view, is to be either “physical” – the rise in hygroscopic aerosols leading to smaller CCNs that influence cloud formation, or “artifact” – the contamination of cloud pixels due to retrieval (e.g., Ignatov et al., 2005; Wang et al., 2014). When a cloud cover region is mistaken as a high aerosol loading, as it occurs sometimes, it results in an incorrect aerosol-cloud relationship (Zhang et al., 2005). Therefore, the impact of aerosols on cloud properties is quite complex, and it is often very difficult to derive a valid relationship. An indirect effect of aerosol was initially examined broadly during the Indian Ocean Experiment (INDOEX) (Ramanathan et al., 2001). Consequently, continuous studies were launched to study the aerosol properties over the Indian subcontinent (mainly aerosol hotspot regions like Indo-Gangetic Plain) using several network observations, including AERONET (Tripathi et al., 2005), other campaigns (Satheesh et al., 2009), and ARFINET (Moorthy et al., 2013). Specific programs like the Continental Tropical Convergence Zone (CTCZ; Sengupta et al., 2013), and Cloud-Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX; Kulkarni et al., 2012) were carried out to examine ACI over the Indian region. Recently, Kant et al. (2023) reported that aerosol loading enhances the low-level cloud cover (warm cloud) over the Northern Bay of Bengal region. Also, the study of Panda et al. (2023) over the South-Asia region revealed that ACI can be associated with the Indian Ocean Dipole scenarios, and the governing meteorological factors, may be the possible reason for the growth of increasing optically thick middle and high-level clouds. There is no doubt that these studies have advanced our understanding of the role of aerosols in governing cloud processes over the Indian region, but they are limited to specific seasons and time periods.

However, these investigations were not focused on the response of aerosols to warm clouds under different meteorological instances across urban areas having diverse climatic features. Bhubaneswar, situated in eastern India, has a tropical climate and substantially better air quality than most Indian cities. Meanwhile, Rourkela is in the western portion of the state and has a more of temperate climate with relatively poor air quality because of industrial operations. Based on the backdrop of the observational and modeling studies, and distinct climatic conditions (although both are in eastern parts of India), it would be motivating to examine ACI over these cities. In this study, we attempt to understand the aerosol indirect effect on warm cloud properties over the above-mentioned land areas under different meteorological scenarios by utilizing the 19 years (January 2003 – December 2021) of MODIS/Aqua-derived data sets. The details of the study areas, data, and adopted methodology are described in the following section. Section 3 discusses the results, and the last section summarizes the findings of this study.

2 Geographical area, data, and methodology

2.1 Study area

The present study is carried out over Bhubaneswar (20.29°N, 85.82°E; 33 m above the sea level) and Rourkela (22.26°N, 84.85°E; 197.29 m above the sea level), which are major cities or urban agglomerations of Odisha, an eastern Indian state. According to the Köppen classification (Köppen et al., 2011), both Bhubaneswar and Rourkela belong to the tropical climate zone. The annual mean temperature over these cities is 27.4°C and 26.5°C, respectively. Summer is hot and humid with mean temperatures of ~40°C and ~35°C, respectively, and winter is much cooler with mean temperatures of ~12°C and 8°C, respectively, over Bhubaneswar and Rourkela. The Indian summer monsoon brings moisture to this region, and major rainfall occurs during June–September, with annual rainfall of 1542 and 1448 mm, respectively. Also, the pre-monsoon thunderstorms contribute to the annual rainfall budget over these two cities during the months March–May. Since these two cities are important to Odisha in many aspects and are continuously growing, they are included in the Government of India’s smart city project. The present study intends to analyze the aerosol–warm cloud interaction over these two cities by utilizing satellite observations during 2003–2021 as per the data availability. This would help understand the characteristics of the warm clouds giving rise to rainfall over these two cities irrespective of seasons.

2.2 Data used

The present study uses 19 years (January 2003–December 2021) Aqua satellite retrieved MODIS (Moderate Resolution Imaging Spectroradiometer) level 2 (Collection 6) data set with a resolution of 0.1° × 0.1°. MODIS from Aqua has multi-band spectral coverage with a swath width of ~2300 km (King et al., 2003). The satellite passes over the study region at 13:30 IST (local time) and thus is very useful in obtaining the data relevant to warm clouds located over land where they are usually well developed. MODIS uses 36 bands from the visible to infrared wavelengths (0.41 to 15 µm) to measure the cloud properties. The AOD product (MYD04_L2) is derived from the 5 min interval data with a swath of 10 km from the cloud-free pixel with a resolution of 500 m at nadir (Remer et al., 2013). The AOD is retrieved from three MODIS channels, whereas the cloud properties are retrieved from six spectral visible and near-infrared wavelength channels. Level 2 (MYD06_L2) product is used in this study for the parameters CER, Cloud Fraction (CF), CWP, COD, and cloud top pressure (CTP). The detailed information regarding the aerosol and cloud properties algorithm adopted in MODIS data retrieval is mentioned in Remer et al. (2005, 2008). Notably, the uncertainty in the MODIS retrieved AOD data over land is ∆AOD = ± 0.05 ± 0.15AOD, ranges from 10% to 21% for COD (Minnis et al., 2004), and 1.0 to 3.0 µm for CER (Platnick et al., 2017).

Besides cloud properties, ERA5 hourly data with 0.25° spatial resolution is used in the study for elucidating meteorological variables, including temperature (at 1000 and 700 hPa), relative humidity (RH), pressure vertical velocity (PVV), and wind speed (WS) at 850 hPa. Details of ERA5 data sets are mentioned in Table ST1. Notably, ERA5 is a global re-analysis data set and has been used in meteorology and aerosol-cloud interaction-related studies (e.g., Kant et al., 2019b; Sarkar et al., 2021, 2022; Paul et al., 2022; Kumar et al., 2023; Panda et al., 2023).

2.3 Methodology

Aerosol characteristics are determined from the cloud-free pixels retrieved by the cloud detection system (Liu et al., 2017). Overestimation may result from biases caused by instrumental errors, calibration problems, and uncertainties in satellite retrieval techniques in the AOD measurements from the MODIS-Aqua satellite. The retrieval method can erroneously identify cloud particles as aerosols, resulting in overestimation of AOD as well. The cloud detection system is not perfect, and some residual clouds may be missed resulting in AOD overestimation (Kaufman et al., 2005). The MODIS AOD retrieval algorithm (Remer et al., 2005; Remer et al., 2008) reduced cloud contamination by using a sensitive cloud detection technique (Martins et al., 2002). Another reason for overestimation may be the misclassification of high AOD regions like the locations with significant desert dust. To avoid the overestimation, cases with AOD > 1.0 are omitted over Bhubaneswar and Rourkela (Kant et al., 2019a, 2019b, 2023). Also, cases with COD < 2 are omitted to avoid the potential effect of the surface overestimation or contamination in cloud retrieval (Kant et al., 2023; Panda et al., 2023). In this study, warm clouds are only considered with cloud top temperature (CTT) > 273 K, CTP < 680 hPa, and CWP < 200 g⁻² to avoid including deep convective clouds (Wang et al., 2014; Kant et al., 2023).

As indicated previously, meteorological conditions significantly influence the cloud microphysical properties (Stevens and Feingold, 2009). Meteorological parameters used in this study include RH, WS, PVV, and lower tropospheric stability (LTS). LTS is defined as the variance or difference between the potential temperature (LTS = θ₇₀₀ − θ_surface) of the free troposphere (700 hPa) and the surface (1000 hPa), and serves as an indicator of the thermodynamical conditions of the atmosphere (Klein and Hartmann, 1993).

To estimate the aerosol-cloud interaction, MODIS retrieved fixed CWP is divided into eight bins within the range of values 0–200 g⁻² at an interval of 25 g⁻². ACI is estimated for every bin using Eq. (1), similar to that adopted by Feingold et al. (2001), Qiu et al. (2017), and Kant et al. (2019b):

(1)\mathrm{A}\mathrm{C}\mathrm{I}=-\frac{{\tau }_a}{r_e}\frac{{\mathrm{\Delta }r}_e}{{\mathrm{\Delta }\tau }_a}=-\frac{{\mathrm{\partial }\mathrm{l}\mathrm{n}r}_e}{{\mathrm{\partial }\mathrm{l}\mathrm{n}\tau }_a}{|}_{\mathrm{C}\mathrm{W}\mathrm{P}},

where τ_a indicates MODIS retrieved AOD as a proxy for the aerosol amount, and r_e is the CER for fixed CWP. Notably, ACI quantifies how variations in aerosol loading influence the effective radius of cloud when the CWP is kept constant.

