Efficient design of the realization scheme of the invariant imbedding (IIM) T-matrix light scattering model for atmospheric nonspherical particles

2020 
Abstract The invariant imbedding T-matrix method has been recognized as one of the rigorous computational tools to simulate the light scattering by atmospheric nonspherical particles after the efficient implementation by L. Bi and P. Yang. Compared with the T-matrix model based on the Extended Boundary Condition Method (“EBCM”), it is suitable for the light scattering simulation of particles with any shapes. However, since the T-matrix must be updated in each iterative process, its computational efficiency is relatively lower than EBCM, especially for the particles with large sizes and large aspect ratios. To this end, an efficient realization scheme of the IIM T-matrix method is proposed in this paper. The principle and the structure of our IIM T-matrix code is firstly introduced, and then the symmetric properties of the U-matrix (an important matrix in the IIM T-matrix method) is derived; to simplify the iteration of T-matrix calculation, the implicit matrix inversion scheme is also proposed. To validate the effectiveness of these improvements, the computational efficiency and accuracy is systematically investigated for the particles with different shapes and sizes. The simulation results indicate that the computational time is reduced notably after the model is improved by the symmetry of U-matrix and the implicit matrix inversion scheme, where for the spheroid with (a, b) = (2.5 µm, 3.0 µm) at λ=0.5 µm (the particle is treated as complex geometries without any symmetry), the computational efficiency is improved by nearly 40%. Excellent agreement is also achieved between the scattering parameters obtained by the IIM T-matrix method and the well-test scattering models (such as BECM, DDASCAT, MRTD and IGOA, etc.), indicating that the IIM T-matrix method can simulate the light scattering by nonspherical particles effectively.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    62
    References
    1
    Citations
    NaN
    KQI
    []