Letter | Published:

# Spectroscopy and thermal modelling of the first interstellar object 1I/2017 U1 ‘Oumuamua

Nature Astronomyvolume 2pages133137 (2018) | Download Citation

## Abstract

During the formation and evolution of the Solar System, significant numbers of cometary and asteroidal bodies were ejected into interstellar space1,2. It is reasonable to expect that the same happened for planetary systems other than our own. Detection of such interstellar objects would allow us to probe the planetesimal formation processes around other stars, possibly together with the effects of long-term exposure to the interstellar medium. 1I/2017 U1 ‘Oumuamua is the first known interstellar object, discovered by the Pan-STARRS1 telescope in October 2017 (ref. 3). The discovery epoch photometry implies a highly elongated body with radii of ~ 200 × 20 m when a comet-like geometric albedo of 0.04 is assumed. The observable interstellar object population is expected to be dominated by comet-like bodies in agreement with our spectra, yet the reported inactivity of 'Oumuamua implies a lack of surface ice. Here, we report spectroscopic characterization of ‘Oumuamua, finding it to be variable with time but similar to organically rich surfaces found in the outer Solar System. We show that this is consistent with predictions of an insulating mantle produced by long-term cosmic ray exposure4. An internal icy composition cannot therefore be ruled out by the lack of activity, even though ‘Oumuamua passed within 0.25 au of the Sun.

## Main

Following the announcement of the discovery, we performed spectroscopic observations at two facilities. The 4.2 m William Herschel Telescope (WHT) on La Palma was used with the ACAM auxillary port imager and spectrograph on 25 October 21:45 ut–22:03 ut. An initial analysis of this spectrum revealed an optically red body5. Spectra were also obtained using the X-shooter spectrograph on the European Southern Observatory 8.2 m Very Large Telescope (VLT) on 27 October 00:21 ut–00:53 ut, covering 0.3–2.5 μm. Observation circumstances are given in Table 1 and the resulting binned reflectance spectra at optical wavelengths are shown in Fig. 1.

Active comets possess strong molecular emission bands via electronic transitions within the vibrational ground state due to fluorescence of CN at 0.38 μm and C2 at 0.52 μm6. Although our spectra are noisy, no such emission is seen, in concordance with imaging reports of an apparently inert body3,7,8,9. Asteroid spectra can show significant solid-state absorption features in this region depending on their mineralogy, notably a wide shallow absorption centred at ~0.7 μm due to phyllosilicates (aqueously altered silicates)10. Mafic minerals seen in asteroids (typically pyroxines and olivines) exhibit an absorption band starting at ~0.75 μm and centred at ≥0.95 μm11. Again, no such diagnostic features are observed.

Over the range 0.4 μm ≤ λ ≤ 0.9 μm, the reflectance gradients are 17.0 ± 2.3%/100 nm (one standard deviation) and 9.3 ± 0.6%/100 nm for the ACAM and X-shooter data, respectively. Additional measurements of the spectral slope have been reported from the Palomar Observatory as 30 ± 15%/100 nm over 0.52 μm ≤ λ ≤ 0.95 μm on October 25.3 ut 12, and 10 ± 6%/100 nm over 0.4 μm ≤ λ ≤ 0.9 μm on October 26.2 ut 9. The published photometric colours range from somewhat neutral to moderately red3,8,13,14. While most of these measurements are similar within their uncertainties, the reported (g − r) = 0.47 ± 0.04 is relatively neutral8, while we have a significant red slope in this region. Within our own data, our spectra differ in slope by >3σ. This is due to the ACAM spectrum being redder than the X-shooter spectrum at 0.7 μm ≤ λ ≤ 0.9 μm, with the mean reflectance increasing to 42% and 21% relative to 0.55 μm, respectively.

The measured rotation period is probably in the 7–8 h range based on photometry from different observers7,8,13,14. The most complete reported lightcurve is consistent with a rotation period of 7.34 h and an extremely elongated shape with an axial ratio of ~10:1 and a 20% change in minimum brightness, possibly due to hemispherically averaged albedo differences3. Using this rotation period, our spectra are separated by 0.66 in rotational phase and near opposing minima in the lightcurve. This implies that our spectra viewed different extrema of the body and supports the existence of compositional differences across the surface. We note the October 25 Palomar spectrum would have been obtained during lightcurve maximum, while the October 26 Palomar spectrum would have been near the same rotational phase as our WHT spectrum. Comparable spectral slope variations with rotation have been detected in ground-based data on a few S-type asteroids15 and trans-Neptunian objects (TNOs)16, although these objects are significantly larger than 1I/2017 U1.