To quantify the sensitivity of cloud properties to changes in aerosol loading, a parameter named ‘susceptibility’ is considered. Satellite retrieved cloud parameter susceptibility (S) is defined as

(2)S=-\frac{\mathrm{d}\mathrm{l}\mathrm{n}A}{\mathrm{d}\mathrm{l}\mathrm{n}{N}_d},

where ‘A’ is the change in cloud albedo A for an incremental rise in droplet number concentration N_d with other cloud properties being constant. The logarithmic form of Eq. (2) lowered the sensitivity of S to determine the accuracy of cloud properties and aerosols (Sorooshian et al., 2009; Chen et al., 2014; Kant et al., 2023).

Two sample student’s t-test is performed for statistical analysis to characterize the probability P value of data sets at a 95% significance level for MODIS retrieved AOD and cloud microphysical parameters. The statistical significance test indicates the rejection of the null hypothesis when it is true (Kang et al., 2015; Kant et al., 2019a, 2019b).

3 Results and discussion

3.1 Frequency distribution of AOD with CER

The probability density function (PDF) signifying the percentage of occurrences, and the frequency of AOD, analyzed for two CER regimes, i.e., CER < 14 µm and CER > 14 µm, to understand the relationship between specific sizes of cloud droplets with changing aerosol loading over Bhubaneswar and Rourkela cities (Fig.1). The PDF mode value for AOD is found to be higher in the case of CER > 14 µm (AOD = 0.5) regime than CER < 14 µm (AOD = 0.4) regime over Bhubaneswar (Fig.1(a)). Similar variation for the PDF mode value for AOD is observed over Rourkela, where corresponding values of AOD are 0.36 and 0.28 in the considered regimes (Fig.1(b)). However, the PDF mode value for AOD is higher over Rourkela (PDF = 23.97) compared to Bhubaneswar (PDF = 17.39) in the CER > 14 µm regime (Fig.1(a) and Fig.1(b)). The relative change in AOD frequency between CER > 14 µm and CER < 14 µm regimes, is observed to be higher over Rourkela (28.57%) compared to Bhubaneswar (25%). However, the frequency of the ratio of change of AOD (Fig.1(c)) between the two regimes of CER (> 14 µm and < 14 µm) is observed to be higher in the case of Bhubaneswar (0.29) than Rourkela (0.24). The ratio of AOD change between the two CER regimes greater than 0 in the case of both cities, can be attributed to greater dominance of droplet size of larger sizes (i.e., > 14 µm). The availability of more hydrometeors or droplets with larger sizes implicate that more of them attended sufficient size to precipitate out of clouds (Rosenfeld and Gutman, 1994; Kant et al., 2019a, 2019b; Toll et al., 2019). In such a scenario, the polluted condition tends to encourage precipitation, implying that the cases with CER > 14 µm will surge with increase in AOD. However, most of the droplets in the regime CER ˂ 14 µm are not large enough to precipitate. Thus, there is a higher probability of occurrence of clouds with CER > 14 µm due to coalescence process, and sufficient moisture availability, over Rourkela compared to Bhubaneswar.

3.2 Variation of COD, CTP, and CWP as a function of CER

Variations of COD, CTP, and CWP as a function of CER are examined to understand the effect of aerosols on warm cloud parameters (Fig.2). For this purpose, AOD is considered in three bins, i.e., (i) AOD < 25th percentile (moderately polluted), (ii) AOD = 25th−75th percentile (polluted), and (iii) AOD > 75th percentile (heavily polluted) in accordance to the considerations of Liu et al. (2017). AOD data are divided by bin width of 7th percentile. The two-sample student’s t-test is performed to examine the statistical significance of the data sets by considering P-value, where P < 0.05 indicates that these data sets are statistically significant.

COD decreases with an increase in CER over both cities in all the AOD regimes (Fig.2(a) and Fig.2(d)). Overall, a negative CER-COD relationship (correlations: −0.57 to −0.30) is observed in both places for all AOD cases. The increased aerosol concentrations can encourage larger droplets to compete with one another for available water vapor. In such cases, water vapor condensation on the numerous aerosol particles results in an increased amount of smaller cloud droplets leading to reduction in the effective radius and the optical depth growth, instead of a substantial increase in CER (Kant et al., 2019b, 2023; Panda et al., 2023). The increased smaller droplet number concentration reduces the collision and coalescence efficiency. Consequently, this competition inhibits the evolution of individual droplets while contributing to the reduction of CER and COD growth (Liu et al., 2017; Kant et al., 2019b; Panda et al., 2023). Also, droplet size reduction decreases sedimentation velocity and enhances cloud liquid water content (negative CER-COD and positive CER-CWP relationships exist in both cities) in non-precipitating clouds (Fig.2), leading to more cloud top cooling and increased evaporation and entrainment (Bretherton et al., 2007; Small et al., 2009; Gao et al., 2021; Kant et al., 2023; Lu et al., 2023). And smaller cloud droplets increase the surface area to volume ratio, resulting in faster evaporation and an increase in negative buoyancy at the cloud top (Small et al., 2009; Gao et al., 2021; Kant et al., 2023; Lu et al., 2023; Zhao et al., 2023).

CTP increases with CER for all cases except the heavily polluted case over Rourkela (Fig.2(b) and 2(e)). The negative CER-CTP (−0.44 to −0.02) association is observed over both the locations in all AOD regimes with an exception, where a weak but positive correlation is found for the heavily polluted (0.06) scenario over Rourkela. The positive CER-CTP relationship exhibits increased CTP and large CER due to the predominance of higher updrafts and increased moisture availability (Koren et al., 2005; Liu et al., 2017; Saponaro et al., 2017; Kant et al., 2019b; Panda et al., 2023). Larger CER and higher CTP result from the enhanced coalescence and condensation of cloud droplets, encouraged by the vertical motion of buoyancy-driven water vapor. However, negative CER-CTP relationship indicates reduced CTP and smaller CER due to lesser updraft and moisture availability (Kant et al., 2019b, 2023; Panda et al., 2023).

Also, CWP increases with CER in all cases for both cities (Fig.2(c) and Fig.2(f)). A positive CER-CWP correlation (0.52–0.79 for Bhubaneswar and 0.03–0.62 for Rourkela) is seen for all cases. The data set is statistically significant at a considered significance level (95%) for all the AOD cases over Bhubaneswar and polluted scenarios over Rourkela. It may be noted that the positive CER-CWP relationship (as observed in the present study) indicates the growth of cloud droplets in the presence of water vapor. Another possibility could be clouds with small CER cannot produce rain and will keep the CWP higher with high aerosol loading (Peng et al., 2002; Kant et al., 2019a, 2019b, 2023).

By definition, COD is inversely related to the CER. If the relationship doesn’t display that (i.e., COD increases with an increase in CER), the effect is over-compensated by the CWP (directly proportional to the COD) effect. In other words, as CER increases, CWP also increases, and perhaps the rate of increase of CWP is larger than that of CER in the case of Rourkela. This is quite interesting as plenty of moisture is available in Bhubaneswar for the cloud drops to grow, while it only grows in Rourkela (a relatively drier place) when just sufficient moisture is available. Since CTP is not changing with an increase in CER, one can interpret that CWP controls the cloud droplet growth in Rourkela.

3.3 Variation in CTP as a function of CF and COD

CTP is an important parameter for aerosol-cloud interaction analysis, and it is considered an indicator of the vertical growth of clouds. A higher value of CTP indicates lower cloud top height and vice versa. CTP value is found to be increasing with CF for all AOD cases over Bhubaneswar and Rourkela (Fig.3(a) and Fig.3(b)). A negative CTP-CF (−0.74 to −0.16) relationship is observed for all the regimes of AOD over both cities, excluding the heavily polluted (0.24) case of AOD over Rourkela. The lower value of CTP indicates the presence of taller clouds extending to higher levels of the atmosphere. The CTP value reaching a higher level up to 750 hPa indicates that CF is less sensitive to aerosol loading. This is because the aerosols cannot be effortlessly vertically lofted so that there is no impact on clouds anymore (Wang et al., 2014). Also, the formation of the cloud droplets depends upon the adiabatic cooling generated by the lifting of an air mass. Water vapor started to condense on the cloud nucleus or CCN since the available water vapor reached to lifting condensation level and consequently, the favorable atmospheric condition helped in the growth of high-altitude clouds (Liu et al., 2017). This may indicate the reason for the increase in CTP with decreased CF, irrespective of aerosol loading. However, cloud vertical extension may somewhat depend upon aerosol loading in some specific environmental conditions, like the heavily polluted AOD case, which shows a relatively higher value of CTP. Increase in CF can be due to the meteorological and aerosol-microphysical effects.

