Science of the Total Environment 527–528 (2015) 507–519
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Columnar-integrated aerosol optical properties and classification of different aerosol types over the semi-arid region, Anantapur, Andhra Pradesh Rama Gopal K a,⁎, Arafath S.Md a, Balakrishnaiah G a, Raja Obul Reddy K a, Siva Kumar Reddy N a, Lingaswamy A.P a, Pavan Kumari S a, Uma Devi K a, Reddy R.R a, Suresh Babu S b a b
Aerosol & Atmospheric Research Laboratory, Department of Physics, Sri Krishnadevaraya University, Anantapur, 515 003 Andhra Pradesh, India Space Physics Laboratory, Vikram Sarabhai Space Centre, Trivandrum, 695 022 Kerala, India
H I G H L I G H T S • • • • •
AOD shows higher values during summer and lower values during monsoon. Minimum (maximum) values of a was observed during summer (winter). a2b0 (a2>0) is a characteristic of fine-mode (coarse-mode) aerosols domination. The frequency distribution of α indicates the mixture of fine and coarse particles. The results identified four aerosol types of differing origin, compositions.
a r t i c l e
i n f o
Article history: Received 11 June 2014 Received in revised form 20 April 2015 Accepted 23 April 2015 Available online 22 May 2015 Editor: D. Barcelo Keywords: Aerosol optical depth Angstrom exponent Aerosol classification
a b s t r a c t This study presents a characterization of aerosol columnar properties measured at a semi-arid station Anantapur in the southern part of India during the period from October 2012 to September 2013. Aerosol optical depth (AOD) and Angstrom exponent (α) have been retrieved from Microtops II Sunphotometer over the observation site. The results show that a pronounced spectral and monthly variability in the optical properties of aerosols is mainly due to anthropogenic sources. The results show that the spectral curvature can effectively be used as a tool for aerosol type discrimination, since the fine-mode aerosols exhibit negative curvature, while the coarsemode particles are positive. The classification of aerosols is also proposed by using the values of AOD at 500 nm and Angstrom exponent values (α(380–870)) by applying threshold values obtained from the frequency distribution of AOD. The results of the analysis were identified by four individual components (anthropogenic/ biomass burning, coarse/dust, coarse/marine, clean continental) of different origin and compositions. The most frequent situations observed over the site are that due to the anthropogenic/biomass burning situations which account for about 45.37%, followed by coarse/dust (43.64%), clean continental (7.2%) and coarse/marine (3.82%) during summer. The identification of the aerosol source type and the modification processes are analyzed by using the Gobbi et al. (2007) classification scheme based on the measured scattering properties (α, dα) derived from the Microtops II Sunphotometer. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Atmospheric aerosols are solid or liquid particles suspended in the atmosphere. Aerosols have long been identified as the major controllers of the Earth's climate (IPCC, 2001). Despite the progress achieved during the last decades in understanding the effects of aerosols on climate, their large spatial–temporal variability and heterogeneity still cause
⁎ Corresponding author. E-mail address:
[email protected] (R.G. K).
http://dx.doi.org/10.1016/j.scitotenv.2015.04.086 0048-9697/© 2015 Elsevier B.V. All rights reserved.
