Atmospheric turbidity over a continental station Mysore, India
K E Ganesh1,$,*, T K Umesh2 & B Narasimhamurthy2
1PES Institute of Technology, Department of S&H, BSK III Stage, Bengaluru 560 085, India
2Department of Studies in Physics, University of Mysore, Manasagangotri, Mysore 570 006, India
$E-mail: ganeshke@gmail.com, drganeshke@gmail.com
Received 26 July 2010; revised 15 February 2011; accepted 21 February 2011
The solar radiation is collected at five optical channels at a continental station Mysore, India during December 2003 - June 2006 using a portable sunphotometer MICROTOPS II. Angstrom’s turbidity parameters, α and β, have been calculated and analysed on daily, monthly, seasonal and annual basis. The observations show that α and β vary throughout on individual day because of changes in the atmospheric meteorological parameters. It is observed that β is highest during summer and lowest during winter. An anti-correlation between α and β is observed throughout the day during all seasons indicating continuous redistribution of fine and coarse particles under the influence of meteorological parameters. The yearly comparison of α and β along with the atmospheric visibility is also analysed and presented in the paper.
Keywords: Atmospheric aerosol, Atmospheric water vapour, Aerosol optical thickness, Atmospheric turbidity PACS No.: 92.60.Mt
1 Introduction
The solar radiation gets attenuated while passing through the earth‘s atmosphere. The radiation extinction is mainly due to scattering by air molecules and aerosol particles. On clear sky days, aerosols are the atmospheric constituents that dominate the attenuation of solar radiation at visible and near- infrared wavelengths1. The scattering of solar radiation by matter other than dry air molecules is referred as turbidity of the atmosphere2. The atmospheric turbidity is a measure of total vertically integrated particulate load in the atmosphere.
Turbidity is a dimensionless quantity, which gives the measure of opacity of a vertical column of the atmosphere3. Atmospheric turbidity is an important factor influencing the energetic of solar radiation in the earth’s atmosphere. It is a useful index of atmospheric pollution particularly in studies of long- term changes in the composition of the atmosphere and the resultant global climatic changes. The seasonal variation of Angstroms parameters with atmospheric water vapour and temperature along with the causes for higher values of β during summer and lower values during other seasons have been discussed in the present study Despite of its importance, there is no universally adopted definition of turbidity or universally accepted technique for its
measurement. Some indices of turbidity were proposed and several methods were developed to determine their values4-7. However, Angstrom’s turbidity parameters have been widely adopted4.
Mysore (12°19ʹN and 76°39ʹE), a low latitude station, is situated at an altitude of 767 m above the sea level on the Deccan plateau of peninsular India.
About 300-500 km away in the East, West and South of Mysore, there is water spreads of Bay of Bengal, Arabian Sea and Indian Ocean, respectively. There is the landmass of Asiatic continent in the North. The geographic climate of Mysore is moderate. The monsoon rains, accounting for 73% of the average annual rainfall of 760 mm, are received during June – November every year. Following the rainy season, the winter prevails during December - February8. During winter, the temperature is low and the rainfall is about 3%. An overall temperature lies in the range of 18- 36°C throughout the year. The hot months during March - May constitute the summer and account for 24% rainfall9.
2 Methodology
2.1 Data collection
In order to collect solar radiation, a hand-held multi-band sunphotometer MICROTOPS II developed by Solar Light Company, USA10 has been
used. The instrument is equipped with five accurately aligned optical collimators with a full field view of 2.5°. The internal baffles are also integrated into the device to eliminate internal reflections. Each channel is fitted with a narrow band interference filter and a photodiode suitable for the particular wavelength range. The MICROTOPS II, used in the present study, has optical filters transmitting the radiation centered at wavelengths of 440, 500, 675, 936 and 1020 nm.