Positive CTP-CF correlation can be due to aerosol microphysical effect and the prevailing meteorology so that it produced lots of low-level clouds but no tall clouds. The aerosol effect may be more obvious in a moderate pollution case, where increase in humidity enhances the production of more cloud droplets and consequently increasing the lifetime of the warm clouds (Koren et al., 2005). A negative CTP-CF correlation may indicate the presence of taller convective clouds. This occurs more often in heavily polluted cases. Another possible reason could be the presence of giant CCN in the heavily polluted case, which favors the formation of large drops quickly. Also, there is a possibility of cloud contamination issues during aerosol retrieval (Zhang et al., 2005) and aerosol humidification aspect (Grandey et al., 2014; Gryspeerdt et al., 2014) to contribute to the CTP-CF relationship. In the present scenario that focuses on low clouds, where the invigoration effect is ruled out, the role of meteorology in governing the CTP-CF relationship could be significant if there is a negligible influence on the retrieval artifacts (i.e., overestimation of AOD in the humid condition/high CF, biases in CER retrievals of MODIS).

A negative COD-CTP (−0.90 to −0.71) relationship is observed over Bhubaneswar and Rourkela in all the cases of AOD (Fig.3(c) and Fig.3(d)). The negative relationship between COD and CTP also implies that CTP will be smaller for larger COD. The current finding of a mostly negative COD-CTP relationship agrees with those reported by Liu et al. (2017). A drizzle can occur when the cloud grows taller and develops a larger drop size, and the CWP also becomes higher (Gao et al., 2014). The increase in CWP with CER in this study (Fig.2(c) and 2(f)) indicates the prevalence of the second aerosol indirect effect, which further implies that precipitation suppression may increase CWP and perhaps an additional increase in COD due to an upsurge in water vapor amount (Costantino and Bréon, 2013).

3.4 Effect of relative humidity on ACI

The study by Feingold et al. (2001) suggested that the indirect effect of the aerosol is highly dependent on the hygroscopicity and PVV. However, the interaction between aerosol and cloud may be influenced by thermodynamical and dynamical processes and prevalent meteorological conditions (Quaas et al., 2010; Gryspeerdt et al., 2014; Wang et al., 2014; Christensen et al., 2016; Gryspeerdt et al., 2016; Andersen et al., 2017; Liu et al., 2017; Kant et al., 2019a, 2019b). Thus, briefly examining the impact of the prevailing meteorological conditions on aerosol-cloud interaction over Bhubaneswar and Rourkela is essential. For this purpose, meteorological parameters such as RH (discussed in this section), LTS, PVV, and WS are considered and discussed in different sections. RH is one of the important factors influencing cloud droplet formation and aerosol particle size. High RH at the cloud base strongly influences the interaction between aerosol particles and cloud parameters (Small et al., 2011; Liu et al., 2017; Panda et al., 2023) and thus, the ‘effect of RH’ is essential to be accounted for ACI (Loeb and Manalo-Smith, 2005; Jeong et al., 2007; Quaas et al., 2010). The investigation of ACI in association with RH at 850 hPa is important since the latter one serves as a significant indicator of moisture availability in the atmosphere, determining the microphysical processes within clouds.

Cloud parameters as a function of AOD are analyzed in three RH bins: low (< 25th percentile value), moderate (within 25th−75th percentile), and high (> 75th percentile value) at 850 hPa (Fig.4). The CER value increased with AOD in moderate to high RH bins over both the cities (Fig.4(a) and Fig.4(e)). It is due to the hygroscopic growth of aerosols due to the condensation of water vapor (Feingold et al., 2013). However, an increased RH in the atmosphere encourages further activation of cloud droplets and the growth of already present droplets (Jones et al., 2009). This indicates that the presence of moderate and high RH helps grow larger droplets due to the sufficient availability of water vapor (Liu et al., 2017). A positive AOD-CER (0.06−0.27) relationship is observed over both the cities in all the regimes of RH except low (−0.05), and high (−0.29) regimes over Bhubaneswar (Fig.4(a) and Fig.4(e)). A positive AOD-CER relationship indicates the dominance of other processes like dynamical and microphysical effects, which is possibly counteracting the aerosol indirect effect on cloud droplets (known as Anti-Twomey effect). The surface mixed layer RH increases, because of the decrease in boundary layer turbulent kinetic energy and the entrainment over the temperature inversion at the boundary layer top (Gao et al., 2021; Li et al., 2022; Kant et al., 2023; Lu et al., 2023). Consequently, cloud droplets form in the thicker saturated layer, and higher RH provides the droplets additional buoyancy subsequently, which causes an increase in the effective radius. Another possible reason for a positive relationship could be the presence of giant CCN and somewhat soluble organic particles (Yuan et al., 2008; Kant et al. 2019b, 2019c; Panda et al., 2023). It is worth mentioning that more activation of CCN (hygroscopic aerosol particles) leads to an increase in cloud droplet numbers. However, the negative AOD-CER relationship is the result of a decrease in CER with an increase in AOD (Twomey, 1977; Panda et al., 2023).

The response of cloud fraction to aerosol loading is complex. On the one hand, increased aerosol loading reduces cloud droplet collision, leading to an increase in cloud fraction by suppressing precipitation. On the other hand, smaller cloud droplets are more susceptible to complete evaporation due to entrainment, which means that increased aerosol may result in smaller spatial distributions for thin clouds. In the current scenario, CF increases with AOD over both cities (Fig.4(b) and Fig.4(f)). A higher value of CF is observed within the AOD range of 0.2–1.0. CF is much higher in moderate and high RH bins. A negative AOD-CF (−0.44 to −0.05) association is found in all the RH regimes over Bhubaneswar. A positive AOD-CF relationship (correlation coefficient in the range 0.35–0.42) is observed in the low and moderate cases of RH, excluding the higher regime over Rourkela. The possible reason for the positive correlation between AOD and CF could be the hygroscopic growth of aerosols (Grandey et al., 2014) over both places. However, an increase in CF with AOD to a larger magnitude with highly loaded water vapor regions (Albrecht, 1989) is possibly due to the influence of a stronger updraft and, consequently, increases the lifetime of clouds. Also, the cloud cover increases with aerosol concentration and, thus, modifies cloud properties. This is because the low atmospheric pressure (i.e., higher altitude) region tends to provide favorable conditions for the formation and growth of clouds in the presence of aerosols and water vapor. The negative AOD-CF relationship indicates the presence of hydrophilic aerosol types (sea salt, organic matter, and sulfate) in the atmosphere (Gryspeerdt et al., 2016). As aerosols were located above the liquid cloud layer, it can increase the potential temperature difference across the inversion layer causing the reduction of cloud top entrainment rate, which leads to cloud thickening resulting in weaker warming of boundary layer, and consequently increased CF (Johnson et al., 2004; Wilcox, 2010; Kant et al., 2023). However, aerosols located below the liquid cloud layer can increase the boundary layer warming to result in increased stability, and consequent moisture content reduction leads to decrease in the CF (Hansen et al., 1997; Wilcox, 2010; Kant et al., 2023).

