REVUE INTERNATIONALE D'HELIOTECHNIQUE ENERGIE - ENVIRONNEMENT - N° 36 (2007) 22-26
http:\\www.comples.org
___________________________________________________________________________________
F.Ellouz1* – M. Masmoudi1 and K. Medhioub2
1 Department of physics, Faculty of Science, Sfax, B.P 802, 3018, Tunisia
faten_ellouz@yahoo.fr Fax : 216 74 274 437
2 Institut Préparatoire aux Etudes d’Ingénieurs de Sfax, (IPEIS) Sfax, Tunisia
Reçu 28/04/07 . En ligne le 23/09/2007
ABSTRACT
Linke turbidity factor (TL) has been estimated at a Mediterranean costal site in Tunisia (Sidi Bou Said), during 1998-1999 period using the total direct solar radiation measurements
The urban site is in the north region of Tunisia and is influenced by air masses from maritime and continental origins. Solar direct radiation measurements have been made continuously with a pyrheliometer installed at the meteorological station of Sidi Bou Said.
The obtained results indicate high variability of the diurnal and monthly of the atmospheric turbidity factor. TL increases in summer with a maximum in august afternoon and decreases in winter. The increase of TL is an indication for increasing atmospheric turbidity level (pollution).
The correlation between atmospheric turbidity and the local weather conditions shows that the increase of TL is particularly due to the transport of particles of continental origin in the west sector. In the East sector, the atmospheric turbidity can be explained by both local effect and the effect of water vapour.
KEYWORDS
Solar radiation; Turbidity; Linke turbidity factor; Aerosols.
1- Introduction
Radiation from the sun as it passes through the atmosphere gets attenuated by the constituent gases, suspended particles called aerosols and clouds. The presence of aerosols makes the atmosphere turbid and the determination of this turbidity is of importance in climatology and for monitoring of atmospheric pollution. A number of atmospheric turbidity indices have been introduced during the past decades in order to estimate this turbidity. Among several turbidity coefficients the most frequently used are the Linke turbidity factor TL [1] and the Angstrom turbidity coefficient ß [2]. Atmospheric turbidity indices have been estimated in many locations, either on the basis of spectral irradiance or on all wavelengths direct irradiance data [3-10]. The Angstrom turbidity coefficient ß is an indication only of the amount of aerosols in the atmosphere. ß is obtained from spectral measurement.
Unlike Angstrom turbidity coefficient, Linke turbidity factor refers to the whole spectrum and
describes the optical thickness of the atmosphere due to both absorption by water vapour and absorption and scattering by the aerosol particles relative to dry and clean atmosphere [11-13]. The value of the Linke factor can be derived from data obtained from pyrheliometric measurement. Its value normally varies from 1 when the atmosphere is fully clean and dry to 8.
The objective of this study is to report results on atmospheric turbidity using linke’s factor calculated from radiation measurements in a coastal and tourist location of Sidi Bou Said, Tunisia and to increase the knowledge of Linke factor in the North of Africa. This site is influenced by air masses from different origins. Their effects on the characteristics and behaviour of the atmospheric turbidity, computed during two years, are analyzed and discussed. The daily, hourly and monthly variations of TL factor are also analysed and discussed.
On the other hand, atmospheric turbidity depends on the local weather conditions and on the climate of the site. Particularly, the wind direction and speed measured in the site are used to discuss the correlation with atmospheric turbidity.
2- Experimental measurements
Sidi Bou Said is an urban, tourist and coastal region in Tunisia. It is located at about 20 Km from the Tunisian capital (Fig 1). (Latitude North 36°52’, longitude East 10°21’).
Hourly measurements of direct solar radiation over Sidi Bou Said in the period of 1998-1999 are used to estimate the Linke turbidity factor. These measurements have been made continuously with a pyrheliometer installed at the meteorological station of Sidi Bou Said. It is important to define a clear sky condition because turbidity is an index for the comparison of cloudless atmospheric environments. In this work, the hours in which normal beam radiation is less than 200 w/m2 are ignored in the selection of clear skies [14].