Aerosol optical thickness (AOT) was determined by Langley method assuming the validity of the Bouguer-Lambert-Beer law. The optical depth due to Rayleigh scattering has been subtracted from the total optical depth to obtain AOT. Initially, several MICROTOPS II settings have been made with the help of GPS receiver, which include universal date and time, geographic coordinates, altitude and atmospheric pressure of the measurement site.
For the observations, the MICROTOPS II was mounted on a tripod in order to minimize the Sun targeting error and was stationed on the open terrace of a two-storey building located in an unpolluted area.
The measurements have been made on the days of clear sky, viz. when there are no clouds. The data has been collected during 04:30:00 - 11:00:00 hrs UT at 15 min interval.
2.2 Calibration
The instrument was calibrated at regular intervals.
The degradation of the filters or the drift in the calibration values has been found to be marginal.
Using the standard Langley technique, the instrument was calibrated atop Sri Chamundeshwari Hills, which is at about 300 m from the ground level. The calibration constants obtained from the data collected atop the hill did not show large variations from the values obtained from the calibrations at factory11.
2.3 Angstrom turbidity formula
Angstrom suggested an empirical formula by considering the attenuation effects of scattering and absorption by aerosols. According to his formulation, the aerosol optical thickness (τpλ) is related to wavelength (λ) through the equation
τpλ = βλ-α … (1)
where, α and β, are known as Angstrom parameters.
The index, α, is the wavelength exponent; and β, turbidity coefficient representing the amount of aerosols present in the atmosphere in the vertical direction. The value of β generally varies from 0.0 to
0.5. The wavelength, λ, is in micrometers. It is called turbidity because of scattering of solar radiation by matter other than dry air molecules of the atmosphere in the optical sense. Consequently, τpλ includes attenuation due to dry as well as wet dust particles.
The wavelength exponent, α, is related to the size distribution of the aerosol particles. Large values of α indicate a relatively high ratio of small particles to large particles. It appears obvious that α varies in the range 0-4 when the aerosol particles are very small, of the order of air molecules; α should approach 4 and it should approach zero for very large particles2.
The wavelength exponent, α, is not a constant of nature but simply a parameter defined by the arbitrary procedure chosen for its determination12. Generally, α has a value between 0.5 and 2.5; a value of 1.3 is commonly employed, since it was originally suggested by Angstrom22. A good average value of α for most natural atmospheres is 1.3 ± 0.5.
However, this assumption has a rather limited validity since many researchers have found that α is a variable13-16. Throughout an individual day, changes in temperature cause evaporation or condensation of moisture in the atmosphere. The aerosols get affected both in number and size. Consequently, there is a variation in the values of turbidity parameters2.
2.4 Procedure for determining the values of Angstrom turbidity parameters, αααα and ββββ
The Angstrom’s equation [Eq. (1)] upon linearizing on a logarithmic format takes the form as17-20
ln τλ = ln β - α ln λ … (2)
This method yields the values of α and β by least squares analysis of spectral optical thickness. The values of α and β determined by this procedure are most adequate17.
In the present investigation, the Angstrom turbidity parameters have been determined for each scan. The precision limit of 90-98% in the correlation coefficient is set to estimate the values of α and β. By this procedure, a three thousand sets of α and β have been obtained. These values form the data base for further analysis of diurnal, monthly, seasonal and inter-annual features.
3 Discussion
3.1 Diurnal behaviour of αααα and βββ β
Like many other climatic variables, the values of α and β undergo a change throughout an individual day
because of changes in the atmospheric meteorological parameters. These changes may either decrease or increase the values of the Angstrom parameters. In the present study, the variations in α and β have been examined.
Figures 1-3 show the panel diagrams of the diurnal plots of α and β for the three aerosol years 2003-04, 2004-05 and 2005-06, each one covering June to May of the next year. In each of these years, a day in
monsoon (except monsoon 2003), a day in winter and a day in summer are shown as typical ones depicting the diurnal changes.