significant uncertainties at global scales (IPCC, 2007). Atmospheric aerosol is a complex dynamic system with temporal variability caused by the nature of aerosol sources, synoptic conditions (air mass changes) and the meteorological factors (humidity, precipitation). Anthropogenic and natural aerosols are important atmospheric constituents that significantly contribute to Earth's radiation budget through a variety of pathways such as direct effects on scattering and absorption of solar radiation, indirect effects on cloud microphysics (modify cloud amount and life time), and semi-direct effects (Kaufman et al., 2002; Yoon et al., 2005; Ramanathan et al., 2007; Ramanathan and Carmichael, 2008). They scatter (angular redistribution of energy) and absorb
508
R.G. K et al. / Science of the Total Environment 527–528 (2015) 507–519
(conversion of energy into either heat or photochemical change) optical radiation depending upon their size distribution, refractive index and total atmospheric loading. This results in attenuation or extinction of solar radiation reaching the earth's surface (Ranjan et al., 2007). On a global scale, the natural sources of aerosols are more important than the anthropogenic aerosols, but regionally anthropogenic aerosols are more important (Ramanathan et al., 2001; Satheesh and Moorthy, 2005). However, as a consequence of their high spatial and temporal variability, these effects are strongly regional in magnitude and sign (Nakajima et al., 2003; El-Metwally et al., 2011). Aerosol optical depth is an important parameter that characterizes the integrated extinction of solar radiation suffered in its transit through the atmosphere. Also many studies concerning atmospheric turbidity and aerosol optical properties in different parts of the world have been performed (Rapti, 2000; Hand et al., 2004; Zhao et al., 2013). In order to improve our scientific knowledge of the climatic role of aerosols for the semi-arid region, and the accuracy of radiation budget model results, more and better data on aerosol properties and loads are required. This study is intended to consider and analyze the aerosol optical properties for the period October 2012 to September 2013 over the semi-arid station, Anantapur in Southern India. The spectral and monthly variables in the optical properties of aerosols are presented. The temporal variation of Angstrom exponent and turbidity factor and also discrimination of aerosol types are studied. The results of groundbased observation would provide insights into the relationship between aerosol optical properties and aerosol transport mechanism in the atmosphere. The identification of the aerosol source type and the modification processes are analyzed by using the Gobbi et al. (2007) classification scheme based on the measured scattering properties of aerosols. It is also expected that the results obtained from the present study will constitute the base for further aerosol research over the measurement region. 2. Site description, instrumentation and methodology Collocated measurements of aerosol optical depth (AOD) were carried out at the Department of Physics, Sri Krishnadevaraya University (SKU, 14.62° N, 77.65° E, 331 m asl), situated at the outskirts south of Anantapur town. Anantapur represents a very dry interior region of Andhra Pradesh, India. It is geographically situated on the boundary of a semi-arid and rain shadow region. Within a 50 km radius, the region has a number of anthropogenic sources such as, cement plants, limekilns and slab polishing units, brick making and other small industries and besides these industries, the national highways (NH 7 and NH 205) pass the town area. The climate is hot and dry in the summer (March–May), hot and humid during the monsoon and post monsoon (June–November) period and dry during the winter (December–February). The normal rainfall during the southwest monsoon period is 350 mm, which is more than 60–70% of the total rainfall, whereas in the northeast monsoon period it is only ~ 150 mm, which forms just 40–30% of annual rainfall. The measurements of AOD were carried out using a sunphotometer, the Microtops II of the Solar Light Company Inc., USA. The sunphotometer measures the intensity of direct solar irradiance at five narrow band spectral channels centered at 380, 500, 870, 936 and 1020 nm. The sunphotometer works on the principle of measuring solar radiation intensity at a specified region. Complete details of the instrument and measurement techniques can be found in the User Guide, Microtops II Sunphotometer version 5.6, Solar Light Company Inc. The instrument is calibrated (Solar Light Company) by Langley plot method or by simultaneous comparison with a reference sunphotometer that was calibrated at Mauna Loa Observatory (http://www.solarlight. com). The filters used in all channels have a peak wavelength precession of ±1.5 nm, and a full width at half maximum (FWHM) bandwidth of 2.4 ± 0.4 nm for the 380 nm channel and 10 ± 1.5 nm for the other channels. The full angular field of view is 2.5°. The front quartz window
of the instrument was cleaned regularly for keeping the accuracy of data. The aerosol optical depth (AOD) calculation algorithms are programmed in the Microtops II and the final results of all stored scans can be conveniently viewed on the LCD. The raw data is also stored to allow retrospective adjustments of algorithms. The raw data collected by the Microtops II, as well as calculated results are stored in nonvolatile memory. And each data point is annotated with date, time, site coordinates, solar angle, altitude, pressure and temperature. The spectral dependence of AOD is used in this work to compute the Angstrom exponent (α). The Angstrom parameters were obtained from each data point at 15 min interval AODs (440–870 nm) by using Angstrom power law. Angstrom suggested an empirical formula for the attenuation of scattering and absorption by aerosols. According to his formula, the aerosol optical depth, AODλ, is related to wavelength (λ in μm) through Angstrom's equation, given by, AODλ ¼ βλ−α
