Near-infrared spectral monitoring of Triton with IRTF/SpeX I: establishing a baseline for rotational variability

We present eight new 0.8 to 2.4 μm spectral observations of Neptune's satellite Triton, obtained at IRTF/SpeX during 2002 July 15-22 UT. Our objective was to determine how Triton's near-infrared spectrum varies as Triton rotates, and to establish an accurate baseline for comparison with past and future observations. The most striking spectral change detected was in Triton's nitrogen ice absorption band at 2.15 μm; its strength varies by about a factor of two as Triton rotates. Maximum N₂ absorption approximately coincides with Triton's Neptune-facing hemisphere, which is also the longitude where the polar cap extends nearest Triton's equator. More subtle rotational variations are reported for Triton's CH₄ and H₂O ice absorption bands. Unlike the other ices, Triton's CO₂ ice absorption bands remain nearly constant as Triton rotates. Triton's H₂O ice is shown to be crystalline, rather than amorphous. Triton's N₂ ice is confirmed to be the warmer, hexagonal, β N₂ phase, and its CH₄ is confirmed to be highly diluted in N₂ ice.     

The diameter and reflectance of Triton

New image-tube spectra of Triton are analyzed for a determination of the reflectance of the satellite between 0.32 and 0.74 μm. Comparison of the violet reflectance of Triton with that of terrestrial minerals, lunar samples, and meteorites, gives evidence that the satellite surface is composed largely of rocky material having the same sources of violet opacity (mineral charge transfer and crystal field transitions). New radiometric observations set a stringent upper limit to the satellite radius (r ? 2600 km) and a lower limit to the geometric albedo (p_v ? 0.19). The albedo can be somewhat higher and still within the range allowed by a rocky surface. No useful constraints can be put on the mean density of Triton because of remaining uncertainties in the radius and the mass. The image-tube spectra show no evidence of gaseous absorption in the methane bands, though a stronger band has been found in the infrared at 2.3 μm (Cruikshank and Silvaggio, 1979, in press; the near-infrared photometric colors may be affected by the CH₄ band. Rayleigh scattering computations of a potential inert atmospheric component of Triton appear to preclude the presence of large quantities of nitrogen and the noble gases.     

Near-infrared spectral monitoring of Triton with IRTF/SpeX II: Spatial distribution and evolution of ices

This report arises from an ongoing program to monitor Neptune’s largest moon Triton spectroscopically in the 0.8 to 2.4 μm range using IRTF/SpeX. Our objective is to search for changes on Triton’s surface as witnessed by changes in the infrared absorption bands of its surface ices N₂,CH₄,H₂O, CO, and CO₂. We have recorded infrared spectra of Triton on 53 nights over the ten apparitions from 2000 to 2009. The data generally confirm our previously reported diurnal spectral variations of the ice absorption bands (Grundy and Young, 2004). Nitrogen ice shows a large amplitude variation, with much stronger absorption on Triton’s Neptune-facing hemisphere. We present evidence for seasonal evolution of Triton’s N₂ ice: the 2.15 μm absorption band appears to be diminishing, especially on the Neptune-facing hemisphere. Although it is mostly dissolved in N₂ ice, Triton’s CH₄ ice shows a very different longitudinal variation from the N₂ ice, challenging assumptions of how the two ices behave. Unlike Triton’s CH₄ ice, the CO ice does exhibit longitudinal variation very similar to the N₂ ice, implying that CO and N₂ condense and sublimate together, maintaining a consistent mixing ratio. Absorptions by H₂O and CO₂ ices show negligible variation as Triton rotates, implying very uniform and/or high latitude spatial distributions for those two non-volatile ices.     