3.2 Monthly characteristics of αααα and ββββ
Over the observation period of three years, about three thousand sets of α and β distributed over 120 clear sky days of different months have been analyzed. In order to bring out the monthly features of
Fig. 1 — Diurnal plots of α vs β for the aerosol year 2003-04
Fig. 2 — Diurnal plots of α vs β for the aerosol year 2004-05
Fig. 4 — Day average of α and β for the month of December over three years
turbidity parameters, the data set is grouped under the months.
Thus, the values of α and β for a month, say December, of three years are treated as a single unit.
For each day, the ensembles of α and β are averaged out to obtain the mean values of α and β for the day.
Such day averages in the same month of the three years have been analysed. Months with more than five sets of α and β are only taken into account to bring out the monthly features.
Fig. 5 — Day average of α and β for the month of March over three years
Figures 4 and 5 show plots of day averages of α and β in the months December and March, respectively during winter and summer. The trend in α and β is not observed during the monsoon season because of the unavailability of the data. From these figures, the minimum and maximum values of each year have been derived.
3.3 Seasonal variations of αααα and βββ β
The varying meteorological parameters are the characteristics of the seasons of a year. It is, therefore,
Fig. 3 — Diurnal plots of α vs β for the aerosol year 2005-06
expected that the atmospheric aerosols be subjected to changes in number and size. Consequent to this, the extinction of solar radiation by aerosols is also affected. The turbidity parameters, which depend on the spectral aerosol optical thickness, are expected to exhibit changes through the seasons. The values of α and β are shown by bar graph in Fig. 6. The seasonal values of α and β are indicated by the height of the bars for each of the aerosol year. A slight increasing trend in β values is observed in the winter season from 2003-04 to 2005-06 with lowest β values during the winter of 2003-04. The β values for the monsoon of two years exhibit a slightly increasing trend.
During the summer of 2005-06, β is highest and it reduces to almost half the value during the other two summers. These variations are a reflection in the population of coarse particles during the three years.
The high value of β indicates large number of coarse particles. The values of α exhibit an anti-correlation with β values. Accordingly, the variation in the population of finer particles is opposite to that of coarse particles. With the increase in the day temperature, evaporation rate of water also increases, because of which there will be rise in atmospheric water vapour content. This increase in water vapour content influences the particle growth. Thus, small
sized (~0.1 µm) particle is converted to large size (0.1-1 µm) and a large sized particle is converted to coarse particle (>1 µm) which settles down to the ground under the influence of gravity. Hence, this leads to the observed variations in the values of α and β of the seasons.
3.4 Average seasonal trends of αααα and βββ over three years β
Seasonal averages of α and β have been presented in Fig. 7. Monsoon values are available only for two years 2004-05 and 2005-06. During these two years, both α and β show an increasing trend. Winter values of β during three years exhibit a slow increasing trend, whereas α is highest during the winter of 2003 and 2004 followed by reduced and almost constant values during the other two years. Summer β values have an increasing trend and the α values have a decreasing trend. The trends of α and β are indicative of the changes in the population of small and coarse particles during the seasons of the years. Summer exhibits larger changes. The β values of monsoon and winter indicate a smaller change of particle population. A significant change in finer particles is indicated by a larger decrease in α from 2003-04 to 2004-05, thereafter, the change is negligible.
Fig. 6 — Monthly average of α and β for the period of three years
3.5 Inter-annual trends in αααα and ββββ
A year is an overall cycle of seasons combined together. An average representation of α and β for a year would be helpful in the study of aerosol climatology. With this in view, the annual trend of turbidity parameters for three years has been presented in Fig. 8. From the figure, it is seen that β has increased from 2003-04 to a slightly higher value during 2004-05 and reaches high value during 2005- 06. Similarly, α with a high value during 2003-04 decreases to a small value and remains almost same during next two years. From these observations, it can be said that the atmosphere is loaded with coarse particles during 2005-06. A high concentration of fine-particle is seen during 2003-04. The α value does not vary significantly during the other two years except 2003-04, with an average value of α ~ 1.7.