The X-shooter spectrum contained a weak but measurable signal at 1.0 ≤ λ ≤ 1.8 μm. Beyond 1.8 μm, the sky background is much brighter than the object. We therefore excluded this spectral region at longer wavelengths from further analysis. In Fig. 2, we show the ACAM and X-shooter spectra, binned to a spectral resolution of 0.02 μm at λ > 1 μm. Although the signal-to-noise is low, it is apparent that the reflectance is relatively neutral in this spectral region; a weighted least-squares fit gives a slope of −1.8 ± 5.3%/100 nm at these near-infrared (NIR) wavelengths. There is a suggestion of decreasing reflectance beyond 1.4 μm, but the uncertainties are large due to the very weak flux from the object. There is no apparent strong absorption band due to water ice at 1.5 μm, as observed on some large TNOs. The only other reported NIR data are J-band photometry (1.15–1.33 μm) from October 30.3 ut 8, where (r − J) = 1.20 corresponds to a slope of 3.6%/100 nm. Our spectrum gives a larger slope of 7.7 ± 1.3%/100 nm over 0.63–1.25 μm. Again assuming a rotation period of 7.34 h would give a rotational phase difference of 0.4, indicating a small change in optical-infrared reflectance properties around the body.

Comparing our spectra with the reflectance spectra for different taxonomic classes of asteroid in the main belt and trojan clouds17, the closest spectral analogues are L- and D-type asteroids (Fig. 3). L-type asteroids are relatively rare in the asteroid belt. They exhibit a flattened or neutral spectrum beyond 0.75 μm and sometimes weak silicate absorption bands, indicating a small amount of silicates on their surfaces. These bands are not strong enough to be visible in our data. D-type asteroids form the dominant populations in the outer asteroid belt and Jupiter trojans. Most D-type reflectance spectra exhibit red slopes out to at least ~2 μm, in disagreement with our spectra, although some show a decrease in the spectral slope at λ > 1 μm, similar to 1I/2017 U118.

Looking at both trojan asteroids and more distant bodies beyond 5 au we find a good match in spectral morphology with 1I/2017 U1 as shown in Fig. 3. The spectral slopes of cometary nuclei tend to be red in the visible range but shallower in the NIR19. Some TNOs also exhibit a red optical slope but a more neutral NIR reflectance20. We show a spectrum of the large active centaur (60558) Echeclus, whose optical slope falls between our ACAM and X-shooter spectra, demonstrating similar behaviour of a red optical slope that decreases in the NIR.

The reddish optical spectra of D-type asteroids, cometary nuclei and TNOs are believed to be a result of irradiated organic-rich surfaces. The spectra presented here would place 1I/2017 U1 in the less red class of dynamically excited TNOs21. Irradiation of carbon-rich ices produces refractory organic residues with a wide range of slopes depending on original composition but consistent with the diversity of slopes observed in the outer Solar System22. To produce such changes in the optically active upper micron of surface only requires exposure to the local interstellar medium of <107 years23. Hence, we conclude that the surface of 1I/2017 U1 is consistent with an originally organic-rich surface that has undergone exposure to cosmic rays.

It was expected that discovered interstellar objects (ISOs) would be mostly icy objects due to both formation and observational biases. Planet formation and migration can expel large numbers of minor bodies, most of which would contain ices because they originated beyond the snow-line in their parent systems and would be ejected by the giant planets that form quickly in the same region1. Additionally, ISOs will have been produced from Oort clouds via the loss mechanisms of stellar encounters and galactic tides2. Our Oort cloud is expected to hold 200 to 10,000 times as many ‘cometary’ bodies than asteroidal objects24 and we assume that exo-planetary systems’ Oort clouds may form and evolve in a broadly similar manner. Therefore, both ISO production mechanisms should produce a population dominated by ice-rich bodies.

