Probable Detection of Water Ice in the Coma of the Inbound Long-period Comet C/2017 K2 (PanSTARRS)

The discovery of comet C/2017 K2 (PanSTARRS) at R_H ∼ 16 au from the Sun (Wainscoat et al. 2017) in an apparent steady-state of activity since at least R_H ∼ 23.7 au (Jewitt et al. 2017; Hui et al. 2018) was unprecedented but not unexpected. Many comets are active beyond where the sublimation of water ice is sufficiently vigorous to power mass loss, meaning that the sublimation of some other ice (e.g., CO or CO₂) or some other process (e.g., crystallization of amorphous water ice, Prialnik & Bar-Nun 1990) must be the driver of activity. These alternative activity drivers appear to function very differently than the traditional sublimation of water ice (Bauer et al. 2015; Womack et al. 2017; Wierzchos & Womack 2020). Understanding how comets become active and sustain activity in the much colder thermal environment of the outer solar system compared to the better-studied inner solar system is critical to assessing how these different volatiles are stored and released.

However, detection and characterization of comets at large helio- and geo-centric distances is observationally challenging. The realization that C/2017 K2 was stably active at R_H = 23.7 au, where the equilibrium temperature of an object with a typical cometary albedo is just ∼60 Kelvin, is strong evidence against amorphous ice being the driver, making hypervolatile ices the only obvious culprit. The recent detection of CO emission toward C/2017 K2 supports this scenario (Yang et al. 2021) and is bolstered by models suggesting that the object had been active sine ∼35 au inbound (Jewitt et al. 2021). C/2017 K2 is already a fairly accessible target (m_V  ∼ 15 depending on aperture) to even moderate-sized telescopes (D = 1 − 3m), so understanding what powers K2's activity, how it changes in time, and what could be inferred about how it fits into the broader distant comet population could be pursued by a broad swath of the community using a variety of techniques.

We observed C/2017 K2 (PanSTARRS) with the SpeX instrument (Rayner et al. 2003) on 2021 April 30 at a heliocentric distance of R_H = 6.488 au and a geocentric distance of Δ = 6.199 au starting at 13:31 UTC. We "bookended" observations of the target with observations of a nearby G-type star to correct for strong telluric absorption by the atmosphere in the near-infrared and further corrected the spectrum to that of a well-studied Solar Analog star (SAO 120107). Due to the extended nature and brightness of the target, we utilized the 60'' long slit, the "prism" mode (R ∼ 100), and 200 s long exposures. We obtained a total of ∼93 usable minutes on target (28 200 s exposures). The reductions were primarily conducted within the "spextool" program (Cushing et al. 2004) in IDL. We note that the 60'' long slit is not formally supported for every step of the spextool reduction pipeline, so where applicable we used new custom-written Python scripts and publicly available packages. ¹ We extracted the spectrum of the comet with a radius of 4'', which corresponds to ∼18,000 linear kilometers at the distance of the target. Our normalized near-infrared reflectance spectrum of the coma of C/2017 K2 (PanSTARRS) is shown in Figure 1.

Zoom In Zoom Out Reset image size

The reflectance spectrum of C/2017 K2's coma is shown to be red and concave downwards with decreased reflectivity near ∼1.5 μm and ∼2.0 μm, consistent with the presence of water ice. The model shown in Figure 1 utilizes intimate mixing of water ice and amorphous carbon and converges to a volumetric ice fraction of 14% ± 2%, but further exploraton of the parameter space is required to properly characterize how much ice there is, whether it is amorphous or crystalline, and whether it is intimately mixed or separate from the refractory materials that make up typical cometary grains. (The under-estimation of the continuum near 1.7 μm is a sign that our telluric correction and/or this modeling approach needs at least some refinement.) We aim to complete this more detailed work in the coming months. Assuming that our detection of water ice is not spurious and can be confirmed by continuing observations of this compelling object, it seems quite possible that this object has been releasing material containing significant water ice for as long as it has been active this apparition.

The coming years surrounding K2's 2022 December perihelion at just q = 1.80 au will provide an unparalleled opportunity to profile a comet's changing activity between at least 23.7 au inbound where supervolatiles must power cometary activity all the way down to inside where water ice sublimation is thought to dominate a comet's activity. Continued observations probing K2's solid and gas comae, especially for the presence of water ice and gas, would be incredibly useful and this Research Note suggests that monitoring these ice features might not require an abundance of telescope time. In particular, finding out where water sublimation becomes detectable and the water ice coma begins to disappear could help understand how this extended ice coma contributes to the overall volatile production rate of the comet and could be compared to and build on Protopapa et al. (2018)s seminal study of time-dependent ice abundance in the coma of another long period comet, C/2013 US10 (Catalina). We are excited to see how C/2017 K2 (PanSTARRS) continues to evolve and hope these observations are useful to the community writ-large.

The authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of Maunakea has always had within the indigenous Hawaiian community. We are most fortunate to have the opportunity to conduct observations from this mountain.

This work was funded in part by NASA Near Earth Object Observations grant NNX17AJ19G (PI: Reddy).

We also thank the small bodies paper discussion group of the Lunar and Planetary Laboratory for motivating us to obtain these observations, as well as John W. Noonan for encouraging us to reduce this data set.

Facility: IRTF(SpeX). -

Software: spextool(Cushing et al. 2004).