The value of COD represents the attenuation of light by the absorption and scattering of cloud droplets when passing through the cloud. It has numerous applications in climate change scenarios, including the computation of Earth’s radiation budget (Prasad et al., 2004). Fig.4(c) and Fig.4(g) show that a negative AOD-COD correlation exists for low (−0.3) and relatively high (−0.16) regimes of RH over Bhubaneswar and moderate (−0.15) RH scenarios over Rourkela. However, a positive AOD-COD relationship is observed in the moderate (0.02) RH bin over Bhubaneswar and low (0.05), and high (0.55) RH scenarios over Rourkela. Two possible reasons may be responsible for the negative correlation (COD decreases with AOD), viz., 1) the prevalent radiation-absorbing aerosols evaporate cloud droplets due to heating and the cloud becomes thinner (Alam et al., 2014; Kant et al., 2019a, 2019c, 2023); 2) the absorption of solar radiation by absorbing aerosols result a reduction in cloud reflectance measured by the satellite sensor (Meyer et al., 2013, 2015; Li et al., 2014; Panda et al., 2023). A positive AOD-COD relationship indicates that wet deposition or the presence of a high amount of water vapor or moisture increases COD. Reduced turbulent heat flux lowers vertical mixing, which leads to surface mixed layer warming. This warming effect reduces entrainment across the temperature inversion at the boundary layer top causing surface layer RH to increase (Wilcox et al., 2016; Kant et al., 2023). Consequently, droplets form in the thicker saturated layer, and enhanced RH gives the droplets extra buoyancy, leading to an increase in COD. A decrease in COD in the presence of absorbing aerosols encourages cloud droplets to disperse and consequent thinning (Alam et al., 2014; Kant et al., 2019a, 2019b, 2023). The entrainment-mediated evaporation is increased in the presence of enhanced aerosol loading, which often results in increased concentration of droplets with smaller sizes. The smaller droplets entrain into the area with less moisture, causing evaporation due to the greater surface area. It could consequently reduce the COD (Gao et al., 2021; Liu et al., 2023; Panda et al., 2023).

A positive AOD-CTP (0.12 to 0.54) correlation is found in all RH bins over Bhubaneswar, and Rourkela except the low (−0.3) RH regime over Rourkela (Fig.4(d) and Fig.4(h)). The positive AOD–CTP relationship observed here contradicts the primary previous understanding that high aerosol loading results in high CTP (Wang et al., 2014; Liu et al., 2017; Kant et al., 2019b). It also indicates that lesser CTP does not constantly attend high aerosol loading. Thus, aerosols do not necessarily help in generating higher and more convective clouds in every scenario (Rennó et al., 2013). Also, the variation in large-scale meteorological conditions (Alam et al., 2014) may govern the relationship. The negative AOD-CTP relationship indicates the possibility of higher clouds (i.e., higher cloud top altitude) with lower CTP that could be associated with higher AOD, resulting in the invigoration of the clouds (Sharif et al., 2015).

3.5 Influence of stability on ACI

To understand the impact of atmospheric thermodynamics and dynamics on aerosol-cloud interaction over Bhubaneswar and Rourkela, we divided MODIS-derived data sets according to different LTS bins. Notably, LTS is used as a measure of the thermodynamic state of the lower atmosphere (up to 700 hPa) and defines the atmospheric tendency to enhance or suppress vertical motion (Klein and Hartmann, 1993; Medeiros and Stevens, 2011). Although the classification of thermal stability of the lower atmosphere that is determined based on positive and negative values of LTS, is straightforward, the percentile-based approach allows a complex understanding of stability patterns. This approach is mainly useful when dealing with data sets showing variability and non-normal distributions. The percentile-based classification essentially recognizes that stability is a continuous rather than a binary state, and its distribution can vary across different LTS percentiles. Previous studies such as Liu et al. (2017), Panda et al. (2023), and Wang et al. (2014) have used percentile-based technique to present their analysis. The current study adopts the percentile-based approach to divide cloud parameters as a function of AOD on the log-log scale into three different LTS bins, i.e., Low (< 25th percentile value), moderate (within 25th−75th percentile), and high (> 75th percentile value). LTS bins respectively signify unstable, intermediate, and stable atmosphere over both the cities (Fig.5).

A positive AOD-CER (0.06 to 0.45) relationship is observed in all the bins of LTS over both the locations except moderate (−0.28), and high (−0.5) regimes over the Bhubaneswar (Fig.5(a) and Fig.5(e)). The CER value is higher in the unstable atmospheric condition (i.e., low LTS) than that of the stable one, irrespective of the AOD values. This scenario indicates higher growth of cloud droplets during the unstable atmospheric condition (Anti-Twomey effect) and thus shows a positive relationship over both cities. This is due to the enhanced vertical mixing of water vapor, resulting in stronger aerosol-cloud interaction (Liu et al., 2017). The decrease in turbulent heat flux causes the surface mixed layer to become warmer and thinner because of increased aerosol radiative heating. Due to this effect, entrainment decreases across the temperature inversion at the boundary layer top resulting in an increase in RH at the surface layer (Johnson et al., 2004; Koch and Del Genio, 2010; Wilcox, 2010; Wilcox et al., 2016; Kant et al., 2023). This consequently increases vertical velocity, and subsequently, LWP causes an increase in vertical mixing of water. The negative AOD-CER relationship serves as an indicator of inhibiting cloud droplet growth (Twomey effect) over Bhubaneswar in moderate and high LTS regimes. However, the AOD-CER relationship over Rourkela indicates a favorable condition for cloud droplet growth in both moderate and high LTS regimes.

The negative AOD-CF (−0.6 to −0.06) relationship is observed in all the regimes of LTS over Bhubaneswar, and the positive AOD-CF (0.16 to 0.36) relationship is found over Rourkela in all the bins of LTS (Fig.5(b) and 5(f)). A positive AOD-CF relationship indicates the prevalence of a stronger vertical inversion, which would inhibit vertical cloud growth and vertical mixing, but keeps a moist boundary layer active to generate a favorable condition for the formation and growth of clouds (Wang et al., 2014; Kant et al., 2023) over Rourkela. The negative association infers a decrease in the cloud cover with an increase in the aerosol loading attributing to a semi-direct effect over Bhubaneswar.

A positive AOD-COD relationship is found in the low LTS regime, whereas a positive association is observed in the moderate regime over both cities (Fig.5(c) and Fig.5(g)). However, the correlation is positive for Rourkela and negative for Bhubaneswar in the high LTS regime. The positive AOD-COD relationship indicates the prevailing unstable condition would help in the formation and growth of cloud droplets due to the possible presence of a sufficient amount of water vapor resulting in increased COD. Whereas the negative relationship supports inhibition of the droplet growth due to the presence of absorbing aerosols, consequently dispersing cloud droplets to generate thinner clouds or a radiative effect (Kant et al., 2019a, 2019b). The positive AOD-CTP (0.11 to 0.45) association is observed in all LTS bins over both cities, excluding the high (−0.09) regime over Rourkela (Fig.5(d) and Fig.5(h)). This scenario indicates an increase in the stability at the lower troposphere would enhance turbulent heat flux response resulting in vertical mixing suppression and, consequently, trapping the moisture to encourage the low-level (warm) cloud cover (Kant et al., 2023) over Rourkela region. Whereas, the overwhelming majority of a positive semi-direct effect over Bhubaneswar results in an opposing association between aerosols and warm cloud cover.

3.6 Dynamic effect on ACI

Various meteorological parameters influence the ACI interaction, including PVV (Wang et al., 2014) and wind speed, which is usually termed the dynamic effect of aerosols. PVV measures the dynamic convection strength of the atmosphere and is important for the formation and growth of cloud droplets (Liu et al., 2017). PVV can also be used for determining the strength of cloud development over a particular region. The positive PVV signifies the downward motion, whereas the negative PVV indicates the upward motion. PVV can influence the atmospheric conditions for the growth and development of cloud droplets and vice versa (Jones et al., 2009). Cloud parameter as a function of AOD on a log-log scale with consideration of low (< 25th percentile value), moderate (within 25th−75th percentile range), and high (> 75th percentile value) PVV bins at 850 hPa is analyzed (Fig.6).

A positive AOD-CER association is observed in moderate (0.2 and 0.07) and high (0.12 and 0.22) regimes of PVV, whereas a negative relationship is found in the low PVV scenario over both cities (Fig.6(a) and Fig.6(e)). Relatively larger CER values were observed in the case of low and moderate regimes compared to the high case of PVV over both cities. Low and moderate regimes of PVV indicate that air parcels moving in an upward direction may encourage the growth and development of bigger cloud droplets and, therefore, influence ACI (Jones et al., 2009).