In terms of discovery, active comet nuclei are much easier to detect than asteroids of the same diameter; their dust comae make them visible over much greater distances and are more likely to attract follow-up observations that would establish their ISO nature. Before the discovery of 1I/2017 U1, ISO models suggested that the typical discovered asteroidal ISO would have a perihelion distance of q < 2 au while the typical cometary ISO would have perihelia 2 to 3 times larger, because they can be detected at greater distances25. Thus, the combination of the ISO production process and strong observational bias towards detecting active cometary ISOs makes the 1I/2017 U1 discovery particularly surprising. However, its perihelion distance, eccentricity and inclination are in excellent agreement with the predicted orbital elements of detectable asteroid-like ISOs25.

Given the spectral similarity with presumed ice-rich bodies in our Solar System, it might be expected that 1I/2017 U1 would have been heated sufficiently during its close (q = 0.25 au) perihelion passage to sublimate sub-surface ices and produce cometary activity. However, it has been shown that cosmic-ray irradiation of organic ices plus heating by local supernovae can produce devolatilized carbon-rich mantles26. Estimates of the thickness of this mantle range from ~0.1 m to ~2 m (ref. 4). Assuming that this object has such a mantle, we have modelled the thermal pulse transmitted through the object during its encounter with our Sun, assuming a spin obliquity of 0° and physical parameters that would be expected for a comet-like surface (see Methods). We find that the intense but brief heating 1I/2017 U1 experienced around perihelion does not translate into heating at significant depth. As shown in Fig. 4, the heat wave passes only slowly into the interior and, while the surface reached peak temperatures of ~600 K, H2O ice buried >20 cm deep would only commence sublimation weeks after perihelion. Layers 30 cm deep or more would never experience temperatures high enough to sublimate H2O ice. Taking the unphysical extreme of a surface continuously exposed to the Sun during the orbit only increases the depth of the ice sublimation layer by ~10 cm. Therefore, we conclude that if there is no ice within ~40 cm of the surface, we would expect to see no activity at all, even if the interior has an ice-rich composition. Simple thermal approximations give a similar surface temperature and thermal skin depth14.

Would a body with interior ice have significant strength to resist rotational disruption? Assuming a low density of ≤1,000 kg m−3, the required strength is estimated to be in the range 0.5–3 Pa3,13. Weak materials like talcum powder have a strength of ~10 Pa, sufficient to maintain the body structure. The inactive surface of comet 67P had a tensile strength ranging from 3–15 Pa27. Therefore, the unusual shape of 1I/2017 U1 does not rule out an internal ice-rich comet-like composition.

We recognize one obvious problem with this model—that Oort cloud comets should have undergone similar mantling due to cosmic-ray exposure over 4.6 Gyr, yet many show significant activity via sublimation of near-surface ice during their first perihelion passage28. 1I/2017 U1 cannot have had a significantly longer exposure to cosmic rays; even if it was formed around one of the earliest stars, it will not be more than ~3 times the age of our Solar System. More likely, 1I/2017 U1 dates from the more recent generations of stars as it could not be formed before the Universe had created enough heavy elements to, in turn, form planetesimals29. It may have become desiccated through sublimation of surface ices during close passages to its parent star before being ejected from its natal system. Damocloid objects in our own Solar System are thought to be similar cometary bodies that have developed thick insulating mantles preventing sublimation30. Alternatively, the cause could be the relatively small size of 1I/2017 U1 compared with active Oort cloud nuclei with radii of ≥1 km. The possible minimum radius of only ~20 m may have allowed most of the interior ice to escape over its unknown history. In this case, we should expect that the Large Synoptic Survey Telescope will find many small devolatized ‘comets’ from our own Oort cloud, in addition to more ISOs like 1I/2017 U1.

## Methods

### Observations

The apparent magnitude and position of 1I/2017 U1 relative to the Earth and Sun at the time of the two sets of observations are given in Table 1. Details of the instrument setup for each observation are given below. At both telescopes, the observations were performed by observatory staff in service mode. Each set of data was subsequently independently reduced by two of the authors; intercomparison of the resulting spectra showed no significant differences for the individual instruments.