ð1Þ
where AODλ is the aerosol optical depth at the wavelength λ, β is the Angstrom's turbidity coefficient which equals AOD at 1 μm wavelength; and α is the Angstrom exponent that depends on the size distribution parameter of aerosol. The Angstrom turbidity parameters give valuable information of aerosol properties and important aerosol types. These parameters are dimensionless measures of the opacity of a vertical column of the atmosphere and a numerical quantification of the column extinction of transmitted radiation by atmospheric aerosols from a fixed band in the solar electromagnetic spectrum. While classifying the aerosol type observations, the error that occurred with a difference in wavelength dependence, is recovered from linear interpolation technique to calculate the AOD at an unknown wavelength (440 and 675 nm) using power law (Prasad et al., 2007) as follows, AODλ1 ¼ AODλ2
λ1 λ2
−α
ð2Þ
where α is the Angstrom exponent, determined from spectral aerosol optical depth measured by a Microtops II Sunphotometer as shown below. The wavelength dependence of AODλ which can be characterized by the Angstrom parameter, is a coefficient of the following regression. ln AODλ ¼ −α ln λ þ lnβ
ð3Þ
The value of α is approximately 1.3 for average normal size distribution. Value higher than 1.3 is a result of the relatively higher frequency of smaller particles as compared with the large particles having a radius greater than 0.5 μm (Tóth, 2013). The variation in α with wavelength is related to the empirical relationship between aerosol extinction and wavelength, when it is simulated by a second order polynomial fit (Pedros et al., 2003; Kaskaoutis and Kambezidis, 2006) ln AOD ¼ a2 ln λ2 þ a1 ln λ þ a0
ð4Þ
where the coefficient a2 accounts for a curvature often observed in sunphotometry measurements. The curvature can be an indicator of the aerosol particle size, with negative sign indicating the aerosol size distributions dominated by fine mode and the positive curvature is indicated by the size distributions with significant contribution by coarse mode aerosols (Schuster et al., 2006; Balakrishnaiah et al., 2011). In addition, the coefficients a1 and a2 are obtained from the second order polynomial fit to the AOD values at four channels (380, 500, 870 and 936 nm). Choosing the selected range of wavelengths relies on the fact that these are highly accurate channels of the sunphotometer. AOD at 1020 nm values was omitted from the fits because it contains larger uncertainties due to the water vapor effect.
R.G. K et al. / Science of the Total Environment 527–528 (2015) 507–519
509
30
I N D I A
25
20
Anantapur
15
ARABIAN OCEAN
BAY OF BENGAL
10
5
INDIAN OCEAN
0 60
65
70
75
80
85
90
95
100
Fig. 1. Schematic site map of the sampling site Sri Krishnadevaraya University (SKU) showing the details mentioned in the text. Also shown is the station (Anantapur) over the India peninsula, in the inset. Satellite aerial view of monitoring site building in the SKU campus indicated with an arrow head.
For the classification of aerosol types, AOD and alpha values have to be used (Holben et al., 2001). In the present study, the contour maps were constructed using 0.3 and 1.0 steps for both AOD500 and α380–870 values, respectively. The threshold values are basically deduced by a statistical analysis of the frequency distribution of those quantities. In these maps, the rectangle areas denote different aerosol types namely, coarse/ dust (CD), anthropogenic/burning (AB), clean continental (CC) and coarse/marine (CM) aerosols, and the boundaries of these areas correspond to the selected threshold values of AOD500 and α380–870. Therefore, (1) values of AOD500 b 0.3 with α380–870 N 1 represent clean continental (CC) aerosols, (2) AOD500 N 0.3 with α380–870 N 1 can be used to characterize anthropogenic biomass burning (AB) aerosols, (3) AOD500 b 0.3 associated with α380–870 b 1 is indicative of coarse/marine (CM) aerosols and (4) AOD500 N 0.3 with α380–870 b 1 is considered as coarse/dust (CD) particles. This method has been used in a number of studies (Elias et al., 2006; Kalapureddy and Devara, 2008; Kaskaoutis
et al., 2009; Obergon et al., 2012) and is based on the sensitivity of the two parameters to different, somewhat independent, microphysical aerosol properties. The Angstrom exponent depends on the particle size distribution while AOD500 depends mainly on the aerosol column density (Fig. 1). To analyze the aerosol modification processes, we have followed the Gobbi et al. (2007) proposed scheme to visualize the contribution of fine aerosols to AOD, and this has revealed the modal radius of the fine particles (Rf). We defined the Angstrom exponent difference dα (α(440–675) − α (675–870)) as a measure of the Angstrom exponent curvature dα / dτ. In these coordinates, we have further classified aerosols by representing their AOT by different colors. To interpret the data in these coordinates, we determined reference points corresponding to bimodal size distributions characterized by a variety of fine mode (Rf) and coarse mode (Rc) modal radius combined to lead to prescribed fractions (η) of the fine mode to total AOD (at 500 nm).
