Uncertainties in measured and modelled asymmetry parameters of mineral dust aerosols

https://doi.org/10.1016/j.jqsrt.2005.11.035Get rights and content

Abstract

The error caused by the uncertainty in the refractive index in the determination of the asymmetry parameter g is studied for a variety of mineral dust aerosol samples at two different optical wavelengths. Lorenz–Mie computations for spherical model particles are compared with results based on laboratory-measured phase functions in conjunction with a commonly used extrapolation method. The difference between the g-value based on measurements and the g-value based on Lorenz–Mie simulations is generally on the same order of magnitude as the error caused by the uncertainty in the refractive index m. For larger effective radii the error in g related to the use of spherical model particles is even larger than that related to the uncertainty in m. This indicates that the use of spherical model particles can be among the major error sources in the determination of the asymmetry parameter of dust aerosols.

Introduction

Understanding the impact of mineral dust aerosols on irradiance is of particular relevance for estimating the local direct climate forcing effect of aerosols in and near arid regions [1], [2], [3], [4]. Reliable estimates of the aerosols’ impact on the radiative net flux depend on accurate knowledge of the extinction optical depth, the single scattering albedo, and the asymmetry parameter of the aerosols. A main difficulty in determining these optical properties is the large number of uncertainties concerning the physical and chemical properties as well as the distribution of mineral aerosols in the atmosphere.

Information on aerosol optical depth (AOD) is often available from independent measurements, such as satellite observations. The single scattering albedo ϖ and the asymmetry parameter g are often obtained by Lorenz–Mie calculations that take into account additional information or assumptions about the aerosols’ size distribution (SD) and the complex refractive index m=n+iκ. The uncertainties associated with the SD and with m cause errors in ϖ and g, which contribute substantially to the error in radiative forcing simulations of dust aerosols. The m-uncertainty is generally considered the single most important error source [5].

Simulations of the aerosol optical properties and the local direct climate forcing effect of mineral aerosols are usually based on representing the aerosols by spherical model particles. It is generally taken for granted that this approximation introduces only small errors that are negligible in view of the large number of other error sources. However, recent case studies [6], [7] indicated that this error source may have been underestimated. At least in those few cases that have been studied so far [6], [7] the net-flux error caused by the use of spherical model particles has been comparable to the error caused by the uncertainty in the refractive index. The origin of this large net-flux error is a misrepresentation of the phase function and thus of the asymmetry parameter g of dust aerosols by spherical model particles. If those indications should be substantiated they would have far-reaching consequences for climate forcing simulations of mineral aerosols, at least on a local scale, and possibly even for other types of aerosol.

Information about the asymmetry parameter of mineral dust aerosols can also be obtained from laboratory measurements of phase functions. However, since it is technically difficult to measure scattering in and near the exact forward-scattering direction, such measurements need to be extrapolated, e.g. by use of Lorenz–Mie simulations. The basic idea of such an extrapolation is that the forward-diffraction peak is rather insensitive to particle shape and can therefore be reproduced reasonably well by use of spherical model particles. Such an extrapolation method has been developed and tested by Liu et al. [8]. However, since the forward-diffraction peak of the phase function is sensitive to the refractive index m of the aerosols, it is still unclear how reliable this method is if m is not known exactly.

In this paper we compare asymmetry parameters obtained from phase function measurements and from Lorenz–Mie simulations for nine different types of aerosols and at two different optical wavelengths. By means of this comparison we can assess the importance of the uncertainty in g related to the use of spherical model particles in relation to the error caused by the uncertainty in the refractive index. The details of the methodology are described in the following section. Results are presented and discussed in Section 3. Concluding remarks are given in Section 4.

Section snippets

Methods

The aerosol samples employed in this study have been taken from the Amsterdam Light Scattering Database (http://www.astro.uva.nl/scatter/; see also Refs. [9], [10], [11]). The experimental setup to measure the angular dependence of light scattering by aerosol samples in a laboratory is described in detail, e.g., in Ref. [9]. A nephelometer-type instrument specifically built for the purpose is used to measure the scattered power, which can be measured for scattering angles from 5 to 173°.

