Astronomy for the developing world
IAU Special Session no. 5, 2006
J.B. Hearnshaw and P. Martinez, eds.
c 2007 International Astronomical Union
doi:10.1017/S1743921307006825
Effect of altitude on aerosol optical
properties
Aziza Bounhir1,2 †, Zouhair Benkhaldoun2 , El Arbi Siher2.3
and L. Masmoudi4
1
Faculté des Sciences et Techniques, Département de Physique Appliquée, Marrakech, Maroc
email: bounhiraz@yahoo.fr
2
Faculté des Sciences Semlalia, Département de Physique, LPHEA, Marrakech, Maroc
email: zouhair@ucam.ac.ma
3
Faculté des Sciences et Techniques, Département de Physique, BP: 523, Béni Mellal, Maroc
email: siher@ucam.ac.ma
4
LETS, Faculté des Sciences, avenue Ibn Battouta, BP: 1014 Rabat, Maroc
Abstract. The ELT project is currently under way in Europe and North America. Astronomical
sites depend critically on sky transparency and on aerosol loadings. A quantitative survey of
aerosol optical properties at candidate ELT sites is an essential part of the site selection process.
There are basically two methods to characterize aerosol properties: ground based measurements
and satellite measurements. In this paper we will establish a full climatology of two sites very
close to each other, but at a difference of 2300m in altitude: Izaña and Santa Cruz located in
the Canary Islands. Both have sun photometers from the AERONET network. We also use the
aerosol index determined from TOMS satellite data to determine how aerosol optical properties
vary with altitude. We establish a correlation between the TOMS index and the aerosol optical
thickness in both sites. Aerosol optical properties show very good correlation between Izaña
and Santa Cruz. As a result we establish a set of relationships helpful to characterize sites at
elevated altitude from data of neighbouring sites at low altitude.
Keywords. Site testing, atmospheric effects, aerosols, sun photometer
1. Introduction
It is important for astronomers to identify high-quality observatory sites. An exhaustive
approach towards prospecting for good sites is to make measurements at all potential
candidate sites. However this is very difficult and time consuming. The ideal is, if possible,
to make measurements at accessible low-altitude places near the candidate sites and then
to extrapolate the results to more elevated places. An important study is to see how
geophysical data affecting astronomical observations vary with altitude. In this paper,
we focus on aerosol optical properties at two sites located in the Canary Islands; Izaña
at an altitude of 2367 m and Santa Cruz at an altitude of 52 m and separated from each
other by some dozen kilometres.
2. Data and method
The instruments used for ground measurements are the CIMEL sun photometers from
the AERONET Network. These radiometers make measurements of direct sun and diffuse
sky radiance within the spectral range of 340–1020 nm (Holben, et al. (1998)). The direct
† Present address: Faculty of Science and Technology, Department of Applied Physics, Marrakech, Morocco
111
112
A. Bounhir et al.
0.9
AOT Izana
AOT Santa−Cruz
0.8
Aerosol optical thickness (440 nm)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
mar−05 april
may
june
july
aug
sept
oct
nov
dec
jan−06
feb
Figure 1. Daily values of the aerosol optical thickness (AOT) at Izaña and Santa Cruz.
sun measurements are acquired in eight spectral channels 340, 380, 440, 500, 670, 870,
940, 1020 nm. The 940 nm band is used to estimate total precipitable water vapour
content (WVC). The bandwidths of the interference filters vary from 2–10 nm. The
aerosol optical thickness (AOT) is computed from the Bouguer-Beer-Lambert law. The
Angstrom parameter (ANG) is derived from a multispectral log-linear fit to the classical
equation: AOT ∝ λ−AN G .
The sky radiance almucantar measurements are acquired at 440, 670, 870 and 1020
nm. A flexible algorithm for retrieval of aerosol physical properties developed by Dubovik
& King, (2000) was used for retrieving aerosol size distribution over a range of sizes
from 0.05–15 µm, together with spectrally dependent complex refractive index and SSA
(Single Scattering Albedo) from spectral and sky radiance data. An automated and
computerized cloud screening algorithm (Smirnov, et al. (2000)) was applied to direct
sun measurements.
The use of satellite observations is the most efficient way to determine aerosol physical
properties on large temporal and spatial scales. Among these instruments, the Total
Ozone Mapping Spectrometer (TOMS) has the capability to sense aerosols (Hermann,
et al. (1997)) and derive their optical properties (Torres, et al. (1998), (2002)) over both
land and ocean, through the aerosol index (AI).
3. Results
The period of study extended from March 2005 until February 2006. We will characterize atmospheric optical conditions by the aerosol optical thickness at 440 nm (AOT),
the water vapour content (WVC) and the Angstrom parameter (ANG) (870-440 nm).
Daily AOT values in Fig. 1 depict very low values for Izaña compared to Santa Cruz,
except during summer time, where dust events occur. The annual mean for Izaña is 0.08,
which is 2.8 times less than the annual mean of Santa Cruz (0.23). From this Figure we
can deduce that the aerosol layer is below Izaña’s altitude most of the time, except in
summer when dust events occur. During July and August the aerosol layer is higher than
2400 m as reported by (Hsu, et al. (1999)).
The annual mean of the Angstrom parameter is 1.2 for Izaña and 0.6 for Santa Cruz.
Small-particle aerosols dominate Izaña’s atmosphere. The Angstrom parameter at Izaña
113
Effect of altitude on aerosol ...
