Rationale and Objective: Understanding the composition and internal structure of asteroids and protoplanets like (4) Vesta is crucial for modeling early solar system planet formation 14, 8 and cataloging asteroids for future research, most prominently in de- veloping fields such as asteroid mining and planetary defense. The surface composition of these bodies, determined by spectral remote sensing, must be quantitatively recon-ciled with the bulk properties of the bodies to infer their interior structure 5. This study’s objective is to predict the crustal grain density of (4) Vesta based on its global Visible and Near-Infrared (VNIR) spectrum and compare it to the known bulk density, demonstrating the viability of this method for future application to a larger dataset. Novelty and Method: The work applies a high-rigor methodology that addresses two critical challenges in spectral remote sensing: the non-linearity of light scattering in particulate regolith, which is solved by applying the Hapke (1993) radiative transfer model 6, 9 to invert reflectance to linearly-additive single-scattering albedo, and the spectral redundancy of large endmember libraries, which is resolved through heuristic filtering and Principal Component Analysis (PCA) 7. This suite is input into the Hapke (1993) model to yield endmember mass ratios (fi). The resulting composition and literature- derived endmember densities (ρi) are used to calculate the crustal grain density (ρgrain) via the harmonic mean, subject to a full error propagation 2.Results and Contribution: The spectral unmixing yields a crustal composition of primarily low-Ca pyroxene (45%) and high-Ca pyroxene (30%). Calculating error propagation results in a crustal grain density of ρgrain = 3.26 ± 0.07 g/cm3. This value is statistically distinct from the bulk density (ρbulk = 3.46±0.01 g/cm3) measured by the Dawn mission 14, 13. Conclusion: The positive density differential, ∆ρ = ρbulk−ρgrain = 0.20±0.07 g/cm3, is statistically significant beyond 2σ. This paper provides quantitatively robust evidence, supported by full uncertainty analysis, that the interior of (4) Vesta must possess a composition or structure (e.g., a high-density core) distinct from that of its surface. The general methodology introduced here serves as a replicable framework for inferring internal structure from remote sensing data for any solid body.
Aidan Gray (Sun,) studied this question.