510
R.G. K et al. / Science of the Total Environment 527–528 (2015) 507–519
0.9
380 936
0.8
500 1020
3.2. Monthly variation of AOD and alpha
870
0.7
AOD
0.6 0.5 0.4 0.3 0.2 0.1 8
10
12
14
16
Local time (Hrs) 0.50
Angstrom ecponent (α)
0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 8
10
12
14
16
18
Local time (Hrs) Fig. 2. Diurnal variation of aerosol optical depth at different wavelengths and Angstrom exponent α(380–1020) during the study period. The vertical bars indicate ±1 standard deviation of the mean for the period.
3. Results and discussion 3.1. Diurnal and monthly variation of AOD and alpha Fig. 2 shows the diurnal variation of average aerosol optical depth (AOD) at all wavelengths and Angstrom exponent (α) during the study period over the measurement site. The morning values are generally higher than those of afternoon because of capping action of aerosols released due to various anthropogenic activities by the previous night. By noon or shortly after, the radiative inversion normally starts to break up under the action of solar heating. In general, this type of behavior of AOD has been observed on most of the days during the study period. Devara et al. (1996) have observed similar variation in diurnal pattern of AOD at Pune. The daily variation of the AOD lies in the range of 0.15 to 0.71 over the Anantapur region. The values of α undergo a change throughout the day because of changes in the atmospheric meteorological parameters and anthropogenic activities. These changes may either decrease or increase in the values of the Angstrom parameters. In general, everyday observations show that an increase in α during the forenoon was followed by a slight decrease throughout day. The dominance of small particles (during morning hours) coupled with high values of α point indicates the possibility of the aerosol loading mainly from the vehicular emissions coming from nearby areas (Rama Gopal et al., 2014).
The monthly mean variations of aerosol optical depths for the entire period over the site are shown in Fig. 3 for all wavelengths. The figure reveals that the AOD500 has attained peak (0.58 ± 0.1) in the month of May while dip (0.3 ± 0.12) is seen in September. The maximum AOD values were observed from March to May. A possible explanation for high AOD during the summer is due to the forest fires in central and northern India and transport of biomass burning plumes or dust transport by northerly winds (Badarinath et al., 2007, 2010). Coarse particle presence in the summer attributed to the lift of the loose soil by high wind speed of that region. Low AOD values obtained from July to September can be attributed to the effects of strong wet removal process of aerosols (Kumar et al., 2009; Balakrishnaiah et al., 2011). Furthermore, drastic reduction in anthropogenic activities like biomass burning and brick making is due to frequent rains in the site of study. In the post-monsoon season, the aerosol production mechanism related to wind and surface conditions is weak. During December through February, AOD is slightly lower as it is mostly associated with fine-mode particles. The variations in meteorological parameters are the characteristics of the environmental changes in the months/seasons of a year. It is, therefore expected that the atmospheric aerosols be subjected to changes in the number and size distribution. The α is an index for the aerosol size distribution and it depends on the ratio of concentration of small and large aerosols, whereas, α decrease is due to the increasing AOD values and vice-versa. In sharp contrast, the α values show a pronounced decrease in the monsoon months. This strong decrease in α value may partly be attributed to cloud contamination in the retrievals caused by thin undetected cirrus clouds. Minimum values of α in the monsoon indicate the presence of coarse-mode aerosol particles during this period. A high value of α during the summer is the indication of the presence and contribution of fine-mode particles. A possible explanation for this could be the fact that this is the time with peaks in forest fires in central and northern India (Badarinath et al., 2007) and transport of biomass burning plumes by northerly winds (Badarinath et al., 2009). 