For WHT, two 900 s exposures were obtained with ACAM31 in spectroscopic mode using a slit width of 2 arcsec at the parallactic angle. Subsequent inspection of the data showed that the second spectrum was contaminated by a late-type star passing through the slit, hence only the first spectrum was usable. The reflectance spectrum was obtained though division by a spectrum of the fundamental solar analogue 16 Cyg B taken directly afterwards with the same instrumental setup. Flux calibration was performed via a spectrum of the spectrophotometric standard BD + 25 4655 obtained through a 10-arcsec-wide slit.

For VLT, X-shooter contains three arms covering the ultraviolet/blue (UVB), visible and NIR spectral regions, separated by dichroic beam-splitters to enable simultaneous observation over the 0.3–2.5 μm range32. Four consecutive exposures were obtained, with 900 s exposures in the UVB and NIR arms and 855 s exposure in the visible arm (as the UVB and visible arms share readout electronics, this allows the most efficient use of the telescope while maximizing the flux in the low-signal ultraviolet region). It was found that the signal in the last two exposures was very poor and these were not used in the analysis. Subsequent matching with published photometry shows we were near lightcurve minimum at that time3, potentially explaining the drop in flux. Slits with widths 1.0, 0.9 and 0.9 arcseconds were used in the UVB, visible and NIR arms, respectively, all of which were aligned with the parallactic angle at the start of the observations. Observations of the solar analogue star HD 1368 were obtained with the same setup to allow calculation of reflectance spectra. Flux calibration was performed via observations of the spectrophotometric standard LTT 7987 obtained through a 5-arcsec-wide slit.

For the spectra from both facilities, the reflectance spectra were calculated from the median reflectance in spectral bins. There was enough flux at λ < 1 μm to allow binning over 0.01 μm bins in wavelength, but in the NIR the detected flux was so low bins had to be increased to 0.02 μm to obtain a reasonable spectrum. A robust estimation of the dispersion of the original spectral reflectance elements in each wavelength bin was performed using the ROBUST_SIGMA routine in IDL or equivalent code in Python. The reflectance uncertainty in each bin was then calculated by dividing by the square root of the number of original spectral elements in the bin.

### Spectrum comparison

In Figs. 2 and 3, we compare our observed spectra of 1I/2017 U1 with various Solar System minor bodies. Spectral types for asteroids are taken from the Bus–DeMeo taxonomy definitions established in ref. 17 and available at http://smass.mit.edu/busdemeoclass.html. For outer Solar System bodies, we define the red centaur, trojan and comet zones based on observed spectra of extreme examples. The centaur zone upper limit is the Pholus spectrum taken from ref. 33, while the lower limit is (55576) Amycus34. The trojan spectra are also defined by previously published data35. The X-shooter spectrum of Echeclus was obtained by the authors (W.C.F. and T.S.) and reduced in the same manner as the I1/2017 U1 data. This will be fully described in a forthcoming paper.

For comet nuclei, there are relatively few observations in the NIR, due to the fact that nuclei are very faint targets when far enough from the Sun to be inactive, but previous observations have shown the dust spectra of weakly active comets to match their nuclei (for example, 67P/Churyumov–Gerasimenko observed simultaneously with X-shooter and from Rosetta19,36). To define the comet zone, we take the upper limit from 19P/Borrelly from spacecraft data37 and the lower limit from C/2001 OG108 (ref. 38), as it covers a wide wavelength range.

### Thermal modelling

To determine the surface and sub-surface temperature of 1I/2017 U1 as a function of time, we solve the one-dimensional heat conduction equation with a suitable surface boundary condition. For temperature T, time t, and depth z, one-dimensional heat conduction is described by

$$\frac{dT}{dt}=\frac{k}{\rho C}\frac{{d}^{2}T}{d{z}^{2}}$$

where k is the thermal conductivity, ρ is the material density and C is the heat capacity39. These properties are assumed to be constant with temperature and depth. For a surface element located on 1I/2017 U1, conservation of energy leads to the surface boundary condition

$$f(1-{A}_{{\rm{B}}})\frac{{F}_{\odot }}{{r}_{{\rm{h}}}{(t)}^{2}}+k{\left(\frac{dT}{dz}\right)}_{z=0}-\varepsilon \sigma {T}_{z=0}^{4}=0$$

where A B is the Bond albedo, F is the integrated solar flux at 1 au (1,367 W m−2), r h(t) is the heliocentric distance in au of 1I/2017 U1 at time t, ε is the bolometric emissivity and σ is the Stefan–Boltzmann constant. f is a multiplying factor to take into account the different illumination scenarios we considered. For instance, f has a value of 1/π to give the rotationally averaged temperature of a surface element located on the equator of 1I/2017 U1 when considering a pole obliquity of 0°. If the surface element is permanently illuminated by the Sun during the encounter, f = 1. The true solution for 1I/2017 U1 will therefore lie between these two illumination condition extremes.

