Aerosol-Cloud-Precipitation Interactions over Major Cities in South Africa : Impact on Regional Environment and Climate Change

In this study, we have used the Terra satellite onboard of the Moderate Resolution Imaging Spectroradiometer (MODIS) to investigate the spatial and temporal relationship between aerosol optical depth (AOD) and cloud parameters namely, water vapor (WV), cloud optical depth (COD), cloud fraction (CF), cloud effective radius (CER), cloud top pressure (CTP), and cloud top temperature (CTT) based on 10 years (from January 2004 to December 2013) of dataset over six locations in South Africa (SA). The obtained results indicated seasonal variation in AOD, with high values during spring (September to November) and low values in winter (June to August) in all locations of study. In terms of temporal variation, AOD was lowest at Bloemfontein 0.06 ± 0.04 followed by Cape Town 0.08 ± 0.02, then Potchefstroom 0.09 ± 0.05, Pretoria and Skukuza had 0.11 ± 0.05 each and with the highest at Durban 0.13 ± 0.05. The mean Angstrom exponent (AE) values for each location showed a general prevalence of fine-mode particles which dominates the AOD for most parts of the year. A hybrid single particle Lagrangian integrated trajectory (HYSPLIT) model was used for trajectory analysis in order to determine the origin of airmasses and to understand the variability of AOD. We then studied the relationship between AOD, cloud parameters and precipitation over selected locations of SA so as to provide a better understanding of aerosol-cloud-precipitation interactions. All these correlations examined over six sites were observed to be depended on the large-scale meteorological variations.


SM-1. Validation reports of MODIS with ground-based data
The spatio-temporal data analysis carried out between satellite-and ground-based instruments help both in identifying the uncertainties of these retrievals and their local spatial behavior (Ichoku et al., 2002).Previous studies have also used AERONET measured AODs to validate MODIS derived AODs.Chu et al. (2002) and Diner et al. (2001) performed validation over SA with a limited data for short period from July to September of 2000.Prasad et al. (2007) found a good correlation (R 2 =0.47) between MODIS and AERONET over Kanpur during winter but a poor correlation (R 2 =0.29) in summer.Gupta et al. (2013) compared MODIS derived AODs with AERONET derived AODs over Lahore and found a good correlation between the two (R 2 =0.72).Recently, Bibi et al. (2015) reported over all good agreement between MODIS and AERONET AODs with correlation coefficients (R 2 ) of 0.71, 0.67, 0.76, and 0.61 over Karachi, Lahore, Jaipur, and Kanpur, respectively.

SM-2. Relationship between AOD and WV
In order to understand the aerosol impact and the process of the hydrological cycle, we investigated the change in column water vapour in relation to aerosols (Myhre et al., 2007).
There are five near-infrared bands in MODIS just around water vapor band (940 nm) for remote sensing of clear sky column water vapour amount.It is so designed to observe water vapor absorption by solar radiation reflected by bottom surface at near-infrared.Through the use of the ratios of water absorbing bands with atmospheric window bands, the variation of surface reflectance with wavelength for most land surfaces is being removed (King et al., 1992).The The temporal correlation (Fig.  between AOD 550 and WV shows that both quantities only co-vary at the beginning of the year but they tend to have opposite trend in the rest of the year over all the locations.The correlation coefficients between for the above relationship observed to -0.463, 0.254, -0.464, 0.123, 0.461 and 0.054 for the locations Bloemfontein, Cape Town, Durban, Potchefstroom, Pretoria and Skukuza, respectively.It is evident that Pretoria has the highest correlation coefficient which agrees well with the time series plot indicates that summer months has higher correspondence between AOD 550 and WV (Ranjan et al., 2007).It may be also noted that aerosol influence the cloud formation over Pretoria as the location experiences higher rainfall during that season (Adesina et al., 2014).Aerosol may not be playing a significant role in cloud formation over Bloemfontein, Durban and Skukuza.
The water absorbing ability of aerosols (hygroscopic nature) depends upon the particular mixing of different types of aerosols particles as well as on the meteorological conditions (Kaufman et al., 2005).Lee et al. (2009) suggested that water uptake of atmospheric aerosols can alter both the size and composition of particles and hence their optical particles.Wright et al. (2010) and Xie et al. (2011) depicted that the radiative forcing on climate significantly influenced by the changes in water uptake behavior of aerosols and hence cloud formation.Guo et al. (2014) and Luo et al. (2013) concluded that natural and anthropogenic aerosols over China play an important role in influencing the convective cloud formation and hence causing climatic implications to the overall hydrological cycle.

SM-3. Relationship between AOD and CTP
The spatial correlation between AOD and CTP presented in Fig. SM-2a shows that Durban, Potchefstroom and Bloemfontein have positive correlation, while the other locations observed to have negative correlation.CTP tends to have higher negative correlation at higher latitude as reported by some studies (Kaufman et al., 2005;Myhre et al., 2007;Sekiguchi et al., 2009) as in the case of Pretoria and Skukuza.These correlation facts may be caused by the largescale meteorological variations.Earlier researchers have reported that except for some regions of low AOD, CTP decreased in most of the areas (higher cloud altitude) as AOD increased (Alam et al., 2014 and references therein).This might have resulted from the suppression of the precipitation by increasing cloud lifetime and thus also affecting the cloud albedo and changing the CTP (Kaufman et al., 2005).Further, Ramanathan et al. (2001) and Lee et al. (2009) suggested since the CER increases with decrease of CTP and thereby, decreasing the CTP with increasing AOD.In the time series plot for the spatial averages (Fig. SM-2b), we can see same pattern of CTP increasing to a maximum value from January to July in all the stations, except Cape Town, where it first decreased to a minimum value in May then increased to July before it started to decrease at the end of the year.The correlation coefficients are observed to be Bloemfontein (0.260), Cape Town (0.182), Durban (0.203), Potchefstroom (-0.225),Pretoria (-0.452), and Skukuza (-0.327).
spatial correlation of AOD 550 with WV during clear sky is shown in Fig. SM-1a.It is found to be positive correlation (0.1-0.2) over Skukuza and Pretoria and negative correlation (>-0.3) over rest of the other locations.
Fig. SM-1.(a) Spatial correlation map and (b) time series plots with standard deviation between AOD and WV over six locations in SA during 2004-2013.