Formation of radical species in photolyzed CH4:N2 ices

We report photochemical studies of thin cryogenic ice films composed of N₂, CH₄ and CO in ratios analogous to those on the surfaces of Neptune’s largest satellite, Triton, and on Pluto. Experiments were performed using a hydrogen discharge lamp, which provides an intense source of ultraviolet light to simulate the sunlight-induced photochemistry on these icy bodies. Characterization via infrared spectroscopy showed that C₂H₆ and C₂H₂, and HCO are formed by the dissociation of CH₄ into H, CH₂ and CH₃ and the subsequent reaction of these radicals within the ice. Other radical species, such as C₂, , CN, and CNN, are observed in the visible and ultraviolet regions of the spectrum. These species imply a rich chemistry based on formation of radicals from methane and their subsequent reaction with the N₂ matrix. We discuss the implications of the formation of these radicals for the chemical evolution of Triton and Pluto. Ultimately, this work suggests that , CN, HCO, and CNN may be found in significant quantities on the surfaces of Triton and Pluto and that new observations of these objects in the appropriate wavelength regions are warranted.     

Large seasonal variations in Triton's atmosphere

Triton's seasons differ materially from those of Pluto owing to four important differences in the governing physics: First, the obliquity of Triton is significantly less than Pluto's obliquity. Second, Triton's inclined orbit precesses rapidly about Neptune so that a complicated seasonal variation in the latitude of the Sun occurs for Triton. Third, Neptune's orbit is much more circular than Pluto's orbit so that the sunlight intercepted by Triton's disk does not vary seasonally. Finally, Triton's atmosphere cannot be saturated at the lower latitudes so that the mass of the atmosphere is controlled by the temperature of the high-latitude ices or liquids (polar caps), as for CO₂ on Mars. The consequences of Triton's entire surface being covered with volatile substances have been examined. It is found that the circularity of Neptune's orbit then implies that Triton would have hardly any seasonal variation at all in surface temperature or atmospheric bulk, in spite of the complicated precessional effects of Triton's orbit. The only seasonal effect would be the migration of surface ices and liquids. This scenario is ruled out because it implies a column CH₄ abundance much higher than that observed and because it quickly depletes the lower latitudes of volatiles. It is concluded that Triton's most volatile surface substances are probably relegated to latitudes higher than 35° and probably form polar caps. The temperature of the polar caps should be nearly equal, even during midwinter/midsummer when the insolation of the summer pole is greatest. If the summer pole completely sublimates during one of the “major” summers, Triton's atmosphere may begin to freeze out over the winter caps. It is therefore expected that Triton's atmosphere undergoes large and complex seasonal variations. Triton is currently approaching a “maximum southern summer”, and over the remainder of this century, a dramatic increase in CH₄ abundance above the current upper limit of 1 m-Am may be witnessed.     

Reassessing the origin of Triton

The origin of Neptune’s large, circular but retrograde satellite Triton has remained largely unexplained. There is an apparent consensus that its origin lies in it being captured, but until recently no successful capture mechanism has been found. Agnor and Hamilton (Agnor, C.B., Hamilton, D.P. [2006]. Nature 441, 192-194) demonstrated that the disruption of a trans-neptunian binary object which had Triton as a member, and which underwent a very close encounter with Neptune, was an effective mechanism to capture Triton while its former partner continued on a hyperbolic orbit. The subsequent evolution of Triton’s post-capture orbit to its current one could have proceeded through gravitational tides (Correia, A.C.M. [2009]. Astrophys. J. 704, L1-L4), during which time Triton was most likely semi-molten (McKinnon, W.B. [1984]. Nature 311, 355-358). However, to date, no study has been performed that considered both the capture and the subsequent tidal evolution. Thus it is attempted here with the use of numerical simulations. The study by Agnor and Hamilton (Agnor, C.B., Hamilton, D.P. [2006]. Nature 441, 192-194) is repeated in the framework of the Nice model (Tsiganis, K., Gomes, R., Morbidelli, A., Levison, H.F. [2005]. Nature 435, 459-461) to determine the post-capture orbit of Triton. After capture Triton is then subjected to tidal evolution using the model of Mignard (Mignard, F. [1979]. Moon Planets 20, 301-315; Mignard, F. [1980]. Moon Planets 23, 185-201). The perturbations from the Sun and the figure of Neptune are included. The perturbations from the Sun acting on Triton just after its capture cause it to spend a long time in its high-eccentricity phase, usually of the order of 10 Myr, while the typical time to circularise to its current orbit is some 200 Myr, consistent with earlier studies. The current orbit of Triton is consistent with an origin through binary capture and tidal evolution, even though the model prefers Triton to be closer to Neptune than it is today. The probability of capturing Triton in this manner is approximately 0.7%. Since the capture of Triton was at most a 50% event - since only Neptune has one, but Uranus does not - we deduce that in the primordial trans-neptunian disc there were some 100 binaries with at least one Triton-sized member. Morbidelli et al. (Morbidelli, A., Levison, H.F., Bottke, W.F., Dones, L., Nesvorný, D. [2009]. Icarus 202, 310-315) concludes there were some 1000 Triton-sized bodies in the trans-neptunian proto-planetary disc, so the primordial binary fraction with at least one Triton-sized member is 10%. This value is consistent with theoretical predictions, but at the low end. If Triton was captured at the same time as Neptune’s irregular satellites, the far majority of these, including Nereid, would be lost. This suggests either that Triton was captured on an orbit with a small semi-major axisa ? 50R_N (a rare event), or that it was captured before the dynamical instability of the Nice model, or that some other mechanism was at play. The issue of keeping the irregular satellites remains unresolved.     