A finite difference numerical technique was used to solve the one-dimensional heat conduction equation and a Newton–Raphson iterative technique was used to solve the surface boundary condition40. In particular, the depth down to 5 m was resolved into 1 mm steps and time was propagated in increments of 1 s. Zero temperature gradient was also assumed at maximum depth to give a required internal boundary condition. The simulation was started 6,500 days before perihelion when 1I/2017 U1 was over 100 au away from the Sun. Low albedo isothermal objects have a temperature of ~30 K at such heliocentric distances, as calculated from

$$T={\left(\frac{{F}_{\odot }(1-{A}_{{\rm{B}}})}{4\varepsilon \sigma {r}_{{\rm{h}}}^{2}}\right)}^{1/4}$$

and so the initial temperature at all depths was set to this value. The hyperbolic orbital elements of 1I/2017 U1 were then used to calculate the heliocentric distance at each time step.

Regarding the material properties of 1I/2017 U1, cometary bodies typically have low albedo and highly insulating surfaces41,42, and we assume that 1I/2017 U1 is similar. Therefore, we assume a Bond albedo of 0.01, a bolometric emissivity of 0.95, a thermal conductivity of 0.001 W m−1 K−1, a density of 1,000 kg m−3 and a heat capacity of 550 J kg−1 K−1. The latter three properties combine to give a thermal inertia of ~25 J m−2 K−1 s−1/2, calculated using $${\rm{\Gamma }}=\sqrt{k\rho C}$$, which is comparable to that measured for several comets43 and outer main-belt asteroids39.

For the two illumination scenarios considered, the thermal model was propagated forwards from its initial starting point and run until 6,500 days after perihelion. The temperature at depths of 0, 10, 20, 30 and 40 cm was recorded at 1-day intervals in the model. As shown in Fig. 4, the most significant temperature changes occur during the 400 days centred on perihelion.

We note that as the thermal penetration depth is proportional to $$\sqrt{k{\rm{/}}(\rho C)}$$ our results can be scaled to different thermal property values. Identical temperature profiles can be found at depths given by

$$z={z}_{0}\sqrt{\frac{(k{\rm{/}}0.001)}{(\rho {\rm{/}}1000)(C{\rm{/}}550)}}$$

For example, if the thermal inertia was ~250 J m−2 K−1 s−1/2 (the 3σ upper limit determined for comet 103 P/Hartley 2; ref. 44), the depth of the temperature profiles would be ten times higher if the difference in thermal inertia was solely due to a difference in thermal conductivity. However, the depth would be less if the increased thermal inertia was spread equally across its three components. Furthermore, if the geometric albedo of 1I/2017 U1 is very low, the temperatures are also relatively insensitive to factor of two changes in this parameter.

### Data availability

The ACAM and X-shooter spectra that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## References

1. 1.

Charnoz, S. & Morbidelli, A. Coupling dynamical and collisional evolution of small bodies: an application to the early ejection of planetesimals from the Jupiter–Saturn region. Icarus 166, 141–156 (2003).

2. 2.

Dones, L., Weissman, P. R., Levison, H. F. & Duncan, M. J. in Comets II (eds Festou, M. C., Keller, H. U. & Weaver, H. A.) 153–174 (Univ. Arizona Press, Tucson, 2004).

3. 3.

Meech, K. J. et al. A brief visit from a red and extremely elongated interstellar asteroid. Nature https://doi.org/10.1038/nature25020 (2017).

4. 4.

Guilbert-Lepoutre, A. et al. On the evolution of comets. Space Sci. Rev. 197, 271–296 (2015).

5. 5.

Fitzsimmons, A., Hyland, M., Jedicke, R., Snodgrass, C. & Yang, B. Minor planets 2017 SN_33 and 2017 U1. Cent. Bur. Electr. Telegr. 4450, 2 (2017).

