Evidence against non-gravitational acceleration of 1I/2017 U1 ‘Oumuamua

  • J. I. KatzEmail author


Micheli et al. (Nature 559:223, 2018) reported that a seven-parameter fit to the orbit of 1I/2017 U1 ‘Oumuamua indicated a non-gravitational acceleration in the anti-Solar direction, and attributed it to recoil from comet-like outgassing. The implied gas to dust ratio is at least 100 times greater than that of known Solar System comets. The reported collapse of the scatter of nearly contemporaneous coordinate residuals upon inclusion of the non-gravitational term in the orbital fits is difficult to understand. There are grounds for skepticism.


Asteroids Comets 

Meech et al. (2017) discovered the interstellar object 1I/2017 U1 (‘Oumuamua) and estimated an upper limit of \(1.7 \times 10^{-3}~\mbox{kg}/\mbox{s}\) on its rate of outflow of dust, describing it as an asteroid rather than a comet. However, Micheli et al. (2018) added a possible non-gravitational force, directed away from the Sun with an inverse square law dependence on distance, to the orbital fit and found a significant amplitude, effectively multiplying the Solar gravitational acceleration by a factor of 0.99917. Such a force might be expected if ‘Oumuamua is outgassing under the influence of Solar radiative heating, although a quantitative inverse square law might not be expected because of the complexity of heat transfer in a transiently heated body.

From this result Micheli et al. (2018) inferred an outgassing rate of 3.6 kg/s, implying a ratio of dust to gas in the outflow \({<} 0.5 \times 10^{-3}\). This should be compared to the dust to gas ratios of Solar System comets that are 0.1–1 (Singh et al. 1992; Sanzovo et al. 1996).

Micheli et al. (2018) also set an upper bound of 1 kg on the dust present within a cylinder of radius \(r = 750~\mbox{km}\) (\(2.5^{\prime \prime }\)) projected separation. Using their outgassing rate of \({\dot{M}}_{\mathit{gas}} = 3.6~\mbox{kg}/\mbox{s}\) and assuming roughly spherical symmetry of outgassing and a plausible ejection speed \(v = 300~\mbox{m}/\mbox{s}\) leads to a mass of gas in this cylinder
$$ M_{\mathit{gas}} = \frac{{\dot{M}} }{4 \pi } \int \,d\varOmega \frac{r }{v \sin \theta } = \frac{\pi }{2} \frac{{\dot{M}} r }{v} = 1.4 \times 10^{4}\ \text{kg}, $$
where \(\theta \) is the angle from the direction to the observer. This would imply a dust to gas ratio \({<} 10^{-4}\), even less than that inferred by comparing their outgassing rate to the bound on the dust outflow rate of Meech et al. (2017).

The claimed outgassing rate would imply that ‘Oumuamua consists of extraordinarily clean ices, unlike any known Solar System cometary body. This is a reason for skepticism of the reported non-gravitational acceleration. Micheli et al. (2018) have suggested that the gas flow might contain comparatively large solid particles that are inefficient scatterers, thus maintaining a solid to gas ratio comparable to that of Solar System cometary bodies while producing very little visible coma. This cannot be excluded empirically, but the phenomenological difference is in the ratio of fine dust to gas; even if these larger particles are present, ‘Oumuamua would still differ from Solar System bodies in its ratio of fine dust to gas. The possibility of such larger particles is constrained by infrared observations (Trilling et al. 2018).

An additional reason for skepticism is that Fig. 2 of Micheli et al. (2018) shows a scatter in the residuals to their 6-element (purely gravitational) orbital fit in observations within a single night or on consecutive nights of \({\sim} 3\) times their formal uncertainty. Adding the seventh fitting parameter, the magnitude of the non-gravitational force, collapses this scatter by at least a factor of three. This would not be expected, even were the non-gravitational forces correct, because they are a smoothly varying function of time. Including it in the fit (and optimizing over all seven parameters) would not be expected to collapse the scatter of nearly contemporaneous observations—the fitted orbit, would, at best, accurately agree with the mean of such observations, but would not be expected to reduce their scatter.

The validity of the fitting procedure can be tested by numerical experiment: Add random noise with dispersion three times the formal uncertainty to the six observations obtained on UTC 2017-10-19. Then perform the six and seven parameter fits to the entire dataset, including these test “data”. If the fitting procedure is valid, the scatter of the contemporaneous data about the best fit solutions should not collapse in the seven parameter fit because the test “data” would not be consistent with the physical model (gravity plus outgassing). If the scatter collapses, the fitting procedure requires further investigation.

Nongravitational acceleration would be explicable if the column density of ‘Oumuamua is \(\sim 0.1\text{--}1~\mbox{g}/\mbox{cm}^{2}\). Bialy and Loeb (2018) have suggested that it is a thin artificial structure, although the inferred column density would be orders of magnitude greater than that of engineered “Solar Sails”. Moro-Martín (2019) suggested it is a fractal aggregate of density \({\sim }10^{-5}~\mbox{g}/\mbox{cm}^{3}\), although such an object would be extremely fragile and would have much lower density than the microscopic fractal aggregates observed in the Solar System.

The arguments presented here are independent of that of Rafikov (2018) but lead to the same conclusion: It is unlikely that ‘Oumuamua has a non-gravitational acceleration as large as that reported by Micheli et al. (2018).



I thank S. Bialy, S. Kenyon and A. Loeb for useful discussions.

Compliance with Ethical Standards

The author has no potential conflicts of interest.


  1. Bialy, S., Loeb, A.: Astrophys. J. Lett. 868, L1 (2018) ADSCrossRefGoogle Scholar
  2. Meech, K.J., et al.: Nature 552, 378 (2017) ADSCrossRefGoogle Scholar
  3. Micheli, M., et al.: Nature 559, 223 (2018) ADSCrossRefGoogle Scholar
  4. Moro-Martín, A.: Astrophys. J. Lett. 832, L32 (2019). ADSCrossRefGoogle Scholar
  5. Rafikov, R.R.: Astrophys. J. Lett. 867, L17 (2018) ADSCrossRefGoogle Scholar
  6. Sanzovo, G.C., Singh, P.D., Huebner, W.F.: Astron. Astrophys. Suppl. 120, 301 (1996) ADSCrossRefGoogle Scholar
  7. Singh, P.D., de Almeida, A.A., Huebner, W.F.: Astron. J. 104, 848 (1992) ADSCrossRefGoogle Scholar
  8. Trilling, D.E., Mommert, M., Hora, J.L., et al.: Astron. J. 156, 261 (2018) ADSCrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.Department of Physics and McDonnell Center for the Space SciencesWashington UniversitySt. LouisUSA

Personalised recommendations