Possible effects of secular resonances in Phobos and Triton

Due to the tides, the orbits of Phobos and Triton are contracting. While their semi major axes are decreasing, several possibilities of secular resonances involving node, argument of the pericenter and mean motion of the Sun will take place. In the case of Mars, if the obliquity (ε), during the passage through some resonances, is not so small, very significant variations of the inclination will appear. In one case, capture is almost certain provided that ε?20°. For Triton there are also similar situations, but capture seems to be not possible, mainly because in S₁ state, Triton's orbit is sufficiently inclined (far) with respect to the Neptune's equator. Following Chyba et al. (Astron. Astrophys. 219 (1989) 123), a simplified equation that gives the evolution of the inclination versus the semi major axis, is derived. The time needed for Triton crash onto Neptune is longer than that one obtained by these authors, but the main difference is due to the new data used here. In general, even in the case of non-capture passages, some significant jumps in inclination and in eccentricities are possible.     

A search for ethane on Pluto and Triton

We present here a search for solid ethane, C₂H₆, on the surfaces of Pluto and Triton, based on near-infrared spectral observations in the H and K bands (1.4-2.45 μm) using the Very Large Telescope (VLT) and the United Kingdom Infrared Telescope (UKIRT). We model each surface using a radiative transfer model based on Hapke theory (Hapke, B. [1993]. Theory of Reflectance and Emittance Spectroscopy. Cambridge University Press, Cambridge, UK) with three basic models: without ethane, with pure ethane, and with ethane diluted in nitrogen. On Pluto we detect weak features near 2.27, 2.405, 2.457, and 2.461 μm that match the strongest features of pure ethane. An additional feature seen at 2.317 μm is shifted to longer wavelengths than ethane by at least 0.002 μm. The strength of the features seen in the models suggests that pure ethane is limited to no more than a few percent of the surface of Pluto. On Triton, features in the H band could potentially be explained by ethane diluted in N₂, however, the lack of corresponding features in the K band makes this unlikely (also noted by Quirico et al. (Quirico, E., Doute, S., Schmitt, B., de Bergh, C., Cruikshank, D.P., Owen, T.C., Geballe, T.R., Roush, T.L. [1999]. Icarus 139, 159-178)). While Cruikshank et al. (Cruikshank, D.P., Mason, R.E., Dalle Ore, C.M., Bernstein, M.P., Quirico, E., Mastrapa, R.M., Emery, J.P., Owen, T.C. [2006]. Bull. Am. Astron. Soc. 38, 518) find that the 2.406-μm feature on Triton could not be completely due to ¹³CO, our models show that it could not be accounted for entirely by ethane either. The multiple origin of this feature complicates constraints on the contribution of ethane for both bodies.     

A coupled microphysical/radiative transfer model of albedo and emissivity of planetary surfaces covered by volatile ices

A new model of albedo and emissivity of planetary surfaces covered by volatile ices in the form of porous slab-like deposits is described. In the model, a radiative transfer model is coupled with a microphysical model in order to link changes in albedo and emissivity to changes in porosity caused by ice metamorphism. Preliminary results for Triton, Pluto, and Io are presented (the martian CO₂ caps will be the subject of a separate publication). The coupled model will aid in the interpretation of ground-based and spacecraft observations and should lead to advances in surface and atmospheric modeling.     