6. 6.

Feldman, P. D., Cochran, A. L. & Combi, M. R. in Comets II (eds Festou, M. C., Keller, H. U. & Weaver, H. A.) 425–447 (Univ. Arizona Press, Tucson, 2004).

7. 7.

Knight, M. M. et al. On the rotation period and shape of the hyperbolic asteroid 1I/Oumuamua (2017) U1 from its lightcurve. Astrophys. J. Lett. (in the press); preprint at https://arxiv.org/abs/1711.01402.

8. 8.

Bannister, M. T. et al. Col-OSSOS: colors of the interstellar planetesimal 1I/Oumuamua. Astrophys. J. Lett. (in the press); preprint at https://arxiv.org/abs/1711.06214.

9. 9.

Ye, Q.-Z., Zhang, Q., Kelley, M. S. P. & Brown, P. G. 1I/2017 U1 (`Oumuamua) is hot: imaging, spectroscopy and search of meteor activity. Astrophys. J. Lett. (in the press); preprint at https://arxiv.org/abs/1711.02320.

10. 10.

Rivkin, A. S. et al. in Asteroids IV (eds Michel, P., DeMeo, F. E. & Bottke, W. F.) 65–87 (Univ. Arizona Press, Tucson, 2015).

11. 11.

Reddy, V., Dunn, T. L., Thomas, C. A., Moskovitz, N. A. & Burbine, T. H. in Asteroids IV (eds Michel, P., DeMeo, F. E. & Bottke, W. F.) 43–63 (Univ. Arizona Press, Tucson, 2015).

12. 12.

Masiero, J. Palomar optical spectrum of hyperbolic near-earth object A/2017 U1. Preprint at https://arxiv.org/abs/1710.09977v2 (2017).

13. 13.

Bolin, B. T. et al. APO time resolved color photometry of highly-elongated interstellar object 1I/’Oumuamua. Astrophys. J. Lett. (in the press); preprint at https://arxiv.org/abs/1711.04927v4.

14. 14.

Jewitt, D. et al. Interstellar Interloper 1I/2017 U1: observations from the NOT and WIYN telescopes. Astrophys J. Lett. 850, L36 (2017).

15. 15.

Mothé-Diniz, T., Lazzaro, D., Carvano, J. M. & Florczak, M. Rotationally resolved spectra of some S-type asteroids. Icarus 148, 494–507 (2000).

16. 16.

Fraser, W. C., Brown, M. E. & Glass, F. The Hubble Wide Field Camera 3 test of surfaces in the outer solar system: spectral variation on Kuiper belt objects. Astrophys. J. 804, 31 (2015).

17. 17.

DeMeo, F. E., Binzel, R. P., Slivan, S. M. & Bus, S. J. An extension of the bus asteroid taxonomy into the near-infrared. Icarus 202, 160–180 (2009).

18. 18.

Emery, J. P., Burr, D. M. & Cruikshank, D. P. Near-infrared spectroscopy of trojan asteroids: evidence for two compositional groups. Astron. J. 141, 25 (2011).

19. 19.

Snodgrass, C. et al. Distant activity of 67P/Churyumov–Gerasimenko in 2014: ground-based results during the Rosetta pre-landing phase. Astron. Astrophys. 588, A80 (2016).

20. 20.

Merlin, F., Hromakina, T., Perna, D., Hong, M. J. & Alvarez-Candal, A. Taxonomy of trans-Neptunian objects and centaurs as seen from spectroscopy. Astron. Astrophys. 604, A86 (2017).

21. 21.

Pike, R. E. et al. Col-OSSOS: z-band photometry reveals three distinct TNO surface types. Astron. J. 154, 101 (2017).

22. 22.

Brunetto, R., Barucci, M. A., Dotto, E. & Strazzulla, G. Ion irradiation of frozen methanol, methane, and benzene: linking to the colors of centaurs and trans-Neptunian objects. Astrophys. J. 644, 646–650 (2006).

23. 23.

Strazzulla, G., Cooper, J. F., Christian, E. R. & Johnson, R. E. Irradiation ionique des OTNs: des flux mesurés dans l’espace aux expériences de laboratoire. C. R. Phys. 4, 791–801 (2003).

