ALIS: An efficient method to compute high spectral resolution polarized solar radiances using the Monte Carlo approach

https://doi.org/10.1016/j.jqsrt.2011年03月01日8 Get rights and content

Abstract

An efficient method to compute accurate polarized solar radiance spectra using the (3D) Monte Carlo model MYSTIC has been developed. Such high resolution spectra are measured by various satellite instruments for remote sensing of atmospheric trace gases. ALIS (Absorption Lines Importance Sampling) allows the calculation of spectra by tracing photons at only one wavelength. In order to take into account the spectral dependence of the absorption coefficient a spectral absorption weight is calculated for each photon path. At each scattering event the local estimate method is combined with an importance sampling method to take into account the spectral dependence of the scattering coefficient. Since each wavelength grid point is computed based on the same set of random photon paths, the statistical error is almost same for all wavelengths and hence the simulated spectrum is not noisy. The statistical error mainly results in a small relative deviation which is independent of wavelength and can be neglected for those remote sensing applications, where differential absorption features are of interest.
Two example applications are presented: The simulation of shortwave-infrared polarized spectra as measured by GOSAT from which CO2 is retrieved, and the simulation of the differential optical thickness in the visible spectral range which is derived from SCIAMACHY measurements to retrieve NO2. The computational speed of ALIS (for 1D or 3D atmospheres) is of the order of or even faster than that of one-dimensional discrete ordinate methods, in particular when polarization is considered.

Highlights

► Very efficient high spectral resolution solar radiative transfer simulations including polarization. ► New approach based on Monte Carlo method. ► Impact of cloud inhomogeneity on trace gas retrievals.

Introduction

Monitoring of atmospheric trace gases is important to understand atmospheric composition and global climate change. In particular, climate models require information about the concentration and global distribution of trace gases like, e.g. H2O, CO2, O3, or CH4. The trace gases can be observed by measuring solar radiation which is scattered and absorbed by the molecules. Several instruments have been developed: satellite instruments provide global observations, local measurements can be taken from the ground, from air-plane or from a balloon. Most instruments designed for trace gas concentrations observations measure radiance spectra with high spectral resolution. In the UV–Vis spectral range, absorption of radiation is due to molecular transitions; at the same time vibrational and rotational transitions can take place, which results in band spectra where the individual absorption lines cannot be distinguished. Nevertheless, each molecule type has its specific absorption features, so that the measured spectra include information about the various trace gas concentrations. In the near-infrared spectral range there are mainly vibrational transitions; here individual lines can be identified and used for trace gas measurements.
Examples for currently operating satellite instruments that measure high resolution radiance spectra of scattered solar radiation are SCIAMACHY on the ENVISAT satellite [1], OMI on AURA [2], GOME-2 on METOP [3] and TANSO-FTS on GOSAT [4]. SCIAMACHY and TANSO-FTS have the advantage of measuring not only the radiance but also the polarization state of the radiation. While extraterrestrial solar radiation is unpolarized, the radiance arriving at the satellite is polarized due to scattering by molecules, aerosols or clouds and due to surface reflection. The polarization information may therefore be used to reduce the uncertainties in trace gas retrievals introduced by aerosols, clouds and surface reflection.
The retrieval of trace gas concentrations from radiance spectra requires a so-called forward model, which can simulate such measurements for given realistic atmospheric conditions. For the often used optimal estimate retrieval method [5] it is important that the forward model is fast because it has to be run several times until iteratively the atmospheric composition is found which best matches the measured spectra.
A commonly used method to simulate solar radiative transfer is the discrete ordinate method which was first described by Chandrasekhar [6] and which has been implemented for instance into the freely available well-known software DISORT [7]. The DISORT code, however, has the limitations that it assumes a plane-parallel atmosphere (i.e. horizontal inhomogeneities cannot be taken into account) and that it neglects polarization. Polarization has been included in the VDISORT code [8]. The SCIATRAN code [9] is also based on the discrete ordinate method. It can optionally take into account spherical geometry as well as polarization [10].
Another method for the simulation of solar radiative transfer is the Monte Carlo method [11], [12], which is usually much slower than the discrete ordinate method. For this reason Monte Carlo methods have mostly been used for simulations including inhomogeneous clouds (e.g. [13]) for which the plane-parallel approximation cannot be applied. We have developed a new Monte Carlo method which calculates high spectral resolution radiance spectra very efficiently. The algorithm, named ALIS (Absorption Lines Importance Sampling), does not require any approximations, in particular it can easily take into account polarization and horizontal inhomogeneity. We show that the computational time of ALIS for high resolution radiance spectra is comparable to or even faster than the discrete ordinate approach, especially if polarization is included. This means that the algorithm is sufficiently fast to be used as forward model for trace gas retrieval algorithms. The basis of the ALIS method is that all wavelengths are calculated at the same time based on the same random numbers. This method which is sometimes called "method of dependent sampling" [11] has been used for various applications, e.g. to calculate mean radiation fluxes in the near-IR spectral range [14], to compute multiple-scattering of polarized radiation in circumstellar dust shells [15] or to calculate Jacobians [16]. We have validated ALIS by comparison to the well-known and well-tested DISORT program, originally developed and implemented by Stamnes et al. [7] in FORTRAN77. We use a new version of the code implemented in C [17] with increased efficiency and numerical accuracy.