A negative AOD-CF (−0.50 to −0.18) relationship is found over Bhubaneswar in all the cases of PVV (Fig.6(b) and Fig.6(f)). However, a positive AOD-CF correlation is observed in the low (0.3) and moderate (0.32) regimes of PVV, excluding the high (−0.19) case of PVV over Rourkela. A relatively greater value of CF was observed in the low and moderate PVV regimes compared to the high scenario. A positive AOD-CF relationship indicates the growth of cloud droplets and a rise in cloud cover due to increased aerosol loading and water vapor in the atmosphere, exhibiting higher CCN. This happens due to the impact of the upward motion of air parcels and aerosols, eventually increasing the cloud lifetime. It occurs in regions with low atmospheric pressure and higher tendencies to generate conditions necessary for cloud formation by accumulating aerosols and water vapor (Wright et al., 2010; Kang et al., 2015). This phenomenon is considered quite important when estimating the impact of aerosols on CER and CF (Wang et al., 2014; Liu et al., 2017). The increase in turbulent flux enhances vertical mixing, causing the surface mixed layer to cool. Subsequently, entrainment across the temperature inversion at the boundary layer top increases, resulting in a reduction in surface layer RH (Hansen et al., 1997; Wilcox, 2010; Wilcox et al., 2016; Kant et al., 2023). This results in increased entrainment-induced evaporation in the presence of enhanced aerosols, leading to a reduction in CF over Bhubaneswar.

A weak negative AOD-COD (−0.12 to −0.1) relationship is observed over Bhubaneswar in all PVV bins, and a weak but positive (0.06−0.35) relationship is seen in all the regimes over Rourkela (Fig.6(c) and Fig.6(g)). A weak negative relationship indicates that the COD value is relatively larger in the presence of upward motion of air parcels and highly dependent upon moisture availability and vertical depth of cloud. The suppression of cloud droplet growth occurs in the presence of sinking or downward motion of air parcels and, consequently, decreasing COD. A weak positive relationship implies the growth of cloud droplets occurs due to the upward motion of the air parcel and hence, increasing COD. Also, the presence of an absorbing type of aerosol may encourage the dispersion of cloud droplets (Alam et al., 2014; Kant et al., 2019a, 2019b; Panda et al., 2023). The downward motion of air causes more aerosol particles to be entrained in the cloud layer (Gao et al., 2021; Liu et al., 2022; Panda et al., 2023). This suppresses the coalescence and collision process of droplets in the presence of aerosols. Consequent droplet evaporation caused by entrainment reduces the optical depth of the cloud. However, a positive AOD-CTP relationship (0.02 to 0.51) is observed in all the regimes of PVV over both cities, excluding the high (−0.18) scenario over Rourkela (Fig.6(d) and Fig.6(h)). A positive relationship indicates that the rising motion of air parcels can encourage the growth and development of higher and thicker clouds and vice-versa (Panda et al., 2023). The negative AOD-CTP relationship indicates increased vertical mixing and aerosol particles ascent into the cloud deck caused by PVV. This increased entrainment-mediated evaporation of droplets influences the optical properties of clouds, and results in a decrease in CTP (Panda et al., 2023; Zhao et al., 2023).

Besides, the cloud parameters as a function of AOD on a log-log scale for low (< 25th percentile value), moderate (within 25th−75th percentile range), and high (> 75th percentile value) WS conditions at 850 hPa are analyzed (Fig. S2) by following Qiu et al. (2017). The low WS bin is an indicator of the condition where clouds are mostly influenced by local or regional aerosols, whereas moderate and high WS regimes imply that clouds are mostly affected by the long-range transported aerosols. Most cloud parameters show a positive relationship with AOD in all the bins of WS over both cities (Fig. S1). Considering the geographical locations of the cities, our previous studies over the eastern part of India (Kant et al., 2019a, 2109c; Panda et al., 2023), and the current findings, it may be inferred that both local and long-range transported aerosols have more impact over Bhubaneswar, whereas locally produced aerosols have a dominant role over Rourkela.

3.7 ACI and cloud susceptibility

An increase in aerosol loading reduces CER and consequently, increased COD at constant CWP indicating a surge in the absorption and scattering of solar radiation due to the presence of many small cloud droplets (cloud albedo effect), which is represented in terms of ACI (Patel and Kumar, 2016). The quantitative analysis of ACI is carried out by using the Eq. (1) and considering MODIS retrieved fixed CWP divided into seven bins ranging from 0 to 175 g⁻² with an interval of 25 g⁻² (Fig.7). The negative ACI is observed mostly in the scenarios with CWP < 75 g/m² regimes over both the cities except 0 to 25 g⁻² bins over Bhubaneswar. However, positive ACI is found for all CWP bins > 75 g⁻² over both regions. The positive values of ACI indicate the decrease in CER with an increase in AOD (Twomey effect). However, the negative values of ACI indicate an increase in CER with AOD (Anti-Twomey effect). A decrease in the amount of water vapor in the atmosphere does not favor the growth of liquid cloud droplets and therefore, encourages the Twomey effect, whereas the sufficient amount of water vapor present in the atmosphere favors the growth of liquid cloud droplets and thus, encouraging the Anti-Twomey effect (Twomey, 1977). The non-hygroscopic aerosols present in the atmosphere may be responsible for negative ACI for liquid clouds over these two cities. The decrease in boundary layer turbulent kinetic energy and entrainment across the temperature inversion at the boundary layer top causes an increase in surface mixed layer RH (Wilcox et al., 2016; Gao et al., 2021; Li et al., 2022; Kant et al., 2023; Lu et al., 2023) as mentioned earlier. Subsequently, cloud droplets form within the thicker saturated layer, and higher RH provides the droplets more buoyancy (Wilcox et al., 2016), which raises the droplet effective radius. Another possible reason for negative ACI could be the presence of absorbing aerosols which are hydrophobic in nature and prevent cloud droplet growth (Panicker et al., 2010). Also, lack of sufficient water vapor in the atmosphere inhibits the growth of fine-mode hygroscopic aerosols, which prevent the development of clouds (Patel and Kumar, 2016) and hence, discourages positive ACI. It is therefore important to understand the cloud susceptibility, i.e. change in cloud properties due to the perturbation in aerosols, and thus, analyzed in different CWP bins over Bhubaneswar and Rourkela (Fig.8).

The cloud susceptibility is derived (using Eq. (2)) in this study using CER, CF, COD, and CTP as a function of AOD. The susceptibility of CER and CF is found to increase in the different regimes of CWP over both cities (Fig.8(a)−Fig.8(b)). Whereas, susceptibility of COD, and CTP is observed to be increasing over the Bhubaneswar, and decreasing over Rourkela (Fig.8(c)−Fig.8(d)). The cloud susceptibility is mostly reliant on the comparative alteration in CER and CWP with increasing aerosol loading. Upsurge in CCN results in more numbers of smaller droplets under constant CWP. The CWP variability act to alter the cloud susceptibility with aerosol loading. When CWP increases with aerosol loading, it results in cloud albedo enhancement establishing the ‘cloud lifetime effect’. In contrast, a decrease in CWP can terminate the Twomey effect resulting in lower cloud albedo (Chen et al., 2014). The current results indicate the cloud cover over Bhubaneswar to be increasing with CWP and more susceptible compared to Rourkela. This could possibly be attributed to the higher competitiveness among droplets for moisture over Bhubaneswar.

4 Conclusions

Natural and anthropogenic aerosol loading in the eastern part of the Indian region including Bhubaneswar and Rourkela cities is an important aspect that motivates the study for examining how aerosols influence warm cloud microphysical properties. Nineteen years (2003–2021) of MODIS/Aqua level-2 retrieved daily aerosol, and cloud parameters are considered along with ERA5 reanalysis dataset for the study of aerosol-warm cloud interaction over the eastern Indian cities Bhubaneswar and Rourkela.

A strong competition between the cloud droplets, and aerosol particle for getting associated with water vapor is realized to be higher over Bhubaneswar as COD decreases with increase in CER for all AOD regimes. However, lower updraft and moisture availability lead to a negative CER-CTP relationship and, as a result, CER decrease and smaller CER due to lower buoyancy driven vertical motion. Nevertheless, increase in CER with CWP occurs due to growth of cloud droplets in the presence of a sufficient amount of water vapor.