24. 24.

Meech, K. J. et al. Inner solar system material discovered in the Oort cloud. Sci. Adv. 2, e1600038 (2016).

25. 25.

Engelhardt, T. et al. An observational upper limit on the interstellar number density of asteroids and comets. Astron. J. 153, 133 (2017).

26. 26.

Jewitt, D. C. in Comets II (eds Festou, M. C., Keller, H. U. & Weaver, H. A.) 659–676 (Univ. Arizona Press, Tucson, 2004).

27. 27.

Groussin, O. et al. Gravitational slopes, geomorphology, and material strengths of the nucleus of comet 67P/Churyumov–Gerasimenko from OSIRIS observations. Astron. Astrophys. 583, A32 (2015).

28. 28.

Meech, K. J. et al. Activity of comets at large heliocentric distances pre-perihelion. Icarus 201, 719–739 (2009).

29. 29.

Zackrisson, E. et al. Terrestrial planets across space and time. Astrophys. J. 833, 214 (2016).

30. 30.

Jewitt, D. Color systematics of comets and related bodies. Astron. J. 150, 201 (2015). 1510.07069.

31. 31.

Benn, C., Dee, K. & Agócs, T. ACAM: a new imager/spectrograph for the William Herschel Telescope Proc. SPIE 7014, 70146X (2008); https://doi.org/10.1117/12.788694.

32. 32.

Vernet, J. et al. X-shooter, the new wide band intermediate resolution spectrograph at the ESO Very Large Telescope. Astron. Astrophys. 536, A105 (2011).

33. 33.

Cruikshank, D. P. et al. The composition of centaur 5145 Pholus. Icarus 135, 389–407 (1998).

34. 34.

Doressoundiram, A. et al. Spectral characteristics and modeling of the trans-Neptunian object (55565) 2002 AW197 and the centaurs (55576) 2002 GB10 and (83982) 2002 GO9: ESO Large Program on TNOs and Centaurs. Plan. Space Sci. 53, 1501–1509 (2005).

35. 35.

Emery, J. P., Burr, D. M. & Cruikshank, D. P. Near-infrared spectroscopy of trojan asteroids: evidence for two compositional groups. Astron. J. 141, 25 (2011).

36. 36.

Capaccioni, F. et al. The organic-rich surface of comet 67P/Churyumov–Gerasimenko as seen by VIRTIS/Rosetta. Science 347, 628 (2015).

37. 37.

Soderblom, L. A. et al. Short-wavelength infrared (1.3-2.6 μm) observations of the nucleus of Comet 19P/Borrelly. Icarus 167, 100–112 (2004).

38. 38.

Abell, P. A. et al. Physical characteristics of comet nucleus C/2001 OG108 (LONEOS). Icarus 179, 174–194 (2005).

39. 39.

Delbo, M., Mueller, M., Emery, J. P., Rozitis, B. & Capria, M. T. in Asteroids IV (eds Michel, P., DeMeo, F. E. & Bottke, W. F.) 107–128 (Univ. Arizona Press, Tucson, 2015).

40. 40.

Rozitis, B. & Green, S. F. Directional characteristics of thermal-infrared beaming from atmosphereless planetary surfaces—a new thermophysical model. Mon. Not. R. Astron. Soc. 415, 2042–2062 (2011).

41. 41.

Fornasier, S. et al. Spectrophotometric properties of the nucleus of comet 67P/Churyumov–Gerasimenko from the OSIRIS instrument onboard the ROSETTA spacecraft. Astron. Astrophys. 583, A30 (2015).

42. 42.

Spohn, T. et al. Thermal and mechanical properties of the near-surface layers of comet 67P/Churyumov–Gerasimenko. Science 349, aab0464 (2015).

43. 43.

Lowry, S. et al. The nucleus of comet 67P/Churyumov–Gerasimenko. A new shape model and thermophysical analysis. Astron. Astrophys. 548, A12 (2012).

44. 44.

Groussin, O. et al. The temperature, thermal inertia, roughness and color of the nuclei of comets 103P/Hartley 2 and 9P/Tempel 1. Icarus 222, 580–594 (2013).

45. 45.