Linke turbidity factor can be derived simply from pyrheliometric measurements of the direct normal irradiance (15): T L = [0.9+9.4. sin (h)]. Ln (I 0/I (h))
Where, I0 is the extraterrestrial solar irradiance at normal incidence (=1367 w/m2) and I (h) is the irradiance of direct normal solar radiation and h is the solar elevation.
This method has generally been used for calculating Linke’s turbidity factor in most works, thus permits a good comparison of our results with other sites [16-17].
The three–hourly standard meteorological data from the Sidi Bou Said station (given by the Meteorological National Institute) are also used to analyse the variability of the wind speed and direction.
3-Results and discussions
3-1-Temporal variation of Linke Turbidity factor
The study of the daily variation of annual averaged TL values together with the standard deviations shows a peak at 9h30 T.U (fig 2) corresponding to rush hours: this peak can be explained by the increase of traffic. Fig 2 shows also that the afternoon values of TL are upper than the morning ones. This fact can be explained by the increase of the traffic activities during summer afternoons, which could supply the atmosphere in dust and particles in suspension. Moreover, in consequence of an important evaporation of sea water in the morning, the sea breeze phenomena can manifest in the afternoon and then favour a load of water vapour in the local atmosphere. On the other hand, Fig. 3 gives the monthly variation of the TL factor computed at Sidi Bou Said. Inspecting fig.3, we can distinguish two seasons in the year. The summer season, which spans from April till August, is characterized by the increase of traffic activities during summer afternoons, and the presence of sirocco winds, which are frequent in July and August. These winds load the atmosphere with dust and coarse particles. In consequence, we recorded important values of TL in the summer season. The winter season, which spans from late September till Mars, is characterized by frequent precipitation. So, the value of the TL factor decreases during this season. This is due to the washing out of dust from the atmosphere by rainwater. The standard deviation varies between 0.96 and 1.34. The minimum value of σ is recorded in July and august. So, a feeble variation of atmospheric turbidity was observed during this period. On the contrary, the increase of σ in winter shows important fluctuations of TL, which can be explained by the preponderant effect of water vapour and the anthropogenic aerosols. These results are comparable to those recorded for other Mediterranean cities: Marrakech city (Morocco) [18] and Greece [6].
The scanning of the computed Linke turbidity factor during two years has allowed us to plot the probability distribution by classes of TL values (fig 4). In 1999, a considered frequency of high turbidity is recorded in a range of 4-6 for TL values (About 45 % of TL are upper than 4). However, in 1998, a considerable frequency of high turbidity is observed in a range of 3-5 for TL values (about 50 % of TL are upper than 3). Moreover, we note that the frequency of TL values upper than 6 is about 18 % in 1998, whereas, it is about 32 % in 1999 (fig 4).
On the other hand, the daily examination of the Linke turbidity factor shows that low atmospheric turbidity is observed in the morning and a considerable number of high turbidity cases is recorded in the afternoon (Fig.5).
3-2- Turbidity, wind direction and wind speed
Atmospheric turbidity depends on local weather conditions and on the climate of the site. The prevailing winds, which may transport moisture or aerosols particles from distant sources, play a major role in the temporal variation of turbidity.
3-2-1- Turbidity and wind speed
In order to analyse the influence of wind direction on the turbidity values, wind data from the meteorological Sidi Bou Said station have been studied during the 2-year period of the study.
A statistical study of the winds data has allowed us to plot the probability distribution by a group of directions:
The percentage of data available from each group of directions is:
1-North sector (from NW to NE) represents 37.28 % of the observed winds
This wind sector could affect the atmosphere by particles of various origins (maritime or industrial from various polluted sources, or by water vapour).
2- East sector (from NE to SE) represents 29.87 % of the observed winds.
Eastern winds carry wet air masses of maritime origin (travelling over the Mediterranean), which are humid, then it can load the atmosphere in water vapour and maritime particles which are rich in salt.
3- West sector (from SW to NW) represents 17 % of the observed winds.