Cryovolcanism on the icy satellites

Evidence of past cryovolcanism is widespread and extremely varied on the icy satellites. Some cryovolcanic landscapes, notably on Triton, are similar to many silicate volcanic terrains, including what appear to be volcanic rifts, calderas and solidified lava lakes, flow fields, breached cinder cones or stratovolcanoes, viscous lava domes, and sinuous rilles. Most other satellites have terrains that are different in the important respect that no obvious volcanoes are present. The preserved record of cryovolcanism generally is believed to have formed by eruptions of aqueous solutions and slurries. Even Triton's volcanic crust, which is covered by nitrogen-rich frost, is probably dominated by water ice. Nonpolar and weakly polar molecular liquids (mainly N₂, CH₄, CO, CO₂, and Ar), may originate by decomposition of gas-clathrate hydrates and may have been erupted on some icy satellites, but without water these substances do not form rigid solids that are stable against sublimation or melting over geologic time. Triton's plumes, active at the time of Voyager 2's flyby, may consist of multicomponent nonpolar gas mixtures. The plumes may be volcanogenic fumaroles or geyserlike emissions powered by deep internal heating, and, thus, the plumes may be indicating an interior that is still cryomagmatically active; or Triton's plumes may be powered by solar heating of translucent ices very near the surface. The Uranian and Neptunian satellites Miranda, Ariel, and Triton have flow deposits that are hundreds to thousands of meters thick (implying highly viscous lavas); by contrast, the Jovian and Saturnian satellites generally have plains-forming deposits composed of relatively thin flows whose thicknesses have not been resolved in Voyager images (thus implying relatively low-viscosity lavas). One possible explanation for this inferred rheological distinction involves a difference in volatile composition of the Uranian and Neptunian satellites on one hand and of the Jovian and Saturnian satellites on the other hand. Perhaps the Jovian and Saturnian satellites tend to have relatively "clean" compositions with water ice as the main volatile (ammonia and water-soluble salts may also be present). The Uranian and Neptunian satellites may possess large amounts of a chemically unequilibrated comet-like volatile assemblage, including methanol, formaldehyde, and a host of other highly water- and ammonia-water-soluble constituents and gas clathrate hydrates. These two volatile mixtures would produce melts that differ enormously in viscosity The geomorphologic similarity in the products of volcanism on Earth and Triton may arise partly from a rheological similarity of the ammonia-water-methanol series of liquids and the silicate series ranging from basalt to dacite. An abundance of gas clathrate hydrates hypothesized to be contained by the satellites of Uranus and Neptune could contribute to evidence of explosive volcanism on those objects.     

Remote sensing D/H ratios in methane ice: Temperature-dependent absorption coefficients of CH3D in methane ice and in nitrogen ice

The existence of strong absorption bands of singly deuterated methane (CH₃D) at wavelengths where normal methane (CH₄) absorbs comparatively weakly could enable remote measurement of D/H ratios in methane ice on outer Solar System bodies. We performed laboratory transmission spectroscopy experiments, recording spectra at wavelengths from 1 to 6 μm to study CH₃D bands at 2.47, 2.87, and 4.56 μm, wavelengths where ordinary methane absorption is weak. We report temperature-dependent absorption coefficients of these bands when the CH₃D is diluted in CH₄ ice and also when it is dissolved in N₂ ice, and describe how these absorption coefficients can be combined with data from the literature to simulate arbitrary D/H ratio absorption coefficients for CH₄ ice and for CH₄ in N₂ ice. We anticipate these results motivating new telescopic observations to measure D/H ratios in CH₄ ice on Triton, Pluto, Eris, and Makemake.     

