15 mins read 06 Aug 2020

Triton Exposed: Atmosphere to Core

In the second instalment from Dr. Graziella Caprarelli, our journey delves deeper into Neptune’s fascinating Moon, Triton. Set as one of the contenders for a new potential mission from NASA to the outer parts of our system, Triton exhibits surface and atmospheric features that make it a rare combination - a plume world, an ocean world and a Kuiper Belt Object.

Credit: NASA/JPL-Caltech/Lunar & Planetary Institute

In 1989, after a 12-year journey, a small human-made spacecraft known as Voyager 2 flew by the planet Neptune, and had a fruitful (though brief) encounter with an exotic and fascinating world: Triton. This small body, 1353.4 km in diameter, originated in the cold trans-Neptunian space occupied by the Kuiper Belt, before being captured into Neptune’s orbit, becoming one of its natural satellites. Its interior comprises a rocky core (950-1000 km in radius), surrounded by two concentric ice shells, possibly separated by a global ocean. The latter has been modelled to have a depth of up to 130 km, and might host life. 

Voyager 2’s observations exposed its icy surface, at equilibrium with a tenuous atmosphere, measuring a pressure of just 14 micro-bar (McKinnon and Kirk, 2014, In: Encyclopedia of the Solar System, pp. 861-881). Images showed a variety of morphologies consistent with recent geological activity. Only 40% of Triton was explored during the flight however, and the observed features left planetary scientists wanting more. 

Thirty-one years after Voyager 2’s encounter with Triton, NASA is now considering a new Discovery-Class mission named Trident, proposed by Dr Louise Prockter (Director of the Lunar and Planetary Institute in Houston) and her colleagues at JPL, Southwest Research Institute and other institutions. If the proposal is chosen to progress to mission development stage, it is possible that new observations of Triton up-close will become available in 2038.

Proposed instrumentation of Trident spacecraft. Credit: L.M. Prockter et al. LPI/JPL/SwRI

The proposed trajectory of the Trident mission, leaving Earth in 2025 and arriving at Neptune in 2038. Credit: L.M. Prockter et al. LPI/JPL/SwRI

A more ambitious proposal, led by Dr Abigail Rymer, from Johns Hopkins APL, calls for a future NASA Large Strategic Science Mission (possible launch window: 2029-2030) to insert a satellite around Neptune’s orbit for a comprehensive study of the Neptune-Triton system. The proposed mission envisions the possibility of landing on Triton to obtain in-situ geochemical data. This, and other proposals, were highlighted at a recent Planetary Mission Concept Study virtual workshop held on 26-27 May.  

In an earlier article I presented a general overview of Triton’s exploration history and discoveries. In this feature I focus on the interaction between the Tritonian atmosphere and cryosphere (i.e., its icy shells), and present some of the most enigmatic features observed on its surface. 

Atmosphere-Cryosphere Interaction

Figure 1: Sketch (not to scale) of simplified cross-section of surface ice layers, showing seasonal ice / atmosphere relation to geological ice. The arrows illustrate the directions of mass transfers across reservoirs, with yellow showing the seasonal sublimation/condensation cycle related to the position of the subsolar point. Also included a possible model for the origin of the jets (red arrow). Credit: Graziella Caprarelli

Before Voyager 2’s encounter with the Neptunian system, ground-based observations using infrared telescopes had detected CH4 (methane) and N2 (molecular nitrogen) ices on the surface of Triton (Cruikshank and Silvaggio, 1979, Astrophys J 233, 1016-1020; Cruikshank et al., 1984, Icarus 58, 293-305). Subsequent studies (Grundy et al., 2010, Icarus 205, 594-604) identified also H2O, and traces of CO2 (carbon dioxide) and CO (carbon monoxide), and showed that the spatial distribution of the ices is not uniform. 

This may reflect actual regional geological differences, or might be a result of cyclical sublimation of the most volatile species from the illuminated areas, and redeposition of the same in the cold traps represented by the dark regions of Triton (Fig. 1). Nitrogen and methane were detected in Triton’s atmosphere by the UV spectrometer on board Voyager 2 (Broadfoot et al., 1989, Science 246, 1459-1466; Herbert and Sandel, 1991, JGR Sp Phys S01, 19241-19252), and atmospheric carbon monoxide was eventually discovered through ground-based observations at infrared wavelengths (Lellouch et al., 2010, Astron Astrophys 512, L8, 6 pp). 

Because Triton’s atmospheric pressure has been observed to increase since Voyager 2’s flyby (Elliott et al., 1998, Nature 393, 765-767), sublimation/condensation cycles may be principally responsible for the non-uniform spatial distribution of the ice species. This is unlikely to be the whole story, however. In fact, the relative abundances of volatile species in the atmosphere do not reflect those in the ice: specifically, the CH4/N2 and CO/N2 atmospheric ratios are higher than it would be expected by mass balance (Lellouch et al., op. cit.). Furthermore, there is considerable uncertainty regarding the actual rate of pressure increase since 2000, with some scientists suggesting that the atmospheric pressure is closer to the equilibrium value than predicted from earlier models (Merlin et al., 2018, Icarus 314, 274-293). Even so, many discrepancies between modelled volatile species ratios and measured values remain to be explained, which we can only hope to do by acquiring new surface and atmospheric data.  

Triton occultation light curves, from the Hubble Space Telescope. Credit: Elliot et al. 1998, Nature, 393.