4 Results
On examining the panels, the following observations emerge:
1. α and β exhibit an anti-correlation throughout the day during all the seasons. This indicates that there is continuous redistribution of fine and coarse particles under the influence of meteorological parameters.
2. During 2003-04, significant diurnal variations in α and β have been observed.
3. During the other two years, α and β do not exhibit large variation. However, a maxima and minima are observed in β and α, respectively at about 1300 hrs LT on 22 November 2004 and 24 December 2004. These are due to the increased
concentration of coarse particles and a simultaneous decrease in the concentration of fine particles, respectively. Constancy of α and β indicates no significant change in the coarse and fine particulates concentration. For Mysore, the average α parameter during these three years turns out to be approximately 2. This value lies within the range of -0.5 to 2.6 reported by Cachorro et al.18-20.
4.1 An empirical relationship between αααα and ββββ
Study of α and β, so far described, exhibits an anti- correlation between the two. However, this does not give a definite connection between the two. In addition, from the Angstrom’s equation, a simple relationship between the two cannot be realized. It is in this context that the numerical data of α and β has been examined to arrive at a plausible empirical relationship between the two.
Since α and β undergo a change in their values due to changing meteorological conditions over a day, a diurnal relationship is contemplated. By examining the graphical plots of α vs β, it is observed that the two exhibit a linear relationship on a large number of days when sky conditions remained stable. It is also evident that in general, throughout a day, β remains either low (≤ 0.05) or high (> 0.05). On the days when no regular trend is exhibited, the scatter may be attributed to abrupt changes in meteorological factors that may give rise to varying fine particle concentration while coarse particles remain unaltered.
This means there would be an influx of fine particles.
On the days when α - β plots show good linearity, by least squares analysis slope and intercept are
Fig. 7 — Seasonal averages of α and β over the years Fig. 8 — Annual average trend of α and β for three years
determined and these values have been presented in Table 1.
Table 1 shows that the values of intercept vary from 1.8 ± 0.2 to 4 ± 0.2. The value of 4 for the intercept indicates that the Rayleigh scattering prevails when the size of the particulates becomes small of the order of molecular species2. On the other hand, the slopes differ largely, in the range 2.3 ± 0.2 to 73 ± 4 (a linear decreasing trend is indicated by the
negative sign). This indicates that the rate of variation of α with β change day-to-day. On low β-days, the slopes are higher as compared to those of high β-days.
On a low β-day, the coarse particulates get depleted fast resulting in larger value of the ratio of fine to coarse particulates. This ratio is an index of α and consequently the values of α are high. The β- parameter is determined by the coarse particulates and low β-results when the coarse particles are depleted.