Infrared study of ion-irradiated N2-dominated ices relevant to Triton and Pluto: formation of HCN and HNC

Infrared spectra and radiation chemical behavior of N₂-dominated ices relevant to the surfaces of Triton and Pluto are presented. This is the first systematic IR study of proton-irradiated N₂-rich ices containing CH₄ and CO. Experiments at 12 K show that HCN, HNC, and diazomethane (CH₂N₂) form in the solid phase, along with several radicals. NH₃ is also identified in irradiated N₂ + CH₄ and N₂ + CH₄ + CO. We show that HCN and HNC are made in irradiated binary ice mixtures having initial N₂/CH₄ ratios from 100 to 4, and in three-component mixtures have an initial N₂/(CH₄ + CO) ratio of 50. HCN and HNC are not detected in N₂-dominated ices when CH₄ is replaced with C₂H₆, C₂H₂, or CH₃OH.The intrinsic band strengths of HCN and HNC are measured and used to calculate G(HCN) and G(HNC) in irradiated N₂ + CH₄ and N₂ + CH₄ + CO ices. In addition, the HNC/HCN ratio is calculated to be ∼1 in both icy mixtures. These radiolysis results reveal, for the first time, solid-phase synthesis of both HCN and HNC in N₂-rich ices containing CH₄.We examine the evolution of spectral features due to acid-base reactions (acids such as HCN, HNC, and HNCO and a base, NH₃) triggered by warming irradiated ices from 12 K to 30-35 K. We identify anions (OCN⁻, CN⁻, and N3−) in ices warmed to 35 K. These ions are expected to form and survive on the surfaces of Triton and Pluto. Our results have astrobiological implications since many of these products (HCN, HNC, HNCO, NH₃, NH₄OCN, and NH₄CN) are involved in the syntheses of biomolecules such as amino acids and polypeptides.     

Infrared reflectance spectra of Hyperion,Titania, and Triton

Medium-resolution infrared (1–2.5 μm; Δλ/λ ∽ 0.05) photometry of Triton, Titania, and Hyperion and medium-resolution (1.5–2.4 μm; Δλ/λ ? 0.01) spectroscopy of Triton are presented. Hyperion and Titania have spectra roughly similar to the laboratory spectrum of water frost, while the spectrum of Triton is inconsistent with the spectra of frosts likely to be major surface constituents.     

Methane on Triton: Physical state and distribution

Infrared spectrophotometric measurements of Neptune's satellite Triton obtained between 1980 and 1982 in the spectral range 0.8–2.5 μm show six individual absorption bands attributable to methane. An additional band in the Triton data is not methane. The Triton spectral data conform more closely to a laboratory spectrum of frozen methane than to a synthetic spectrum of methane gas computed for conditions of low temperature expected at the satellite. Additionally, the strength of the bands vary with Triton's orbital position. The data thus suggest that methane in the ice phase is mostly responsible for the bands in Triton's spectrum, and that the ice is distributed nonuniformly around the satellite's surface.     

Spatial and Compositional Constraints on Non-ice Components and H2O on Pluto's Surface

We present four new near-infrared spectra of Pluto, measured separately from its satellite Charon during four HST/NICMOS observations in 1998, timed to sample four evenly spaced longitudes on Pluto. Being free of contamination by telluric absorptions or by Charon light, the new data are particularly valuable for studies of Pluto's continuum absorption. Previous studies of the major volatile species indicate the existence of at least three distinct terrains on Pluto's surface: N₂-rich, CH₄-rich, and volatile-depleted. The new data provide evidence that each of these three terrains has distinct near-infrared continuum absorption features. CH₄-rich regions appear to show reddish continuum absorption through the near-infrared spectral range. N₂-rich regions have very little continuum absorption. Visually dark, volatile-depleted regions exhibit intermediate continuum albedos with a bluish continuum slope. By analogy with Triton, we expected that careful spectral modeling would reveal strong evidence for the existence of H₂O ice on Pluto's surface, but we found only very weak evidence for its existence in the volatile-depleted regions. These data require H₂O ice to play a much less prominent role on Pluto's surface than it does on Triton's.     