The increase in CER with AOD over both the cities also indicate strong competition between the aerosol particles and cloud droplets for water vapor. However, CF was found to mostly decrease over Bhubaneswar, and increase over Rourkela with AOD in most of the considered scenarios. Also, COD was found to be mostly decreasing with increasing AOD in the presence of absorbing aerosols and thus, supports the formation of thinner clouds over both the cities due to dispersion of cloud droplets. However, decrease in CTP with increase in aerosol loading favors horizontal extension of cloud cover. Whereas, increase in CTP with AOD favors the vertical growth of clouds or development of deeper clouds. The growth of CER is a possible result of the strong competition for water vapor in polluted and heavily polluted cases. However, a reduction in COD with an increase in AOD is possibly due to radiative effects and thinning of clouds.

An atmosphere with higher RH can encourage the hygroscopic growth of aerosols. This can also help in further growth and activation of prevailing cloud droplets (Jones et al., 2009). A stable atmosphere can favor expanding of cloud cover horizontally whereas an unstable atmosphere can encourage vertical cloud development. However, dynamic effects such as upward motion are helping in the development of the higher and thicker clouds. On the other hand, cloud properties are mostly influenced by locally produced aerosols compared to the long-range transported ones (through the prevailing wind circulation). Besides, temperature advection may play an important role in governing the relationship between aerosol and warm cloud properties. The positive value of ACI indicates a decrease in CER with AOD (Twomey effect) and negative ACI implies an increase in CER with AOD (Anti-Twomey effect). The transition from Anti-Twomey to Twomey effect is evident in both cities, with the Twomey effect dominating as CWP exceeds 75 g⁻². Overall results suggest that heating by the aerosol loading increases stability at the lower troposphere resulting in turbulent heat flux response (sensible heat over land) increase, which leads to suppression of vertical mixing, and thus, result in enhancement of warm clouds over the Rourkela region. Conversely, an inverse relationship has been observed between warm cloud cover and aerosols because of the prevalence of a positive semi-direct effect over Bhubaneswar.

Caution is necessary for the analysis of the satellite-retrieved relationship between aerosol and cloud parameter data sets. Uncertainties of satellite-retrieved data sets are possibly due to the dynamic effect and influence of RH on aerosol properties (Yuan et al., 2008). Another possible reason could be assumptions of aerosol size distribution during the retrieval process, and improper cloud detection algorithms resulting in undetected contamination that leads to high AOD value (Liu et al., 2017; Sogacheva et al., 2017; Kant et al., 2019b, 2023). The aerosol and cloud parameter relationships are influenced by the size distribution of aerosols (Small et al., 2011). However, MODIS retrieved aerosol properties do not give any information about aerosol size. Therefore, it is better to examine seasonal ACI with different types and amounts of aerosol emissions.

Although all the urban areas have some common features that can modulate the aerosol-cloud interactions, the background environment or climates over different areas would have influences too. Since the current study could not accommodate this aspect, it would be interesting to isolate the distinct impacts of different climatic and environmental regimes, and industrial characteristics, on ACI over urban areas in future. Therefore, the conclusions drawn from the current study may be considered accordingly.

References

[1] Alam K, Khan R, Blaschke T, Mukhtiar A (2014). Variability of aerosol optical depth and their impact on cloud properties in Pakistan.J Atmos Sol Terr Phys, 107: 104–112

[2] Albrecht B A (1989). Aerosols, cloud microphysics, and fractional cloudiness.Science, 245(4923): 1227–1230

[3] Andersen H, Cermak J, Fuchs J, Knutti R, Lohmann U (2017). Understanding the drivers of marine liquid-water cloud occurrence and properties with global observations using neural networks.Atmos Chem Phys, 17(15): 9535–9546

[4] Bretherton C S, Blossey P N, Uchida J (2007). Cloud droplet sedimentation, entrainment efficiency, and subtropical stratocumulus albedo.Geophys Res Lett, 34(3): L03813

[5] Chen Y C, Christensen M W, Stephens G L, Seinfeld J H (2014). Satellite-based estimate of global aerosol–cloud radiative forcing by marine warm clouds.Nat Geosci, 7(9): 643–646

[6] Christensen M W, Chen Y C, Stephens G L (2016). Aerosol indirect effect dictated by liquid clouds.J Geophys Res Atmos, 121(24): 14636–14650

[7] Costantino L, Bréon F M (2013). Aerosol indirect effect on warm clouds over South-East Atlantic, from co-located MODIS and CALIPSO observations.Atmos Chem Phys, 13(1): 69–88

[8] Feingold G, McComiskey A, Rosenfeld D, Sorooshian A (2013). On the relationship between cloud contact time and precipitation susceptibility to aerosol.J Geophys Res Atmos, 118(18): 10544–10554

[9] Feingold G, Remer L A, Ramaprasad J, Kaufman Y J (2001). Analysis of smoke impact on clouds in Brazilian biomass burning regions: an extension of Twomey’s approach.J Geophys Res, 106(D19): 22907–22922

[10] Gao S, Lu C, Liu Y, Yum S S, Zhu J, Zhu L, Desai N, Ma Y, Wu S (2021). Comprehensive quantification of height dependence of entrainment mixing between stratiform cloud top and environment.Atmos Chem Phys, 21(14): 11225–11241

[11] Gao W, Sui C H, Hu Z (2014). A study of macrophysical and microphysical properties of warm clouds over the Northern Hemisphere using CloudSat/CALIPSO data.J Geophys Res Atmos, 119(6): 3268–3280

[12] Grandey B S, Gururaj A, Stier P, Wagner T M (2014). Rainfall-aerosol relationships explained by wet scavenging and humidity.Geophys Res Lett, 41(15): 5678–5684

[13] Grandey B S, Stier P (2010). A critical look at spatial scale choices in satellite-based aerosol indirect effect studies.Atmos Chem Phys, 10(23): 11459–11470

[14] Gryspeerdt E, Quaas J, Bellouin N (2016). Constraining the aerosol influence on cloud fraction.J Geophys Res Atmos, 121(7): 3566–3583

[15] Gryspeerdt E, Stier P, Partridge D G (2014). Satellite observations of cloud regime development: the role of aerosol processes.Atmos Chem Phys, 14(3): 1141–1158

[16] Hansen J, Sato M, Lacis A, Ruedy R (1997). The missing climate forcing.Philosophical Transactions of the Royal Society of London Series B: Biological Sciences, 352(1350): 231–240

[17] Ignatov A, Minnis P, Loeb N, Wielicki B, Miller W, Sun-Mack S, Tanré D, Remer L, Laszlo I, Geier E (2005). Two MODIS aerosol products over ocean on the Terra and Aqua CERES SSF datasets.J Atmos Sci, 62(4): 1008–1031

[18] Jeong M J, Li Z, Andrews E, Tsay S C (2007). Effect of aerosol humidification on the column aerosol optical thickness over the Atmospheric Radiation Measurement Southern Great Plains site.J Geophys Res, 112(D10): D10202

[19] Johnson B T, Shine K P, Forster P M (2004). The semi-direct aerosol effect: impact of absorbing aerosols on marine stratocumulus.Q J R Meteorol Soc, 130(599): 1407–1422

[20] Jones T A, Christopher S A, Quaas J (2009). A six year satellite-based assessment of the regional variations in aerosol indirect effects.Atmos Chem Phys, 9(12): 4091–4114

[21] Kang N, Kumar K R, Yin Y, Diao Y, Yu X (2015). Correlation analysis between AOD and cloud parameters to study their relationship over China using MODIS data (2003–2013): impact on cloud formation and climate change.Aerosol Air Qual Res, 15(3): 958–973

[22] Kant S, Panda J, Gautam R (2019b). A seasonal analysis of aerosol-cloud-radiation interaction over Indian region during 2000–2017.Atmos Environ, 201: 212–222

[23] Kant S, Panda J, Gautam R, Wang P K, Singh S P (2017). Significance of aerosols influencing weather and climate over Indian region.J Earth Atmosph Sci, 4(1): 1–20

[24] Kant S, Panda J, Manoj M G (2019c). A satellite observation-based analysis of aerosol-cloud-precipitation interaction during the February 2016 unseasonal heatwave episode over Indian region.Aerosol Air Qual Res, 19(7): 1508–1525

[25] Kant S, Panda J, Pani S K, Wang P K (2019a). Long-term study of aerosol-cloud-precipitation interaction over the eastern part of India using satellite observations during pre-monsoon seasons.Theor Appl Climatol, 136(1−2): 605–626