Gasc, S. et al. Change of outgassing pattern of 67P/Churyumov–Gerasimenko during the March 2016 equinox as seen by ROSINA. Mon. Not. R. Astron. Soc. 469, S108–S117 (2017).

## Acknowledgements

We thank the observatory staff at the Isaac Newton Group of Telescopes and the European Southern Observatory for responding quickly to our observing requests. Particular thanks go to R. Ashley, C. Fariña and I. Skillen (Isaac Newton Group) and G. Beccari, B. Haeussler and F. Labrana (European Southern Observatory). A.F., M.T.B. and W.C.F. acknowledge support from Science and Technology Facilities Council grant ST/P0003094/1 and M.T.B. acknowledges support from Science and Technology Facilities Council grant ST/L000709/1. C.S. acknowledges support from the Science and Technology Facilities Council in the form of an Ernest Rutherford Fellowship. B.R. is supported by a Royal Astronomical Society Research Fellowship. The WHT is operated on the island of La Palma by the Isaac Newton Group of Telescopes in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofísica de Canarias. The ACAM spectroscopy was obtained as part of programme SW2017b11. This paper is also based on observations collected at the European Organisation for Astronomical Research in the Southern Hemisphere under European Southern Observatory programme 2100.C-5009.

## Author information

### Affiliations

1. #### Astrophysics Research Centre, School of Mathematics and Physics, Queen’s University Belfast, Belfast, UK

• Alan Fitzsimmons
• , Méabh Hyland
• , Tom Seccull
• , Michele T. Bannister
• , Wesley C. Fraser
•  & Pedro Lacerda
2. #### Planetary and Space Sciences, School of Physical Sciences, The Open University, Milton Keynes, UK

• Colin Snodgrass
•  & Ben Rozitis

• Bin Yang
4. #### Institute for Astronomy, Honolulu, HI, USA

• Robert Jedicke

### Contributions

A.F. led the application and organization of the WHT observations, analysis of these data and writing of the paper. C.S. led the application for VLT observations, organized the observing plan and assisted with analysis and writing. B.R. performed the thermal modelling of 1I/2017 U1. B.Y. was co-investigator on the telescope proposals, assisted in writing the VLT proposal and reduced the X-shooter data. M.T.B. and W.C.F. assisted in interpretation of the spectra in terms of known TNO properties and helped with writing the paper. M.H. reduced the WHT data. T.S. reduced the VLT data and provided the comparison spectrum of Echeclus. R.J. was co-investigator on the telescope proposals and contributed to the analysis and interpretation, especially with respect to observational selection effects. P.L. assisted in interpretation of the variable spectra and helped with writing the paper.

### Competing interests

The authors declare no competing financial interests.

### Corresponding author

Correspondence to Alan Fitzsimmons.

### DOI

https://doi.org/10.1038/s41550-017-0361-4

• ### Mysterious interstellar visitor is a comet — not an asteroid

• Alexandra Witze

Nature (2018)

• ### Ice can survive an interstellar trip

• Karen J. Meech

Nature Astronomy (2018)

• ### Tumbling motion of 1I/‘Oumuamua and its implications for the body’s distant past

• Michał Drahus
• , Piotr Guzik
• , Wacław Waniak
• , Barbara Handzlik
• , Sebastian Kurowski
•  & Siyi Xu

Nature Astronomy (2018)

• ### The tumbling rotational state of 1I/‘Oumuamua

• Wesley C. Fraser
• , Petr Pravec
• , Alan Fitzsimmons
• , Pedro Lacerda
• , Michele T. Bannister
• , Colin Snodgrass
•  & Igor Smolić

Nature Astronomy (2018)

• ### Non-gravitational acceleration in the trajectory of 1I/2017 U1 (‘Oumuamua)

• Marco Micheli
• , Davide Farnocchia
• , Karen J. Meech
• , Marc W. Buie
• , Olivier R. Hainaut
• , Dina Prialnik
• , Norbert Schörghofer
• , Harold A. Weaver
• , Paul W. Chodas
• , Jan T. Kleyna
• , Robert Weryk
• , Richard J. Wainscoat
• , Harald Ebeling
• , Jacqueline V. Keane
• , Kenneth C. Chambers
• , Detlef Koschny
•  & Anastassios E. Petropoulos

Nature (2018)