Section snippets

Methodology

The new method Absorption Lines Importance Sampling (ALIS), which allows fast calculations of spectra in high spectral resolution, has been implemented into the radiative transfer model MYSTIC (Monte Carlo code for the phYsically correct Tracing of photons In Cloudy atmospheres; [18]. MYSTIC is operated as one of several solvers of the libRadtran radiative transfer package [19]. The common model geometry of MYSTIC is a 3D plane-parallel atmosphere to simulate radiances or irradiances in

Simulation of polarized near-infrared spectra in cloudless conditions

The Greenhouse Gases Observing Satellite (GOSAT) determines the concentrations of carbon dioxide and methane globally from space. The spacecraft was launched on January 23, 2009, and has been operating properly since then. Information about the project can be found on the web-page http://www.gosat.nies.go.jp. GOSAT carries the Thermal and Near-Infrared Sensor for Carbon Observation Fourier-Transform Spectrometer (TANSO-FTS) [4]) which measures in four spectral bands (band 1: 0.758–0.775 μm, band

Conclusions

We have developed the new method ALIS (absorption lines importance sampling) that allows to compute polarized radiances in high spectral resolution using the Monte Carlo method in a very efficient way. We sample random photon paths at one wavelength. For these random paths we calculate a spectral absorption weight using the wavelength dependent absorption coefficients of the model boxes. In order to correct for the wavelength dependence of the Rayleigh scattering an importance sampling method

Acknowledgments

We thank Timothy E. Dowling for translating the DISORT code from FORTRAN77 to C which resulted in great improvements regarding numerical accuracy and computation time. Furthermore, we thank Jerôme Vidot for providing NO2 profiles. This work was done within the project RESINC2 funded by the "Deutsche Forschungsgemeinschaft" (DFG).

References (42)

  • J.P. Burrows et al.

    Atmospheric remote-sensing reference data from GOME: part 1. Temperature-dependent absorption cross sections of NO2 in the 231–794 nm range

    J Quant Spectrosc Radiat Transf

    (1998)
  • Gottwald M, Bovensmann H, Lichtenberg G, Noel S, von Bargen A, Slijkhuis S, et al., SCIAMACHY, monitoring the changing...
  • P.F. Levelt et al.

    The ozone monitoring instrument

    IEEE Trans Geosci Remote Sensing

    (2006)
  • Callies J, Corpaccioli E, Eisinger M, Hahne A, Lefebvre A, GOME-2—Metop's second generation sensor for operational...
  • A. Kuze et al.

    Thermal and near infrared sensor for carbon observation Fourier-transform spectrometer on the greenhouse gases observing satellite for greenhouse gases monitoring

    Appl Opt

    (2009)
  • C. Rodgers

    Inverse methods for atmospheric sounding: theory and practice

    (2000)
  • S. Chandrasekhar

    Radiative transfer

    (1950)
  • Stamnes K, Tsay S-C, Wiscombe W, Laszlo I, DISORT, a general-purpose Fortran program for discrete-ordinate-method...
  • G.I. Marchuk et al.

    The Monte Carlo methods in atmospheric optics, Springer series in optical sciences

    (1980)
  • A. Marshak et al.

    3D radiative transfer in cloudy atmospheres

    (2005)
  • T. Zinner et al.

    Testing remote sensing on artificial observations: impact of drizzle and 3-D cloud structure on effective radius retrievals

    Atmos Chem Phys

    (2010)

    • Cited by (50)

    • Modeling polarized radiative transfer in the ocean-atmosphere system with the GPU-accelerated SMART-G Monte Carlo code

      2019, Journal of Quantitative Spectroscopy and Radiative Transfer
      Citation Excerpt :

      Absorption is treated afterwards using the Beer-Lambert law and the record of all photons’ bundle trajectories. During the photon ray-tracing, wavelength-dependent scattering correction factors are applied to the photon for compensating the fact that the photon path is determined at a unique wavelength: this method named ALIS is described in [41]. Trans-spectral processes can be treated easily by changing randomly the wavelength of the propagating photon.

    • IPRT polarized radiative transfer model intercomparison project - Phase A

      2015, Journal of Quantitative Spectroscopy and Radiative Transfer

    • New secondary-scattering correction in DISORT with increased efficiency for forward scattering

      2011, Journal of Quantitative Spectroscopy and Radiative Transfer
    View all citing articles on Scopus
    View full text
    Copyright © 2011 Elsevier Ltd. All rights reserved.