[26] Kant S, Sarangi C, Wilcox E M (2023). Aerosol processes perturb cloud trends over Bay of Bengal: observational evidence.NPJ Climate Atmosph Sci, 6: 132

[27] Kaufman Y J, Remer L A, Tanré D, Li R R, Kleidman R, Mattoo S, Levy R C, Eck T F, Holben B N, Ichoku C, Martins J V (2005). A critical examination of the residual cloud contamination and diurnal sampling effects on MODIS estimates of aerosol over ocean.IEEE Trans Geosci Remote Sens, 43(12): 2886–2897

[28] King M D, Menzel W P, Kaufman Y J, Tanré D, Gao B C, Platnick S, Ackerman S A, Remer L A, Pincus R, Hubanks P A (2003). Cloud and aerosol properties, precipitable water, and profiles of temperature and water vapor from MODIS.IEEE Trans Geosci Remote Sens, 41(2): 442–458

[29] Klein S A, Hartmann D L (1993). The seasonal cycle of low stratiform clouds.J Clim, 6(8): 1587–1606

[30] Koch D, Del Genio A D (2010). Black carbon semi-direct effects on cloud cover: review and synthesis.Atmos Chem Phys, 10(16): 7685–7696

[31] Köppen W, Volken E, Brönnimann S (2011). The thermal zones of the earth according to the duration of hot, moderate and cold periods and to the impact of heat on the organic world (Translated from: Die Wärmezonen der Erde, nach der Dauer der heissen, gemässigten und kalten Zeit und nach der Wirkung der Wärme auf die organische Welt betrachtet, Meteorol Z 1884, 1, 215–226).Meteorol Z (Berl), 20(3): 351–360

[32] Koren I, Feingold G, Remer L A (2010). The invigoration of deep convective clouds over the Atlantic: aerosol effect, meteorology or retrieval artifact.Atmos Chem Phys, 10(18): 8855–8872

[33] Koren I, Kaufman Y J, Rosenfeld D, Remer L A, Rudich Y (2005). Aerosol invigoration and restructuring of Atlantic convective clouds.Geophys Res Lett, 32(14): L14828

[34] Koren I, Martins J V, Remer L A, Afargan H (2008). Smoke invigoration versus inhibition of clouds over the Amazon.Science, 321(5891): 946–949

[35] Kulkarni J R, Maheskumar R S, Morwal S B, Kumari B P, Konwar M, Deshpande C G, Goswami B N (2012). The cloud aerosol interaction and precipitation enhancement experiment (CAIPEEX): overview and preliminary results. Current Sci, 102(3): 413–425

[36] Kumar S, Panda J, Paul D, Guha B K (2023). Impact of environmental variables on the North Indian Ocean tropical cyclones radial parameters.Climate Dynamics, 60(3−4): 813–830

[37] Li J, Yue Z, Lu C, Chen J, Wu X, Xu X, Luo S, Zhu L, Wu S, Wang F, He X (2022). Convective entrainment rate over the Tibetan Plateau and its adjacent regions in the boreal summer using SNPP-VIIRS.Remote Sens (Basel), 14(9): 2073

[38] Li Z, Zhao F, Liu J, Jiang M, Zhao C, Cribb M (2014). Opposite effects of absorbing aerosols on the retrievals of cloud optical depth from spaceborne and ground-based measurements.J Geophys Res Atmos, 119(9): 5104–5114

[39] Liu C, Song Y, Zhou G, Teng S, Li B, Xu N, Lu F, Zhang P (2023). A cloud optical and microphysical property product for the advanced geosynchronous radiation imager onboard China’s Fengyun-4 satellites: the first version.Atmosph Oceanic Sci Lett, 16(3): 100337

[40] Liu Y, de Leeuw G, Kerminen V M, Zhang J, Zhou P, Nie W, Qi X, Hong J, Wang Y, Ding A, Guo H (2017). Analysis of aerosol effects on warm clouds over the Yangtze River Delta from multi-sensor satellite observations.Atmos Chem Phys, 17(9): 5623–5641

[41] Liu Y, Huang J, Wang T, Li J, Yan H, He Y (2022). Aerosol-cloud interactions over the Tibetan Plateau: an overview.Earth-Sci Rev, 234: 104216

[42] Loeb N G, Manalo-Smith N (2005). Top-of-atmosphere direct radiative effect of aerosols over global oceans from merged CERES and MODIS observations.J Clim, 18(17): 3506–3526

[43] Lu C, Zhu L, Liu Y, Mei F, Fast J D, Pekour M S, Luo S, Xu X, He X, Li J, Gao S (2023). Observational study of relationships between entrainment rate, homogeneity of mixing, and cloud droplet relative dispersion.Atmos Res, 293: 106900

[44] Malavelle F F, Haywood J M, Jones A, Gettelman A, Clarisse L, Bauduin S, Allan R P, Karset I H H, Kristjánsson J E, Oreopoulos L, Cho N, Lee D, Bellouin N, Boucher O, Grosvenor D P, Carslaw K S, Dhomse S, Mann G W, Schmidt A, Coe H, Hartley M E, Dalvi M, Hill A A, Johnson B T, Johnson C E, Knight J R, O’Connor F M, Partridge D G, Stier P, Myhre G, Platnick S, Stephens G L, Takahashi H, Thordarson T (2017). Strong constraints on aerosol–cloud interactions from volcanic eruptions.Nature, 546(7659): 485–491

[45] Martins J V, Tanré D, Remer L, Kaufman Y, Mattoo S, Levy R (2002). MODIS cloud screening for remote sensing of aerosols over oceans using spatial variability.Geophys Res Lett, 29(12): 1619

[46] Medeiros B, Stevens B (2011). Revealing differences in GCM representations of low clouds.Clim Dyn, 36(1−2): 385–399

[47] Meyer K, Platnick S, Oreopoulos L, Lee D (2013). Estimating the direct radiative effect of absorbing aerosols overlying marine boundary layer clouds in the southeast Atlantic using MODIS and CALIOP.J Geophys Res Atmos, 118(10): 4801–4815

[48] Meyer K, Platnick S, Zhang Z (2015). Simultaneously inferring above-cloud absorbing aerosol optical thickness and underlying liquid phase cloud optical and microphysical properties using MODIS.J Geophys Res Atmos, 120(11): 5524–5547

[49] Minnis P, Ayers J K, Palikonda R, Phan D (2004). Contrails, cirrus trends, and climate.J Climate, 17(8): 1671–1685

[50] Moorthy K K, Suresh Babu S, Manoj M R, Satheesh S K (2013). Buildup of aerosols over the Indian Region.Geophys Res Lett, 40(5): 1011–1014

[51] Panda J, Kant S, Sarkar A (2023). A satellite-observation based study on responses of clouds to aerosols over South Asia during IOD events of south-west monsoon season.Atmos Pollut Res, 14(9): 101861

[52] Panicker A S, Pandithurai G, Dipu S (2010). Aerosol indirect effect during successive contrasting monsoon seasons over Indian subcontinent using MODIS data.Atmos Environ, 44(15): 1937–1943

[53] Patel P N, Kumar R (2016). Dust induced changes in ice cloud and cloud radiative forcing over a high altitude site.Aerosol Air Qual Res, 16(8): 1820–1831

[54] Paul D, Panda J, Routray A (2022). Ocean and atmospheric characteristics associated with the cyclogenesis and rapid intensification of NIO super cyclonic storms during 1981–2020.Nat Hazards, 114: 261–289

[55] Peng Y, Lohmann U, Leaitch R, Banic C, Couture M (2002). The cloud albedo-cloud droplet effective radius relationship for clean and polluted clouds from RACE and FIRE. ACE.J Geophys Res, 107(D11): 4106

[56] Platnick S, Meyer K G, King M D, Wind G, Amarasinghe N, Marchant B, Arnold G T, Zhang Z, Hubanks P A, Holz R E, Yang P (2017). The MODIS cloud optical and microphysical products: collection 6 updates and examples from Terra and Aqua.IEEE Trans Geosci Remote Sens, 55(1): 502–525

[57] Prasad A K, Singh R P, Singh A (2004). Variability of aerosol optical depth over Indian subcontinent using MODIS data.Photonirvachak (Dehra Dun), 32(4): 313–316