1.5
0.8
Santa−Cruz Angstrom parameter
aerosol optical depth 440 nm at Santa−Cruz
0.9
0.7
0.6
0.5
0.4
0.3
0.2
0.1
1
0.5
0
0
−0.1
0.05
0.1
0.15
0.2
0.25
0.3
0.35
aerosol optical depth 440 nm at Izana
0.4
0.2
0.4
0.6
0.8
1
1.2 1.4 1.6
Izana Angstrom parameter
1.8
2
2.2
Santa−Cruz water vapor content (cm)
6
5
4
3
2
1
0.5
1
1.5
Izana water vapor content (cm)
2
Figure 2. Scattergrams of AOT Santa Cruz versus AOT Izaña, ANG Santa Cruz versus ANG
Izaña and WVC Santa Cruz versus WVC Izaña.
is around 1.5 most of the time except in summer during dust events. The annual mean
water vapour content is 0.5 cm for Izaña and 2.2 cm for Santa Cruz.
The aerosol optical thickness (AOT) frequency distribution is very narrow at Izaña
with a modal value at 0.05 and broader at Santa Cruz with a modal value around 0.1.
About 70% of Izaña’s AOT occur below 0.05, which denotes good observation conditions.
About 70% of Santa Cruz’s AOT occurs below 0.25.
The Angstrom parameter histogram is bimodal at both sites. The modes at Izaña are
0.3 and 1.6. The second one is dominant. At Santa Cruz the modes are 0.3 and 0.9.
The water vapor content histogram modes are 0.3 cm at Izaña and 2.5 cm at Santa
Cruz. About 95% of the occurrence is below 1.2 cm for Izaña and higher than 1.2 cm for
Santa Cruz.
Figure 2 shows scattergrams of the relation between Santa Cruz and Izaña concerning
aerosol optical depth, Angstrom parameter and water vapour content. The resulting
relationships show notable correlations (correlation coefficients of 0.92, 0.78 and 0.66,
respectively).
AOTSantaCruz = 1,5(±0.15)*AOTIzaña + 0.07(± 0,02) R=0.92
ANGSantaCruz = 0.35(± 0.098)*ANGIzaña + 0.25(± 0,11) R=0.66
WVCSantaCruz = 1,6(± 0.31)*WVCIzaña + 1,33(±0,2) R=0.78
No specific correlation was found for monthly mean values of the single scattering
albedo (SSA). Monthly SSA values at 440 nm vary from 0.95 to 0.6 at Santa Cruz and
from 0.9 to 0.7 at Izaña. The values at Santa Cruz are most of the time higher than
114
A. Bounhir et al.
the Izaña values, except in October, January and February, which means that Izaña’s
aerosols are more absorbent.
Concerning the relationships between the Izaña and Santa Cruz aerosol size distributions, we found very high correlations, varying from 0.97 for June and decreasing
progressively to 0.6 for December.
Concerning the satellite measurements, the relationships between the TOMS aerosol
index (AI) and the corresponding aerosol optical thicknesses (AOT) give satisfying correlations: the correlation coefficient of AI and AOT for Izaña is 0.68 and the one for Santa
Cruz is 0.75. The relationships concerning AOT and AI are:
AOTIzaña = 0.12*AIIzaña + 0.05 R = 0.68.
AOTSantaCruz = 0.22*AISantaCruz + 0.16 R = 0.75.
We can thus retrieve the aerosol optical thickness starting from the aerosol index signal.
4. Conclusion
In this work we have characterized the aerosol optical properties of Izaña and Santa
Cruz during a year. For that purpose we used the AERONET data. We have established
the climatology of both sites. We find linear relationships between Izaña and Santa
Cruz aerosol optical properties; aerosol optical thickness, water vapour content and the
Angstrom parameter give good correlations (92%, 78,5% and 66,4%, respectively). Size
distribution correlates well (R varying from 98% to 60%) for June, July, August, September, October, November and December. January, February, March, April and May give
no correlation. The single scattering albedos of both sites do not seem to correlate. One
surprising thing is that the single scattering albedo decreases with increasing wavelength,
even during dust events. We have established correlations between TOMS Aerosol Index
(AI) and the aerosol optical thickness (AOT) (R at Izaña is 68,5% and at Santa Cruz
75,5%). Now the question is, can we use these relationships in other locations close to
the Canary Islands, like the Atlas mountains of Morocco, for example?
Acknowledgements
We thank ESO, especially Marc Sarazin, IAU and the Cadi Ayyad University for
financial suport. Our thanks to TOMS and AERONET for the use of their data.
References
Dubovik, O., King, J.M.D. 2000, J. of Geophys. Res. 105, 673
Hermann, J.R., Barthia, P.K., Torres, O., Hsu, C., Sefter, C. 1997, J. of Geophys. Res. 102,
16911
Holben, B.N., Eck, T.F., Slutsker, D., Tanr, D., Buis, J.P., Setzer, A., Vermote, E., Reagan, J.A.,
Nakajima, T., Lavenu, F., Jankowiak, I., Smirnov, A. 1998, Remote Sensing of Environment
66, 1
Hsu, N.C., Herman, J.R., Torres, O., Holben, B.N., Tanr, D., Eck, T.F., Smirnov, A., Chatenet,
B., Lavenu, F. 1999, J. of Geophys. Res. 104, 6269
Smirnov, A., Holben, B.N., Eck, T.F., Dubovic, O., Slutsker, I. 2000, Remote Sensing of Environment 73, 337
Torres, O., Barthia, P.K., Hermann, J.R., Ahmad, Z., Gleason, J. 1998, J. of Geophys. Res. 103,
99
Torres, O., Herman, J.R., Barthia, P.K., Sinyuk, A. 2002, Advances in Space Research 29, 1771