These winds carry dry air masses of continental origin.
4- South sector (from SW to SE) represents 15.85 % of the observed winds.
These winds carry humid air masses of maritime origin and also dry air masses of continental origin.
The influence of wind speed on atmospheric turbidity has been analysed for all records. Intervals of 1 m/s were chosen for the wind speed and the means of the Linke factor were calculated and then plotted for these wind speed intervals (fig. 6).
Fig. 6 shows that, generally, TL increases with high wind speed.
The standard deviation σ is important for high and low wind speeds. Moreover, a statistical study shows that:
· for high turbidity levels (TL>5.5), 60 % of the wind speeds are upper than 5 m/s and 40 % of wind speeds are less than 5 m/s.
· For moderate and low turbidity levels (TL <5.5), 58 % of the wind speeds are less than 5 m/s and 42 % corresponds to high wind speeds (>5m/s).
In the case of the turbidity observed for low wind speeds, we think that turbidity is due to local effects: the influence of traffic at rush hours, the influence of neighbouring sources of industrial pollution and turbidity can also be caused by the heavy water vapour content of maritime air masses carried by eastern winds. When the speeds are more important, we can explain the increase of turbidity by the transport of aerosol particles from various origins in addition to local effects [19].
3-2-2-Turbidity and wind direction
The atmospheric aerosol content is expressed by Angstrom’s turbidity coefficient (ß). This factor is defined in Angstrom’s formula δa (λ) = ß λ –α (where δa is the optical thickness caused by aerosols). The study of the variation of the Linke’s factor (TL) and the Angstrom turbidity coefficient (ß) illustrates the linear correlation among TL and ß [20]. Besides, the variation of atmospheric turbidity with ß shows that ß increases when wind speeds are more important [21].
In order to analyse the influence of wind direction on the turbidity values, the wind speed influence on atmospheric turbidity has been analyzed in the west (continental air masses) and the east sector (maritime air masses) separately.
In the west sector, when the wind speeds are upper than 5 m/s, TL shows a linear regression with a correlation coefficient =0.77 (Fig. 7).This correlation shows that the increase of TL is particularly due to the transport of particles of continental origin.
In the east sector, for the high and low speeds, the Linke’s factor fluctuate about important values (fig.8). Moreover, we note that, in this sector, wind speed cannot exceed 8m/s. Hence, atmospheric turbidity can be explained by the local effect and the effect of water vapour.
4-CONCLUSION
For two years, direct solar irradiation was used to compute the Linke turbidity factor at the Sidi Bou Said station (Northern Tunisia). These are the first published data for a long period for Tunisian localities. These results increase the sparse knowledge in the coastal Mediterranean area. The mean values and frequency of occurrence of this index have been used to characterize the atmospheric turbidity of the site. Values of the turbidity of the atmosphere where found to vary with seasons, months and hours.
The monthly mean values of TL shows that atmospheric turbidity decreases during the winter season (rainy season) and increases during the summer season (from April to August).
The turbidity of the atmosphere also varies with the time of day; we note that the Linke’s factor increases from noon TU. This situation can be explained by a supplementary load of the atmosphere in water vapour due to sea breeze and transport of aerosols due to the increase of traffic in the site, essentially during summer afternoons.
Atmospheric turbidity depends on the local weather conditions; the prevailing winds may transport aerosol particles from various sources, depending on the speed of the wind and its direction. In the west sector, when the wind speeds are upper than 5 m/s, TL shows a linear regression. Then, the west winds carry continental particles. The east winds carry wet air masses of maritime origin travelling over Mediterranean which are humid. Maritime moisture combined with anthropogenic aerosols give rise to an important fluctuation of turbidity.
ACKNOWLEDGEMENTS
The authors gratefully thank Mr. EZZINE Ali and Mr. ELLEUCH for providing us with the diurnal records of solar radiation components and meteorological parameters.
Special thanks are expressed to Mr. CHAABANE Mabrouk for his help and remarks.