[58] Qiu Y, Zhao C, Guo J, Li J (2017). 8-Year ground-based observational analysis about the seasonal variation of the aerosol-cloud droplet effective radius relationship at SGP site.Atmos Environ, 164: 139–146

[59] Quaas J, Stevens B, Stier P, Lohmann U (2010). Interpreting the cloud cover–aerosol optical depth relationship found in satellite data using a general circulation model.Atmos Chem Phys, 10(13): 6129–6135

[60] Ramanathan V C P J, Crutzen P J, Kiehl J T, Rosenfeld D (2001). Aerosols, climate, and the hydrological cycle.Science, 294(5549): 2119–2124

[61] Remer L A, Kaufman Y J, Tanré D, Mattoo S, Chu D A, Martins J V, Li R R, Ichoku C, Levy R C, Kleidman R G, Eck T F, Vermote E, Holben B N (2005). The MODIS aerosol algorithm, products, and validation.J Atmos Sci, 62(4): 947–973

[62] Remer L A, Kleidman R G, Levy R C, Kaufman Y J, Tanré D, Mattoo S, Martins J V, Ichoku C, Koren I, Yu H, Holben B N (2008). Global aerosol climatology from the MODIS satellite sensors.J Geophys Res, 113(D14): D14S07

[63] Remer L A, Mattoo S, Levy R C, Munchak L A (2013). MODIS 3 km aerosol product: algorithm and global perspective.Atmos Meas Tech, 6(7): 1829–1844

[64] Rennó N O, Williams E, Rosenfeld D, Fischer D G, Fischer J, Kremic T, Agrawal A, Andreae M O, Bierbaum R, Blakeslee R, Boerner A, Bowles N, Christian H, Cox A, Dunion J, Horvath A, Huang X, Khain A, Kinne S, Lemos M C, Penner J E, Pöschl U, Quaas J, Seran E, Stevens B, Walati T, Wagner T (2013). CHASER: an innovative satellite mission concept to measure the effects of aerosols on clouds and climate.Bull Am Meteorol Soc, 94(5): 685–694

[65] Rosenfeld D, Gutman G (1994). Retrieving microphysical properties near the tops of potential rain clouds by multispectral analysis of AVHRR data.Atmos Res, 34(1−4): 259–283

[66] Saponaro G, Kolmonen P, Sogacheva L, Rodriguez E, Virtanen T, de Leeuw G (2017). Estimates of the aerosol indirect effect over the Baltic Sea region derived from 12 years of MODIS observations.Atmos Chem Phys, 17(4): 3133–3143

[67] Sarkar A, Amal K K, Sarkar T, Panda J, Paul D (2021). Variability in air-pollutants, aerosols, and associated meteorology over peninsular India and neighboring ocean regions during COVID-19 lockdown to unlock phases.Atmosph Pollut Res, 12(12): 101231

[68] Sarkar A, Panda J, Kant S, Mukherjee A (2022). Influence of smoke aerosols on low-level clouds over the Indian region during winter.Atmosph Res, 278: 106358

[69] Satheesh S K, Krishna Moorthy K, Suresh Babu S, Vinoj V, Nair V S, Naseema Beegum S, Dutt C B S, Alappattu D P, Kunhikrishnan P K (2009). Vertical structure and horizontal gradients of aerosol extinction coefficients over coastal India inferred from airborne lidar measurements during the Integrated Campaign for Aerosol, Gases and Radiation Budget (ICARB) field campaign.J Geophys Res: Atmosph, 114(D5): D05204

[70] Sengupta K, Dey S, Sarkar M (2013). Structural evolution of monsoon clouds in the Indian CTCZ.Geophys Res Lett, 40(19): 5295–5299

[71] Sharif F, Alam K, Afsar S (2015). Spatio-temporal distribution of aerosol and cloud properties over Sindh using MODIS satellite data and a HYSPLIT Model.Aerosol Air Qual Res, 15(2): 657–672

[72] Small J D, Chuang P Y, Feingold G, Jiang H (2009). Can aerosol decrease cloud lifetime.Geophys Res Lett, 36(16): L16806

[73] Small J D, Jiang J H, Su H, Zhai C (2011). Relationship between aerosol and cloud fraction over Australia.Geophys Res Lett, 38(23): L23802

[74] Sogacheva L, Kolmonen P, Virtanen T H, Rodriguez E, Saponaro G, De Leeuw G (2017). Post-processing to remove residual clouds from aerosol optical depth retrieved using the advanced along track scanning radiometer.Atmos Meas Tech, 10(2): 491–505

[75] Sorooshian A, Feingold G, Lebsock M D, Jiang H, Stephens G L (2009). On the precipitation susceptibility of clouds to aerosol perturbations.Geophys Res Lett, 36(13): L13803

[76] Sporre M K, Swietlicki E, Glantz P, Kulmala M (2014). A long-term satellite study of aerosol effects on convective clouds in Nordic background air.Atmos Chem Phys, 14(4): 2203–2217

[77] Stathopoulos S, Georgoulias A K, Kourtidis K (2017). Space-borne observations of aerosol-cloud relations for cloud systems of different heights.Atmos Res, 183: 191–201

[78] Stevens B, Feingold G (2009). Untangling aerosol effects on clouds and precipitation in a buffered system.Nature, 461(7264): 607–613

[79] Tang J, Wang P, Mickley L J, Xia X, Liao H, Yue X, Sun L, Xia J (2014). Positive relationship between liquid cloud droplet effective radius and aerosol optical depth over Eastern China from satellite data.Atmos Environ, 84: 244–253

[80] Thomas A, Sarangi C, Kanawade V P (2019). Recent increase in Winter Hazy Days over central India and the Arabian Sea.Sci Rep, 9(1): 17406

[81] Toll V, Christensen M, Gassó S, Bellouin N (2017). Volcano and ship tracks indicate excessive aerosol-induced cloud water increases in a climate model.Geophys Res Lett, 44(24): 12492–12500

[82] Toll V, Christensen M, Quaas J, Bellouin N (2019). Weak average liquid-cloud-water response to anthropogenic aerosols.Nature, 572(7767): 51–55

[83] Tripathi S N, Dey S, Tare V, Satheesh S K (2005). Aerosol black carbon radiative forcing at an industrial city in northern India.Geophys res lett, 32(8): L08802

[84] Twomey S (1977). The influence of pollution on the shortwave albedo of clouds.J Atmos Sci, 34(7): 1149–1152

[85] Wang F, Guo J, Wu Y, Zhang X, Deng M, Li X, Zhang J, Zhao J (2014). Satellite observed aerosol-induced variability in warm cloud properties under different meteorological conditions over eastern China.Atmos Environ, 84: 122–132

[86] Wang F, Guo J, Zhang J, Huang J, Min M, Chen T, Liu H, Deng M, Li X (2015). Multi-sensor quantification of aerosol-induced variability in warm clouds over eastern China.Atmos Environ, 113: 1–9

[87] Wilcox E M (2010). Stratocumulus cloud thickening beneath layers of absorbing smoke aerosol.Atmos Chem Phys, 10(23): 11769–11777

[88] Wilcox E M, Thomas R M, Praveen P S, Pistone K, Bender F A M, Ramanathan V (2016). Black carbon solar absorption suppresses turbulence in the atmospheric boundary layer.Proc Natl Acad Sci USA, 113(42): 11794–11799

[89] Wright M E, Atkinson D B, Ziemba L, Griffin R, Hiranuma N, Brooks S, Lefer B, Flynn J, Perna R, Rappenglück B, Luke W, Kelley P (2010). Extensive aerosol optical properties and aerosol mass related measurements during TRAMP/TexAQS 2006–Implications for PM compliance and planning.Atmos Environ, 44(33): 4035–4044

[90] Yuan T, Li Z, Zhang R, Fan J (2008). Increase of cloud droplet size with aerosol optical depth: an observation and modeling study.J Geophys Res, 113(D4): D04201

[91] Zhang J, Reid J S, Holben B N (2005). An analysis of potential cloud artifacts in MODIS over ocean aerosol optical thickness products.Geophys Res Lett, 32(15): L15803

[92] Zhao J, Ma X, Quaas J, Jia H (2023). Exploring aerosol-cloud interactions over eastern China and its adjacent ocean using the WRF-SBM-MOSAIC model.EGUsphere, 2023: 1–27

RIGHTS & PERMISSIONS

Higher Education Press