Table 1 — Slope and intercept values for α - β plots
Date Slope Error Intcpt Error Date Slope Error Intcpt Error
06 Dec 2003 -20.72 3.13 2.90 0.15 22 Feb 2006 -12.84 1.56 2.44 0.10
07 Dec 2003 -39.79 5.17 3.68 0.15 24 Feb 2006 -31.30 2.85 3.29 0.09
09 Dec 2003 -21.06 2.25 3.23 0.18 25 Feb 2006 -12.36 2.73 2.69 0.12
10 Dec 2003 -39.27 6.06 3.70 0.18 27 Feb 2006 -6.46 0.84 2.60 0.08
11 Dec 2003 -21.22 2.10 3.33 0.11 10 Jan 2004 -16.64 1.04 2.91 0.07
16 Dec 2003 -41.06 8.60 3.03 0.24 12 Jan 2004 -13.07 1.70 2.73 0.09
17 Dec 2003 -9.91 0.89 2.54 0.08 13 Jan 2004 -25.73 2.64 3.66 0.13
21 Dec 2003 -37.31 4.00 4.03 0.18 15 Jan 2004 -52.31 6.37 4.30 0.23
23 Dec 2003 -12.50 1.38 2.90 0.10 11 Jan 2006 -10.10 1.29 2.44 0.12
17 Dec 2004 -16.83 0.82 2.78 0.04 12 Jan 2006 -22.80 1.80 3.01 0.06
18 Dec 2004 -19.07 2.10 3.11 0.07 18 Jan 2006 -6.88 0.54 2.49 0.04
25 Dec 2004 -14.89 1.32 2.76 0.06 19 Jan 2006 -37.28 2.43 3.37 0.07
26 Dec 2004 -42.29 2.62 2.88 0.06 20 Jan 2006 -53.29 2.48 3.62 0.07
01 Dec 2005 -3.61 1.01 2.09 0.08 23 Jan 2006 -6.14 0.47 2.31 0.04
06 Dec 2005 -44.13 6.90 3.15 0.19 26 Jan 2006 -2.32 0.22 2.02 0.02
08 Dec 2005 -8.62 1.27 2.39 0.05 30 Jan 2006 -8.28 3.05 2.39 0.12
09 Dec 2005 -18.06 2.13 2.53 0.07 31 Jan 2006 -6.54 0.55 2.32 0.03
13 Feb 2004 -39.73 3.35 3.75 0.11 14 Apr 2004 -6.83 1.16 2.32 0.13
23 Feb 2004 -54.26 3.73 4.11 0.10 17 Apr 2004 -4.70 1.23 1.78 0.18
24 Feb 2004 -26.98 2.44 3.63 0.11 12 Apr 2006 -5.95 0.09 2.47 0.01
09 Feb 2005 -33.49 3.08 3.09 0.09 05 Mar 2004 -39.18 2.99 4.11 0.09
10 Feb 2005 -21.48 0.98 2.78 0.04 01 Mar 2005 -9.68 0.79 2.41 0.05
11 Feb 2005 -18.88 1.42 2.68 0.06 03 Mar 2005 -18.95 1.93 2.76 0.09
12 Feb 2005 -33.79 1.72 2.96 0.05 14 Mar 2005 -8.42 0.65 2.63 0.05
13 Feb 2005 -29.92 2.12 3.16 0.07 17 Mar 2005 -6.63 1.18 2.46 0.09
19 Feb 2005 -21.07 0.98 2.81 0.04 18 Mar 2005 -29.11 1.98 3.66 0.09
28 Feb 2005 -19.22 1.40 2.69 0.05 19 Mar 2005 -35.58 2.07 3.98 0.08
03 Feb 2006 -6.84 0.77 2.33 0.06 26 Mar 2005 -9.06 0.58 2.56 0.07
04 Feb 2006 -10.76 1.65 2.54 0.09 05 Mar 2006 -16.24 1.84 3.18 0.08
07 Feb 2006 -4.63 1.16 1.89 0.10 07 Mar 2006 -11.78 0.79 2.96 0.05
10 Feb 2006 -5.62 0.66 2.04 0.06 08 Mar 2006 -11.22 0.59 2.79 0.04
11 Feb 2006 -13.46 1.08 2.47 0.07 15 Mar 2006 -6.56 0.61 2.53 0.04
15 Feb 2006 -3.91 0.63 2.11 0.05 19 Mar 2006 -11.86 2.42 2.76 0.14
16 Feb 2006 -12.12 1.24 2.65 0.08 17 May 2005 -5.87 0.37 1.96 0.05
17 Feb 2006 -5.12 0.51 2.32 0.05 07 Oct 2004 -8.36 0.48 1.82 0.05
18 Feb 2006 -6.39 0.47 2.31 0.04 08 Oct 2004 -72.63 3.84 3.31 0.09
19 Feb 2006 -13.65 1.00 2.76 0.06 12 Jun 2006 -32.73 2.32 3.71 0.15
20 Feb 2006 -12.42 1.98 2.65 0.14
This phenomena explains large value of slope on low β-day. Under the influence of suitable meteorological parameters, the coarse particles may remain in the atmosphere. This results in large values of β, and α value will be low leading to smaller slopes, thus production-depletion processes control the values of α and β, and hence, the rate of variation between the two parameters.
