Bennu's and Ryugu's top shapes, which exhibit oblate shapes with equatorial ridges, are considered to have evolved by fast rotation in a longer timescale after the reaccumulation-driven formation. Analyses have suggested different evolution processes that enhanced their top shapes. Bennu's shape may have been driven by surface processing, while Ryugu's may have resulted from large deformation due to internal failure. These two scenarios sound inconsistent. Here, we attempt to connect these different processes by (1) revisiting the earlier results from the finite element model (FEM) approach and (2) applying a semi-analytical approach that computes the spatial distribution of the minimum cohesive strength that allows a body element to resist structural failure. In the semi-analytical approach, an asteroid is assumed to be a uniformly rotating triaxial ellipsoid. In both approaches, the structure is considered to be uniform. We investigate the variations in the failure modes of the bodies at different spin periods.

The approaches consistently show that if the bulk cohesive strength is minimal but non-zero (< 10 Pa), the failure mode changes at a spin period shorter than 4.5 h. At a spin period of 4 h─4.5 h, the failed region only appears in a shallow surface layer in the equatorial region. As the spin period becomes shorter, the failure region becomes wider and spreads to the interior. When the spin period is shorter than 3.5 h, the interior becomes the most sensitive in the entire body. The results show that Ryugu's 7.6 h spin period may not significantly induce mass movement, while Bennu's 4.3 h spin period, within the surface failure condition, supports observational evidence of recent mass movement (Walsh et al., 2019; Jawin et al., 2020). These two failure modes infer that top shapes may evolve by surface processing and large deformation-driven internal failure. This explanation supports the arguments that Ryugu's top shape evolved due to relatively fast rotation, while Bennu's top shape was enhanced by recent surface processing.

This work was recently published in Icarus (10.1016/j.icarus.2020.113946). M.H. and R.N. thank support from NASA/SSW (NNH17ZDA001N) and AU/IGP. ANSYS Mechanical APDL 18.1 was used in this work. P.M. acknowledges support from CNES, from the EU's Horizon 2020 under No 870377 (NEO-MAPP) and from the IDEX JEDI of the Université Côte d'Azur. S.S. acknowledges support from JSPS Core-to-Core Program, "International Network of Planetary Sciences."

1: Walsh et al. (2019), Nature Geoscience, 12, 242-246 ; 2: Jawin et al. (2020), LPSC, 2326, 1201.