Elsevier

New Astronomy

Volume 13, Issue 1, January 2008, Pages 41-52
New Astronomy

Aerosol columnar characterization in Morocco: ELT prospect

https://doi.org/10.1016/j.newast.2007.06.011Get rights and content

Abstract

The work presented in this paper focuses on site testing in terms of aerosol loadings where ground based measurements are essential. In our case they are materialized by the aerosol optical thickness (AOT) and are provided by the Aerosol Robotic Network (AERONET) network from four stations, Dakhla and Marrakech in Morocco and Santa-Cruz and Izana in the Canary Islands. To fully scan all the area of the Moroccan territories, satellite measurements are certainly the most efficient way. We used the most popular and reliable products. Total ozone mapping spectrometer (TOMS) aerosol index (AI) provided by both TOMS Earth Probe and TOMS ozone monitoring instrument (OMI) along with aerosol optical thicknesses provided by moderate resolution imaging spectroradiometer (MODIS) and multiangle imaging spectroradiometer (MISR) instruments, onboard Terra platform. The idea is to compare sensing capabilities of each instrument in the region under study, in order to know which is suitable for a given place and when. For that purpose linear regression analysis were performed between satellite data and AERONET observations. Good correlations were observed with the Pearson correlation coefficient, R, varying from 0.68 to 0.92 for MODIS, MISR and TOMS OMI. However for TOMS EP the correlations are fairly poor (from 0.54 to 0.74). A ten years analysis of the TOMS EP index has been performed with a calibration of the aerosol index into TOMS retrieved aerosol optical thickness in the area of interest (Morocco and Canary Islands) and an inter-comparison with the other products was achieved. In the frame of the extremely large telescope (ELT) project prospect, once the appropriate satellite instrument have been chosen and the area scanned, the next step would be to scan aerosol loadings at higher altitude locations. Since vertical distribution of aerosol optical thickness and microphysical properties are not well understood and modelized, we used the relationships related to Izana (Izana’s altitude is 2367 m), as a first attempt, to extrapolate the aerosol optical thickness at higher locations in the Moroccan mountains. Izana and Santa-Cruz very close to each other (30 km) are located in the same satellite pixel and then have the same satellite (AOT) or (AI) whereas AERONET gives very distinct aerosol optical depths. A good linear correlation (R = 0.92) has been observed between the AERONET aerosol optical depths at Izana and Santa-Cruz. The seasonal correlation coefficients are 0.85 for winter, 0.87 for spring, 0.91 for summer and 0.87 for autumn The ratio AODSanta-Cruz/AODIzana has a seasonal behavior, reaches the average of 4.5 in winter and 2 in summer time and the subtraction of the aerosol optical thicknesses has an average of 1.3. Finally we retrieved the aerosol optical thickness at Oukaimeden: a Moroccan observatory located at 2700 m above sea level, and about 70 km from Marrakech city. We then converted the aerosol optical depth into astronomical light extinction and compare with previous records measured at the observatory.

Introduction

Extremely large telescopes are considered worldwide as one of the highest priorities in ground based Astronomy. They will advance astrophysical knowledge and could completely change our understanding of the universe and may answer fundamental questions about exo-planets, dark matter and energy… At the present time, many countries are involved in the ELT project prospect, from site testing and selection to instrumentation. Morocco is part in the extremely large telescope (ELT) project prospect. A team of Moroccan astronomers is busy working on site qualification and testing. Since the performance of large telescope at visible and infrared wavelengths is critically dependant on sky transparency and then on atmospheric aerosol cover (Muñoz-Tuñon et al., 2004, Sarazin, 2006), a quantitative survey of the aerosol loadings and their microphysical and optical properties is an essential part of the site selection process.

The study of sky transparency has been initiated by astronomers, who were carrying about photometric characterization of their observatories. Stellar photometric measurements are carried out in most observatories as routine astronomic observations, since 1950 (1970) for La Silla and (1980) for La Palma. Before daytime photometry being popular, atmospheric scientists took expertise from astronomers (Laulainen, 1977, Laulainen et al., 1977. Formenti et al. (2002)) have used a record of atmospheric nighttime turbidity data from astronomical stellar radiation measurements, in South Africa, to study changes in aerosol concentrations over the past three decades prior to the popularization of sun photometry. Astronomical light extinction coefficient can be converted to aerosol optical depth and vice versa. Since the dilemma concerning the heating of the Earth, the new field of aerosols; their optical and microphysical properties with sun photometry, LIDAR techniques and satellite remote sensing, is emerging and growing very fast. Now astronomers can take advantage of these well developed tools to characterize sky transparency with even more deeper insight.

The astronomical light extinction (A) is determined by comparing the apparent brightness, B, of the standard stars with there intrinsic luminosity at different air masses, through the Beer’s law. The Langley technique relying on the plot of ln(B) versus air masses is used to extract the astronomical light extinction which is the slope of the linear regression. Daylight photometry is based on the same physical principles: Iλ = Iλ0  exp(τtot  m); Iλ is the direct solar irradiance, Iλ0 is the extraterrestrial irradiance, λ is the wavelength, τtot is the total atmospheric optical depth; the sum of the optical depths of molecules (Rayleigh scattering), gazes (absorption) and aerosols (Mie scattering and absorption), m is the inverse of the cosine of the zenith angle θ. As for nighttime photometry, the linear regression Langley procedure is applied to extract the atmospheric total optical depth. The aerosol optical depth is then calculated by subtracting to τtot the aerosol optical depth of Rayleigh scattering τR and the Ozone absorption optical depth τO3 in the Chappuis bands, and the absorptions of water vapor, carbon dioxide, methane and nitroxyde, depending on the wavelength used. The astronomical light extinction coefficient A is related to the total optical depth: A = 1.086  τtot. The scattering by molecules has a predictable profile depending on the wavelength, altitude and the refractive index of molecules. The optical thickness of ozone is calculated using daily averages from TOMS satellite.

