Investigating Possible Spindown of Arrokoth by Collisions with Small Classical Kuiper Belt Objects

Introduction: The New Horizons flyby of Arrokoth revealed an ancient, contact binary planetesimal [1,2]. Arrokoth’s both lobes’ respective principal axes are aligned within a few degrees [2], and such configuration suggests a co-orbiting Arrokoth before the coalescence of its two lobes [3]. One mechanism proposed for Kuiper belt object (KBO) binaries to merge into a bilobate body is a random walk due to collisions with other heliocentric bodies [3,4]. For Arrokoth, one thing that remains resolved is that its present-day spin period (15.92 hr) is slower than that predicted from both lobes’ mutual gravitational pull (11.26 hr), assuming a comet-nucleus-like density of 500 kg m^-3 [3], implying ~30% angular momentum loss. While Arrokoth may simply be less dense, it is worth exploring whether collisions with other KBOs could have substantially altered its spin state over time. Here we adapt our Monte Carlo impact simulation for Ceres and Vesta [5] and investigate Arrokoth’s possible spindown (or spinup) by impacts.

Triaxial Arrokoth and model results: We previously carried out Monte Carlo impact simulations with random impacts onto an Arrokoth modeled as an oblate spheroid [6]. Here we model Arrokoth as a triaxial ellipsoid by matching its cross-sectional areas along the principal axes with that from its shape model. We also scale this triaxial body’s density for its moment-of-inertia (MOI) to match that of Arrokoth. As a result, this model geometry approximates what the actual Arrokoth would undergo, regarding its dynamical evolution by a given flux of impactors. In this way we avoid the unnecessary complications for trying to track the ejecta interactions on a truly bilobate object.

We base our range of impactor sizes from Arrokoth’s measured craters [2] along with impactor-crater scaling laws [7,8]. While the lower bound is well-determined near 10-m, the upper bound is less well constrained. Both scalings [7,8] predict that the largest crater “Maryland” (~7-km wide, [2]) could have been created by an impactor ~1-to-2-km wide with a typical impact speed ~300 m/s. The total number of impactors is extrapolated from the crater counts [2] where 40-50 craters or pits were recognized during the flyby on one side of Arrokoth. Hence, we allow for 100 impacts between 10 and 1000 m in our simulations (we vary the upper limit later), assuming dN/dD ~ D^-1.75  [2,7], where N(>D) is the number of impactors with diameters greater than D. Implicit in our modeling is the assumption that Arrokoth's craters postdate its (plausibly very early) merger [3]. We also track a disruption energy threshold for porous asteroids [9] to test for potential catastrophic breakup.

Figure 1 consists the results after 5000 Monte Carlo simulations starting at 11.26 hr, with impactors from cold classical KBOs (CCKBOs). No disruption is predicted among all our simulations. About 80% of the simulations end within 11.2 ± 0.2 hr (1σ). Only 2% of the runs have a final spin increase or decrease by more than half an hour. The total, integrated impactor mass is only 0.01 ± 0.01% (1σ) of M, while the total mass loss ejected is 3.8 ± 1.6 times (1σ) larger [10], indicating that Arrokoth is losing mass almost all the time after these 100 impacts. This finding is opposite to what we have found on Ceres or Vesta [5] where a porous surface tends to retain mass rather than losing mass overall; this is expected because the escape velocity of Arrokoth is ~5 m/s, much lower than a typical impact velocity among KBOs [7].

Even though the majority of the impactor flux onto Arrokoth comes from CCKBOs [7], other subpopulations of KBOs also contribute to its overall spin history. Based on pre-encounter model [7], we randomly select ~11% of the impactors to be hot classicals (with a higher impact speed, see Fig. 2); an extra set of 5000 simulations is implemented (Fig. 3). The incorporation of hot classicals delivers notable differences. Due to slightly increased overall impact speed, ejecta loss is further enhanced (about 6.2 ± 5.2 times (1σ) larger than total impactor mass), albeit the net mass change is still ~0.1% M. Whereas the majority of the simulation runs clusters within 1σ from the average, more than 6% experience net spin changes greater than 0.5 hr; indeed, a few runs even achieve the required spindown from 11.26 hr to 15.92 hr. Therefore, Arrokoth’s potential spindown by impacts has a higher likelihood under this condition, although this possibility is still quite low from a statistical point of view.

Discussion: Impacts are shown to play a potentially important role in Arrokoth’s spin angular momentum evolution over time. Our simulations show that major changes of a few hours in spin period, solely by impacts, are highly unlikely, but neither should one assume Arrokoth’s present-day rotation is primordial (or fixed post-merger). The most important role for heliocentric impacts probably occurred when Arrokoth was a co-orbiting binary. The greater the binary separation, the greater the relative angular momentum input to the system for a given impact, the cumulative effect of which could be important for ultimate binary merger. A prior, pre-merger formation of “Maryland” crater could have been especially important for Arrokoth’s angular momentum evolution.

Acknowledgments: This research supported by NASA’s New Horizons project.

References: [1] Stern S.A. et al. (2019) Science 364, eaaw9771. [2] Spencer J.S. et al. (2020) Science 367, aay3999. [3] McKinnon W.B. et al. (2020) Science 367, aay6620.  [4] Nesvorný D. et al. (2018) AJ, 155, 246. [5] Mao X. and McKinnon W.B., MAPS, in revision. [6] Mao X. et al. (2020) 51th LPSC, abs. #2592. [7] Greenstreet S. et al (2019) Astrophys. J. Lett. 872, L5. [8] Housen K.R. et al. (2018) Icarus 300, 72-96. [9] Holsapple K.A. and Housen K.R. (2019) Planet. Space Sci. 179, 104724. [10] Housen K.R. and Holsapple K.A. (2011) Icarus 211, 856-875.