# Numerical study of multiple scattering intensities obtained from dust grains of different sizes and compositions in an impact-induced ejecta curtain

**Authors:** U. Keerthana, P. Shalima, Koji Wada, Johannes Markkanen, Hiroshi Kimura  
**arXiv:** [2610.23176](https://arxiv.org/abs/2610.23176)  
**Date:** 2026-10-16  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2610.23176v1/

## Key points

- Purpose: Investigate how grain size, composition, and optical depth shape impact-ejecta brightness, motivated by Hayabusa2’s Ryugu experiment.
- Method: Combine Mie calculations for spherical grains with three-dimensional Monte Carlo radiative transfer to model multiple scattering in an idealized ejecta curtain at a visible wavelength.
- Scope: Examine nine compositions, grain radii of 0.1–1,000 micrometers, several size distributions, and uniform versus height-dependent grain densities.
- Main finding: Brightness depends jointly on absorption, angular scattering, and optical depth—not scattering efficiency alone.
- Grain size and composition: Weakly absorbing materials generally peak in brightness at 0.1 micrometers, moderately absorbing materials at 0.1–0.5 micrometers, and strongly absorbing materials around 0.5 micrometers. Magnesium-rich olivine is particularly bright, while cometary dust is faint.
- Spatial patterns: Submicrometer grains produce U-shaped intensity profiles that flatten as grain size increases.
- Optical depth and scattering: Greater optical depth generally lowers detected brightness through extinction; larger grains scatter strongly forward, contributing less at the observation angles.
- Phase functions: Composition-specific scattering features are absent from the smooth Henyey–Greenstein approximation, highlighting the importance of material-specific modeling.
- Grain distributions: At fixed total number density, size distributions favoring smaller grains are brighter; a density that decreases with height lowers brightness while largely preserving composition differences.
- Application: The model offers a reference for interpreting ejecta observations and planning impact missions, not a definitive reconstruction of Ryugu’s ejecta.
- Limitations: Spherical grains, simplified geometry, a single wavelength, and incomplete temporal evolution restrict realism; fixed-number-density comparisons also involve different ejecta masses.

## Summary

This paper investigates how dust-grain size, composition, and optical depth affect the brightness of impact-generated ejecta curtains. These curtains expose subsurface material that is less altered by solar radiation, potentially revealing the composition and evolutionary history of planetary bodies. Motivated by Hayabusa2’s impact experiment on asteroid Ryugu, the authors assess whether multiple-scattering intensity maps can distinguish grain populations and how their predictions differ from those of simpler scattering approximations. The authors combine Mie calculations for homogeneous spherical particles with a three-dimensional Monte Carlo radiative-transfer model. Their idealized ejecta curtain is a hollow, inverted cone with a height and radius of 10 m, divided into cells measuring 0.1 m on each side. Parallel sunlight illuminates the cone at 45 degrees, while a detector observes it from approximately 1,000 m away. Calculations use a wavelength of 6,000 angstroms, corresponding to visible-light observations by Hayabusa2’s DCAM3 camera. The model considers nine compositions spanning weakly absorbing materials, including magnesium-rich olivine and lunar glass; moderately absorbing materials, including astronomical silicate, tholin, and organic refractory material; and strongly absorbing materials, including carbonaceous substances and iron. Grain radii range from 0.1 to 1,000 micrometers. For each grain population, the model calculates extinction and scattering cross-sections, single-scattering albedo, and angular scattering phase functions. It then tracks weighted photon packets through successive interactions, accounting for absorption and attenuation along the path to the detector. Each simulation uses 10 million photon packets to produce a 2,000-by-2,000-pixel intensity map. A representative convergence test showed that the median intensity nearly stabilized after one million packets. The authors compare monodisperse grains with power-law size distributions having indices of −2, −3, and −4. They examine both uniform number density and a number density that decreases with height according to a power-law index of −1.82. The results show that brightness depends jointly on grain absorption, angular scattering, and optical depth, rather than on scattering efficiency alone. Weakly absorbing materials generally produce their brightest curtains at a grain radius of 0.1 micrometers, with brightness decreasing toward millimeter-sized grains. Moderately absorbing materials peak at approximately 0.1–0.5 micrometers, whereas strongly absorbing materials peak around 0.5 micrometers and then decline more gradually. Magnesium-rich olivine is particularly bright, while cometary dust produces low intensities. Submicrometer grains also generate U-shaped spatial intensity profiles that flatten as grain size increases. Increasing optical depth generally reduces the median detected intensity because scattered photons undergo greater extinction before reaching the observer. Larger grains also scatter strongly forward, reducing their contribution at the approximately 130–140-degree scattering angles relevant to the observations. The calculated phase functions exhibit composition-dependent interference features absent from the smooth Henyey–Greenstein approximation, underscoring the importance of material-specific phase functions. For size distributions with a fixed total number density, more negative power-law indices favor smaller grains and yield higher median intensities. Introducing a vertical density gradient reduces overall brightness but broadly preserves differences among compositions. A qualitative comparison with a DCAM3 image taken 98 seconds after impact yields predicted median intensities of approximately 3 × 10^13 and 9 × 10^13 photons per square meter per second per steradian per angstrom for 1 mm astronomical-silicate and magnesium-rich-olivine grains, respectively. The authors conclude that their model provides a useful reference for interpreting ejecta observations and planning future impact missions, rather than a definitive reconstruction of Ryugu’s ejecta. Important limitations include the assumption of spherical rather than aggregate grains, simplified geometry, the use of a single wavelength, and an incomplete treatment of temporal evolution. Comparisons at fixed number density also imply different ejecta masses, complicating efforts to attribute brightness differences solely to grain properties.