In Fig. 9, three-year data of α and β are presented graphically. The scatter, indicates a polynomial fit, would be required as the day-slopes vary. On a logarithmic format, the scatter graph shows linear trend (Fig. 10).
By least squares fitting, the following equation is obtained:
lnα = -[0.30 lnβ + 0.29] … (3)
It is found that this equation yields, for a given value of β, an α value with an average variation of 13%. This expression is a synoptic equation based on the 3-year data collected at the site of observation.
Whether this equation is valid for any other location needs to be examined to ascertain the spatial applicability.
4.2 Visibility
The turbidity parameters play a significant role in determining the visibility of a location. It is known that at a fixed value of β, a lower value of α signifies higher visibility, that is higher atmospheric trans- parency. By inference, it can be concluded that the lower values of α, which means larger average particle size would result in higher amounts of solar
radiation reaching the ground. Further, at wavelengths shorter than 1 µm, finer particles transmit less energy than the coarser particles.
Selby & McClatchey21 have developed a relationship between the visibility (Vis) and the turbidity parameters as:
β = [0.55]α [3.912/Vis – 0.01162][0.02472(Vis – 5)
+ 1.132] … (4)
where, Vis is in kilometers. This expression is valid for visibilities more than 5 km. By following an iterative procedure in which α and β are inputs, the visibility can be determined.
From Fig. 11, it can be seen that the visibility is highest during winter followed by monsoon and summer.
Fig. 9 — Mass plot of α vs β for the three years
Fig. 10 — Plot of α vs β on a logarithmic scale
Fig. 11 — Season-wise average visibility over the years
Table 2 — Seasonal average visibility Average visibility, km Aerosol year
Monsoon Winter Summer
2003-04 No data 68 29
2004-05 65 65 35
2005-06 43 60 25
2006-07 41 No data No data
4.3 Seasonal trends through the years
The monsoon of 2004 shows highest visibility. A reduction of about 30% is indicated during the other two years. Winter exhibits a slow decreasing trend in visibility through three years. Summer of 2004-05 shows highest visibility and slightly reduced value during other two years with a slightly higher value in 2003-04 than in 2005-06. Table 2 presents visibility of the seasons during the years of observation. The values are the averages of monthly values of each season.
5 Conclusions
Important results have emerged by analyzing three- year data. On any day, α and β vary with respect to the change in meteorological parameters. Seasonally, summer records the maximum value of β followed by winter and monsoon, thereby, indicating more number of large sized particles. Washout and rainout of aerosols is the cause for low concentration of particulates in monsoon. The anti-correlation between α and β is the common feature for all the days of observation. An annual trend shows that β has increased from 2003-04 to a slightly higher value during 2004-05 and reaches a high value during 2005- 06. Similarly, α with a high value during 2003-04 decreases to a small value and almost the same during the next two years. From these observations, it can be said that the atmosphere is loaded with coarse particles during 2005-06. A high concentration of fine-particle is seen during 2003-04. In addition, the visibility is found to be highest during winter followed by monsoon and summer.
Acknowledgements
The authors thank Indian Space Research Organization for the financial assistance through the RESPOND scheme. The authors thank the University of Mysore for the facilities. Special thanks to Late Prof B. Narasimhamurthy for his kind support.
Sincere thanks to the Management of PESIT and to the colleagues in the department for their constant encouragement.
References
1 Szymber R J & Sellers W D, Atmospheric turbidity at Tucson, Arizona, 1956-83: Variations and their causes, J Clim Appl Meteorol (USA), 24 (1985) 725.