Results and discussion

The asymmetry parameter is defined asg=120πp(Θ)cos(Θ)sin(Θ)dΘ,where p denotes the phase function and Θ represents the scattering angle. Table 1 shows results at a wavelength λ=441.6nm for the asymmetry parameter gref of the reference samples (based on the laboratory observations) and for gMie based on Mie simulations. Table 2 presents corresponding results at a wavelength of λ=632.8nm.

Computation of gref depends on extrapolating the measured phase function to the full range of scattering

Summary and conclusions

As part of this investigation it was estimated how strongly an uncertainty in the refractive index affects the asymmetry parameter gref determined from laboratory-measured phase functions in conjunction with the extrapolation method by Liu et al. [8]. It was found that the error is growing as the significance of absorption increases in relation to scattering. More specifically, the uncertainty in gref increases with the imaginary part κ of the refractive index m and with the effective radius r

Acknowledgements

We wish to thank Hester Volten for making the measurement data available. We also thank Hannu Savijärvi, Jouni Räisänen, Petri Räisänen, and an anonymous reviewer for helpful comments on the manuscript.

References (16)

There are more references available in the full text version of this article.

Cited by (29)

  • Scattering asymmetry parameters for a circular cylinder in arbitrary–shaped acoustical sheets

    2021, Communications in Nonlinear Science and Numerical Simulation
    Citation Excerpt :

    Accurate calculations of the asymmetry parameter are of particular importance in scattering and related phenomena, as shown in numerous investigations in optics [9–12] to compute the albedo of snow and ice [13], the scattering of nonspherical particles [14], densely-packed grains [15,16], particle aggregates [17,18] and mineral dust aerosols [19] to name a few examples.

  • Experimental and simulated scattering matrices of small calcite particles at 647nm

    2013, Journal of Quantitative Spectroscopy and Radiative Transfer
    Citation Excerpt :

    Therefore, we also present the extrapolated scattering matrix that is defined in the entire angle range from 0° to 180° [9]. The extrapolation of the phase function is performed using the procedure suggested by Liu et al. [10] and subsequently adopted by, e.g., Kahnert and Nousiainen [11] and Muñoz et al. [9]. Although calcite is not a major component of the Martian surface, it is commonly considered to be particularly important for its link with climate evolution and water resources on Mars [12,13].

  • The Amsterdam-Granada Light Scattering Database

    2012, Journal of Quantitative Spectroscopy and Radiative Transfer
    Citation Excerpt :

    Since September 2003, the Dutch experimental data are freely available in digital form in the Amsterdam Light Scattering Database [17,18]. The success of this database is clearly demonstrated by the increasing number of different research groups (see e.g. [19–37]) that make use of the data. The Amsterdam Light scattering setup was closed in 2007, but a modernized and improved descendant of the Dutch scattering apparatus, the IAA Cosmic Dust Laboratory (CoDuLab), has been constructed at the Instituto de Astrofísica de Andalucía (IAA) in Granada, Spain [38].

  • Laboratory measurements of single light scattering by ensembles of randomly oriented small irregular particles in air. A review

    2011, Journal of Quantitative Spectroscopy and Radiative Transfer
    Citation Excerpt :

    The availability of computer algorithms for computing Lorenz–Mie scattering has favored the practice of treating irregular small particles as if they were spheres. Experimental data can help in studying the effect of particle nonsphericity in the calculated scattering properties [92,109,117–119]. In a more general context, experimental data are needed to validate some numerical codes/approximations used to interpret different natural phenomena in which light scattering by small nonspherical particles are involved [120–125].

  • Light scattering by large Saharan dust particles: Comparison of modeling and experimental data for two samples

    2011, Journal of Quantitative Spectroscopy and Radiative Transfer
    Citation Excerpt :

    Here we use this n=3 fit together with the measured size distribution of the feldspar sample to obtain our first input matrix. By using the modeled rather than the measured matrix, we avoid the issues related to extrapolating and renormalizing the measured scattering matrix [27–29]. The other two input matrices are based on light-scattering simulations conducted for thin calcite flakes by [30].

View all citing articles on Scopus
View full text