The rest of the paper is organized as follows. In Section 2, we describe the data used in the current study, the area of interest, and the methodology adopted. In Section 3, we show the details and the results of the regression analysis performed between ground measurements and the different satellite data. Section 4 deals with comparison of ground measurements of Izana and Santa-Cruz. In Section 5, first the TOMS EP index is converted to aerosol optical thickness mapping the area of study, and then we made a comparison of long term satellite AOT retrievals. Finally, we retrieve the AOT at Oukaimeden observatory, by different approaches. In order to compare to previous measurements of astronomical light extinction coefficient performed at the observatory during the year 1997, we converted the AOT to astronomical light extinction coefficient over 1997 using TOMS EP records.

Section snippets

Study area

Two AERONET stations located in Morocco; Marrakech (lat = 31.6°; lon = −8.15°; alt = 420 m) and Dakhla (lat = 23.7°; lon = −15.95°, alt = 12 m), have delivered data during at least two years. Close to Morocco, two other AERONET stations located in the Canary Islands; Izana (lat = 28.3°, lon = −16.5°; alt = 2367 m) and Santa-Cruz (lat = 28.5°, lon = −16.25°, alt = 52 m) have delivered data during relatively a long period of time. The mape showing the locations of the AERONET sun-sky radiometers used in this study is shown

TOMS aerosol index and AERONET aerosol optical thickness correlation

The only available long term record of atmospheric aerosols over both oceanic and continental areas is provided by TOMS. Indeed TOMS data are delivered since 1978 until now; Nimbus 7 from November 1978 until May 1993, Meteor 3, from August 1991 until December 1994, Earth Probe from September 1996 until December 2005 and OMI from 2004 until now. The primary importance of this instrument resides in the fact that it is possible to study the variability of the aerosols over the last three decades

Altitude effect

Vertical distribution of aerosol properties is not well understood; there is no systematic way of deducing aerosol microphysical and optical properties at a given height from aerosol properties measured on the ground just below. Active sensors like LIDAR can give the average position of the aerosol layer. In this section we will focus on AERONET AOT at Izana and Santa-Cruz and make regression analysis. As the AERONET data at Santa-Cruz and Izana are level 1.5 and then are not manually

Retrieval of Aerosol optical thickness and light extinction coefficient at Oukaïmeden observatory

First, we want to confront long term satellite aerosol loading previsions from different instruments. Fig. 5 illustrates the average aerosol optical thicknesses of MODIS and MISR instruments from 2000 until 2006. These images (MODIS and MISR) were acquired using the GES-DISC Interactive Online Visualization and Analysis Infrastructure (Giovanni) as part of the NASAs Goddard Earth Sciences (GES) Data and Information Services Center (DISC). MISR AOT forecast for the Canary Islands and the

Conclusion and perspectives

In this work we have characterized sky transparency by means of aerosol optical properties trough the aerosol optical thickness parameter in the area of Morocco and the Canary Islands. Ground based measurements are provided by the AERONET Network in four locations: Dakhla, Marrakech, Santa-Cruz and Izana. TOMS Earth Probe aerosol index have been used to have long term aerosol cover in the four AERONET stations. Seasonal effects of aerosol cover have been clearly demonstrated through ten years

Acknowledgements

We would like to thank TOMS and AERONET groups from NASA Goddard Space Flight Center for providing data. Thanks to GIOVANNI group. We thank the investigators: Emilio Cuevas, Hamad Benchekroun, Brent Holben, Philippe Goloub, and Benoit Duchemin and their staff for establishing and maintaining the sites of Marrakech, Santa-Cruz, Izana and Dakhla used in this investigation.

References (28)

  • P. Formenti et al.

    Atmos. Res.

    (2002)
  • B.N. Holben et al.

    J. Remote Sens. Environ.

    (1998)
  • N.S. Laulainen

    Atmos. Environ.

    (1977)
  • N.S. Laulainen et al.

    Atmos. Environ.

    (1977)
  • A. Smirnov et al.

    Remote Sens. Environ.

    (2000)
  • O. Torres et al.

    Adv. Space Res.

    (2002)
  • Benkhaldoun, Z., Siher, E.A., 1998. Photometry análisis using IRIS database. SOHO6/Gong98 Workshop. Boston, USA, ESA...
  • O. Dubovic et al.

    J. Geophys. Res.

    (2000)
  • O. Dubovic et al.

    J. Geophys. Res.

    (1998)
  • O. Dubovic et al.

    J. Geophys. Res.

    (2000)
  • O. Dubovic et al.

    Geophys. Res. Lett.

    (2002)
  • Ginoux P., Torres, O., J. Geophys. Res., 108(D17), 4534....
  • J.R. Herman et al.

    J. Geophy. Res.

    (1997)
  • N.C. Hsu et al.

    J. Geophys. Res.

    (1999)
  • View full text