2 Iqbal M, An introduction to solar radiation (Academic Press, Ontario), 1983.
3 Zakey A S, Abdelwahab M M & Makar P A, Atmospheric turbidity over Egypt, Atmos Environ (UK), 38 (2004) 1579.
4 Angstrom A, On the atmospheric transmission of sun radiation and on dust in the air, Geografis Annal (USA), 2 (1929) 156.
5 Linke F, Transmission coefficient und trubungsfaktor, Beitr Phys Atmos (Germany), 10 (1922) 91.
6 Schuepp W, Die Bestimmung der komponenten der atmospharischen trubung aus aktinometer-messungen, Archiv Meteorol Geophys Bioclimatol B (Austria), (1949) 257.
7 Unsworth M H & Monteith J L, Aerosol and solar radiation in Britain, Q J R Meterol Soc (UK), 98 (1972) 778.
8 Narasimhamurthy B, Raju N V, Thukarama M, Prasad B S N
& Krishnamoorthy K, MWR studies of the temporal and spectral features of atmospheric aerosols over Mysore (12.3 N): Global change studies, ISRO-GBP: GCS-02-98 (Indian Space Research Organization, Bangalore), 1998, 77.
9 Ganesh K E, Umesh T K & Narasimhamurthy B, Aerosol optical thickness measurements on Tsunami day at a continental station, Mysore, J Aerosol Air Qual Res (Taiwan), 9 (2009) 94.
10 MICROTOPS II user manual (Solar Light Company, Philadelphia, USA), 2003.
11 Ganesh K E, Umesh T K & Narasimhamurthy B, Inter- annual variability in time scale characteristics of aerosol optical thickness over Mysore, India, Indian J Radio Space Phys, 39 (2010) 80.
12 Angstrom A, The parameters of atmospheric turbidity, Tellus (Sweden), 16 (1964) 64.
13 Xia Xiang-Ao, Chen Hong-Bin, Wang Pu-Cai, Zong Xue- Mei, Qiu Jin-Huan & Philippe Gouloub, Lageo, Tellus (Sweden), 57B (2005) 28.
14 Cachorro V E, Plinio Duran, Ricardo Vergaz & de Frutos A M, Measurements of the atmospheric turbidity of the North- centre continental area in Spain: Spectral aerosol optical depth and Angstrom turbidity parameters, J Aerosol Sci (UK), 31 (2000) 687.
15 Suresh T & Elgar Desa, Seasonal variations of aerosol over Dona Paula, a coastal site on the west coast of India, Atmos Environ (UK), 39 (2005) 3471.
16 Balkrishna Sapkota & Rajan Dhaubhadel, Atmospheric turbidity over Kathmandu valley, Atmos Environ (UK), 36 (2002) 1249.
17 Cachorro V E, de Frutos A M & Casanova J L, Determination of the Angstrom turbidity parameters, Appl Opt (USA), 26 (1987) 3069.
18 Cachorro V E, Gonzaley M J, de Frutos A M & Casanova J L, Fitting Angstrom’s formula to spectrally resolved aerosol optical thickness, Atmos Environ (UK), 23 (1989) 265.
19 Cachorro A M, Duran P, Vergaz R & de Frutos A M, Columnar physical and radiative properties of atmospheric aerosol in north central Spain, J Geophys Res (USA), 105 (2000 a) 7161.
20 Cachorro A M, Duran P, Vergaz R & de Frutos A M, Measurements of atmospheric turbidity of the North-centre continental area in Spain: Spectral aerosol optical depth and Angstrom turbidity parameters, J Aerosol Sci (UK), 31 (2000 b) 687.
21 Selby J E A & McClatchey R M, Air Force Cambridge Research Laboratories, Report AFCRL-72-0745 (AFCRL, USA), 1972.
22 Angstrom A, The parameters of atmospheric turbidity, Tellus (Sweden), 16 (1964) 64.