References
[1] Linke F.Transmissions-Koeffizient und Trubungsfaktor.Beitr phys Fr Atmos ; 10:91-103 (1922).
[2] Angström A. Techniques of determining the turbidity of the atmosphere. Tellus; 13:214-23 (1961).
[3] Flowers EC, Mc Cormik RA, Kurfis KR. Atmospheric turbidity over the United States, Journal of applied Meteorology, pp.955-962, (1969)
[4] Katz M., Baille A., Mermier M., Atmospheric turbidity in a semi rural site I: Evaluation and comparison of different atmospheric turbidity coefficients, Solar Energy, Vol. 28, (1982).
[5] Kaufman Y. and Tanré D., Strategy for direct and indirect methods for correcting the aerosol effect on remote sensing: From AVHRR to EOS-MODIS, Remote Sensing Environment, Vol.55, PP.65, (1996).
[6] Rapti A.S., Atmospheric transparency, atmospheric turbidity and climatic parameters, Solar Energy Vol. 69, N°2, pp.99-111, (2000).
[7] Hussain M., Khatun S., Rasul M., Determination of atmospheric turbidity in Bangladesh. Renewable Energy, Vol. 20(3), pp.325-357, (2000).
[8] Masmoudi M, Chaâbane M, Tanré D, Goloub P, Blarel L, Elleuch F. Spatial and temporal variability of aerosol: size distribution and optical properties. Atmospheric Research; 25: 1-13 (2003).
[9] Zakey A.S, Abdelwaheb M.M, Makar P.A, Atmospheric turbidity over Egypt, Atmospheric Environnement, Vol.38, pp 1579-1591, (2004).
[10] Chaâbane. M, Masmoudi.M, Medhioub. K and Elleuch.F; Daily and monthly averaged aerosol optical variability deduced from AERONET sun photometric measurements at Thala site (Tunisia) accepted for publication in ‘Meteorology and Atmospheric Physics’(2005)
[11] Kasten F, The Linke turbidity factor based on improved values of the integral Rayleigh optical thickness. Solar Energy 56, 239-244 (1996).
[12] Cucumo M, Marinelli V, Oliveti G, Experimental data of the Linke turbidity factor and estimates of the Angström turbidity coefficient for two Italian localities. Renewable Energy 17: 397-410 (2000).
[13] Chaâbane M, Masmoudi M, Medhioub K. Determination of Linke turbidity factor from solar radiation measurement in northen Tunisia. Renewable Energy 29: 2065-2076 (2004).
[14] Karayel M, Navvab M, Ne’eman E, Selkowitz S. Zenith luminance and sky luminance distributions for daylighting calculations. Energy Bulding ; 6(3): 283-91 (1984).
[15] Kasten F, A simple parameterization of the pyrheliometric formula for determining the Linke turbidity factor. Meteorl. Rdsch ; 33, 124-127(1980).
[16] Ineichen P and Perez. R, A new airmass independent formulation for the Linke turbidity coefficient, Solar Energy, p 151-157 (2002).
[17] Hamdy K, Elminir, R.H.Hamid, El hussainy F., Ahmed E.Ghitas, M.M. Beheary and Khaled M.Abdel-Moneim, The relative influence of the anthropogenic air pollutants on the atmospheric turbidity factors measured at un urban monitoring station, Science of the total environnement, (2006).
[18] Jamali. A, Idliman. A, Kaoua. M, Fliyou. M, Experimental study of the atmospheric turbidity in Marrakech city.Tétouan, Maroc..p 485-92 (2002).
[19] Masmoudi.M, Belghith..I, and Chaabane.M, Element particle size distributions measured and estimated dry deposition in Sfax (Tunisian), Atmospheric Research, 63, 209-219 (2002).
[20] Molineaux B.and Ineichen P; On the broad band transmittance of direct irradiance in a cloudless sky and its application, (1996)
[21] Masmoudi M, Chaabane M, Medhioub K, Elleuch F. Varibility of aerosol optical thickness and atmospheric turbidity in Tunisia. Atmospheric Research 66: 175-88 (2003).