Fair Evaluation of Orientation‐Averaging Techniques in Light‐Scattering Simulations: Comment on "Evaluation of Higher‐Order Quadrature Schemes in Improving Computational Efficiency for Orientation‐Averaged Single‐Scattering Properties of Nonspherical Ice Particles" by Fenni et al. Issue 2 (17th January 2023)

• Record Type: Journal Article
• Authors: Yurkin, Maxim A.
• Is Part Of: Journal of geophysical research. Volume 128:Issue 2(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-01-17
• Subjects: light scattering -- discrete dipole approximation -- orientational averaging -- cubature -- error control; Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1029/2021JD036088
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: In a recent paper Fenni et al. (2021, https://doi.org/10.1029/2020jd034172 ) compared the code MIDAS, based on the direct solution of the volume‐integral equation combined with advanced cubatures for orientation averaging, to the code DDSCAT, a state‐of‐the‐art implementation of the discrete dipole approximation. This comment highlights methodological issues in this comparison and shows that the quantitative claims of Fenni et al. (2021, https://doi.org/10.1029/2020jd034172 ), related to superiority of MIDAS over DDSCAT, are based on very specific test cases with respect to particle symmetries or initial orientation, as well as to the selected scattering quantity of interest. Thus, these claims are not expected to hold for other similar particles. Moreover, the detailed discussion of these issues is relevant for all light‐scattering simulation methods, except those allowing analytical orientation averaging. Thus, the comment constructs general guidelines for fair evaluation of orientation‐averaging techniques in a wide range of light‐scattering methods and computer codes. Plain Language Summary: The paper discusses several issues that appear when one is comparing different orientation‐averaging techniques (cubatures) in combination with the same or different light‐scattering simulation methods. Fair evaluation of cubature performance in realistic general scenarios is important both for practitioners (to choose the most efficient combination of the existing codes and cubatures) and for code developers (to set their priorities on the new features with the largest expected benefits). Unfortunately, the performance of the cubatures is complexly interwoven with the internals of the simulation methods and depends on specific test particles and computed scattering quantities. This questions the generality of conclusions in some previous publications. Based on this discussion, the paper ends with general guidelines for fair evaluation of cubatures, allowing future studies to arrive at general conclusions, so that they can be directly used by other researchers. Key Points: Quantitative conclusions of Fenni et al. (2021) are based on very specific test cases Orientation‐averaging techniques should be compared on non‐symmetric particles, and not with a special initial orientation Any comparison of simulated results should consider their uncertainties accounting for all sources of errors