3.3. Angstrom exponent (α) versus turbidity factor (β) In order to classify the variation of fine and coarse mode particles, changes in the Angstrom parameters are employed. In Fig. 4 (in monthly scale), it is seen that there is an anti-correlation between α and β indicating the continuous redistribution of fine and coarse particles under the influence of meteorological parameters. This shows an inverse relationship between α and β values. Such a trend is found in many other observational sites in India and elsewhere (Xin et al., 2007; Vijaykumar et al., 2012). The α values are found to be maximum in summer and minimum in the monsoon season. The high value of α in the summer is attributed to the domination of fine mode aerosols over coarse mode aerosols. And the greenery which is abundant at the observation site undergoes photosynthetic action due to sun shine which is the reason for the production of small size aerosol (during the summer). The higher value of β (and smaller value of α) during the monsoon and post monsoon signifies higher relative abundance of coarse (super micron) aerosols in the atmosphere. Normally β increases with the increasing AOD values and viceversa, whereas, α decreases with is due to the increase of AOD values and vice-versa. 3.4. Spectral variation of AOD Fig. 5 indicates the spectral variation of AOD for different seasons at the measurement site. It is evident from the figure that there is relatively strong wavelength dependence of optical depth at shorter wavelengths that gradually decrease towards longer wavelengths
R.G. K et al. / Science of the Total Environment 527–528 (2015) 507–519
0.9
0.7
380 nm
0.8
511
500 nm
0.6
0.7 0.5
AOD
AOD
0.6 0.5
0.4
0.4
0.3
0.3 0.2
0.2
0.30
0.30
0.25
0.25
2013
2012
Sep
Aug
Jul
Jun
May
Apr
Sep
Aug
Jul
Jun
May
Apr
Month
Sep
Aug
Jul
Jun
May
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
0.0
Jan
0.05
Apr
0.4 0.2
Dec
Mar
0.6
0.10
Month
Mar
0.8
Mar
0.15
1.0
Feb
0.20
1.2
Jan
0.25
1.4
Dec
Angstrom expoenent (α)
0.30
Nov
Alpha
1.6
Nov
1020 nm
2013
Month
2012
0.35
Feb
2013
Month
Jan
Oct
Sep
Jul
Aug
Jun
May
Dec
2012
Apr
0.05
Mar
0.05
Feb
0.10
Jan
0.10
Nov
0.15
Oct
0.15
Dec
0.20
Nov
0.20
Oct
Feb
936 nm 0.35
AOD
2013
Month
0.40
0.35
2012
Jan
2012
870 nm
0.40
Dec
Oct
Sep
Jul
Aug
Jun
May
Mar
Apr
2013
Month
AOD
AOD
0.40
Feb
Jan
Dec
Oct
Nov 2012
Nov
0.1
0.1
2013
Fig. 3. Monthly mean aerosol optical depth (at different wavelengths) and Angstrom exponent at Anantapur.
irrespective of the seasonal change attributing it to the presence of fine to coarse particles (the curve is steeper during the summer and relatively less steeper during the monsoon season). The presence of a high concentration of the fine-mode particles selectively enhances the irradiance scattering at lower wavelength and therefore, the AOD values are high at the shorter wavelengths. Likewise, the coarse-mode particles provide similar contributions to the AOD at relatively larger wavelengths (Schuster et al., 2006). It is noticed that the magnitude of
mean AOD is higher at longer wavelengths throughout the monsoon months, which indicates the dominance of coarse mode aerosols due to the condensation growth and coagulation mechanism of submicron aerosols, which are more efficient in producing larger aerosols. This type of spectral AOD is consistent with the Mie scattering theory for aerosol particles. There are no significant changes found for the near infrared channels at 870, 936 and 1020 nm. This could also be interpreted that the regional pollutants have high concentrations of finer particles.
R.G. K et al. / Science of the Total Environment 527–528 (2015) 507–519
Alpha
1.6
Beta
700
0.35
600
1.2
0.30
1.0
0.25
0.8
0.20
0.6 0.15 0.4
400 300 200 100
Sep
Jul
Aug
500
0 0.6-0.7
0.5-0.6
0.4-0.5
0.3-0.4
0.2-0.3
Fig. 4. Temporal variation of Angstrom exponent and turbidity factor from October 2012 to September 2013.
0.1-0.2
2013