Watching meteors on Triton

The thin atmosphere of Neptune's moon Triton is dense enough to ablate micrometeoroids as they pass through. A combination of Triton's orbital velocity around Neptune and its orbital velocity around the Sun gives a maximum meteoroid impact velocity of approximately 19 km s⁻¹, sufficient to heat the micrometeoroids to visibility as they enter. The ablation profiles of icy and stony micrometeoroids were calculated, along with the estimated brightness of the meteors. In contrast to the terrestrial case, visible meteors would extend very close to the surface of Triton. In addition, the variation in the meteoroid impact velocity as Triton orbits Neptune produces a large variation in the brightness of meteors with orbital phase, a unique Solar System phenomenon.     

The Temperature-Dependent Spectrum of Methane Ice I between 0.7 and 5 μm and Opportunities for Near-Infrared Remote Thermometry

New infrared absorption coefficient spectra of pure methane ice I were measured at temperatures between 30 and 90 K, over wavelengths from 0.7 to 5 μm, along with spectra of methane ice II at 20 K and liquid methane at 93 K. The spectra were derived from transmission measurements through monocrystalline samples grown in a series of closed cells having interior dimensions ranging from 100 μm to 1 cm. The thicker samples permitted measurement of extremely weak absorption bands, with absorption coefficients as small as 0.003 cm⁻¹. We report 14 new absorption bands, which we tentatively assign to specific vibrational transitions. Two of the new bands are attributed to CH₃D. Measurements of the weaker CH₄ bands are particularly needed for interpreting spectral observations of Pluto and Triton, where a number of weak CH₄-ice absorption bands have been observed. The data presented in this paper complement studies of spectral transmission by thin films of methane ice, which are most suitable for measuring the stronger absorption bands. Temperature-dependent spectral features revealed by the new data offer the opportunity to determine CH₄-ice temperatures remotely, via near-infrared reflectance spectroscopy. This approach could prove particularly valuable for future spacecraft exploration of Pluto.     

Image tube spectra of pluto and triton from 6800 to 9000 Å

To search for a possible atmosphere on Pluto and Triton, spectra of these objects as well as comparison stars were obtained with a three-stage Varo image tube for the spectral region from 6800 to 9000 Å. Ratio spectra indicate an absorption feature near 8900 Å, although the steeply diminishing response of the image tube at that wavelength casts some doubt on the reality of this feature. The feature appears more definitive in the spectrum of Pluto and less certain in the spectrum of Triton. The absorption was analyzed using our recently determined band-model parameters for methane. Under the assumption of a pressure higher than 0.01 atm an abundance of 3 m-amagat was determined. For pressures limited by the methane abundance itself, an abundance of 50 m-amagat and a pressure of 10^?3 atm was derived (using g = 0.20 g⊕ for both Pluto and Triton). This pressure is close to the pressure that can be expected from the equilibrium vapor pressure of a methane frost. If the absorption at 8900 Å is spurious, our analysis will be applicable as an upper limit for the presence of methane gas on Pluto or Triton.     

Nitrogen on Triton

The near-infrared spectrum of Triton is characterized by strong absorption bands of methane, probably in the solid state. An additional absorption band at 2.16 μm is tentatively identified as the density-induced (2-0) band of molecular nitrogen in the liquid state. The fundamental overtones of this band system cannot presently be observed because of limitations of the terrestrial atmosphere or spectral signal precision. Using the absorption coefficient for this band derived from laboratory observations and from the literature, it is calculated that Triton must have a layer of nitrogen at least tens of centimeters deep over much of its surface; this quantity is plausible in terms of the cosmic abundance of nitrogen and by comparison with Titan where a massive atmosphere of nitrogen exists. The Triton spectrum has been modeled with liquid nitrogen and solid methane, and it is found that the shape of the continuum in two spectral regions can be properly accounted for by adding a spectral component corresponding to fine-grained water frost. It is speculated that yet another component, a dark, solid, photochemical derivative of methane, may occur as a trace contaminant of the surface materials. If much of the surface of Triton is liquid, the radiometric observations of the satellite must be reinterpreted to derive the radius and surface albedo. If there is liquid nitrogen exposed on the surface, the atmosphere of Triton is probably dominated by nitrogen rather than methane because of the much higher vapor pressure of the former. At the calculated subsolar temperature of Triton, the vapor pressure of nitrogen implies a surface atmospheric pressure in the range 0.13 to 0.30 atm.