Surface temperature and pressure changes observed on Triton. Credit: Elliot et al. 1998, Nature, 393.

At an average distance of 30 AU from the Sun, Triton’s orbital period around the star is 164.8 years. Together with its primary (Neptune), Triton is slowly traveling toward perihelion (its closest distance from the Sun), which it will reach on 3 September 2042. Therefore, when Voyager 2 approached Triton, it was farther from the Sun than it is now (Fig. 2), which would make it logical to expect warming. 

The eccentricity of the Neptunian system’s revolution around the Sun, however, is only 0.0086, corresponding to a quasi-circular orbit, with a very small difference between its perihelion (29.81 AU) and aphelion (30.33 AU) distances. Indeed, the surface warming/cooling patterns on Triton are very complex and, just like for the Earth, have little to do with the body’s distance from the Sun.

Figure 2: Simplified diagram illustrating the relative orbital positions of the Neptunian system (light blue circles) on August 25, 1989 (Voyager 2’s flyby), and at the present time (July 4, 2020). The points marked by A and P are the aphelion and perihelion, respectively. Orbit eccentricity is exaggerated, Sun and planetary system dimensions and distances are not to scale. Credit: Graziella Caprarelli

However, unlike the Earth, where seasonal changes follow (relatively) simple annual patterns related to the tilt of the Earth’s rotation axis and insolation conditions as our planet revolves around the Sun (Fig. 3), Triton’s orbital peculiarities (introduced in the previous Triton feature) result in centennial oscillations of the subsolar point, which swings dramatically between extreme northern and southern latitudes (Fig. 4). 

Models (Trafton, 1984, Icarus 58, 312-324) indicate that Voyager 2 approached Triton when the subsolar point was in the southern hemisphere and migrating southward, eventually reaching its southernmost latitude (-50°) in the year 2000, before starting its northward journey (Holler et al., 2016, Icarus 267, 255-266). This marks a long period of persistent southern hemisphere illumination, which appears consistent with models explaining the observed increase in atmospheric pressure by seasonal volatilization of large quantities of ice.

Figure 3: Graphical representation of the approximate variation of the subsolar (noon) latitude on Earth during one calendar year: from March (year Y0) to March of the following year (Y1). Credit: Graziella Caprarelli

Figure 4: Millennial variation of the subsolar latitude on Triton (drafted from combination of plots in Fig. 1 from Holler et al., 2016, Icarus 267, 255-66, and Fig. 40.4 from McKinnon and Kirk, 2014, In: Encyclopedia of the Solar System). Vertical axis units: degrees latitude north; horizontal axis units: years (a). Star shows the approximate position of the subsolar latitude of -45.43° (Voyager 2’s flyby, 1989). Subsequently, the subpolar latitude moved further south, reaching the latitude of -50° in the year 2000, before starting to migrate north, reaching a latitude of -38° in 2020. Credit: Graziella Caprarelli

During Voyager 2’s flyby, dark particulate and gas jets reaching a height of 8 km were observed (Soderblom et al., 1990, Science 250, 410-415). Because of their locations (southern hemisphere), the most often cited mechanism for the genesis of the jets involves some form of solid-state greenhouse effect responsible for trapping of solar radiation in a dark (absorbing) geological material underlying the transparent veneer of N2 frost: as a consequence of a long period of insolation, the temperatures at the base of the veneer may become so high that the excess heat must eventually be released through the jets. 

While plausible, this explanation is not conclusive, however: all models depend on (yet to be verified) specific sets of physical and geological parameters to be true. One criticism of the greenhouse mechanism is based on the height reached by the observed jets: if the greenhouse mechanism were the only type of process in play, the jets should be only tens of meters high (Laufer et al., 2013 Icarus 222, 73-80). Thus, it appears likely that an elevated geotherm is also required to power jet activities (Brown et al., 1990, Science 250, 431-435). Internal heat sources are indeed consistent with observed surface morphologies interpreted as formed by volcano-tectonic processes.

Suerface Morphologies

Geology of Triton. Credit: Antonio Ciccolella

Even though only 40% of Triton’s surface was imaged during Voyager 2’s flyby, it was immediately apparent that Triton is home to a variety of geological processes. Furthermore, the relatively fresh morphologies, low density of impact craters, and the observation of plumes, indicated that Triton is a geologically active small body. Maps of Triton’s surface are available from the USGS. A geomorphological map based on the JPL Global Color Mosaic of Triton can be found here

Based on surface morphologies and density of impact craters (Smith et al., 1989, Science 246, 1422-1449), and (variously processed) photometric data (e.g., Hillier et al., 1994, Icarus 109, 296-312, and refs. therein), Triton’s geological domains can be broadly classified as: (1) Smooth Terrains; (2) Cantaloupe Terrain; (3) Bright Polar Regions. Each of these domains can be further subdivided into regions identifiable by specific morphological characteristics. 

Before introducing some of the most striking features, however, it is important to remember here that even though the morphologies may appear familiar, the “rocks” on Triton are in fact made of ice. Therefore, when we talk about lava and volcanic flows on Triton, we are describing “slushes”, probably originating in the warmer interior of Triton, and eventually erupted and solidified on the surface. Volcanism on Triton is therefore more appropriately referred to as “cryovolcanism” i.e., “cold” or “icy” (from Ancient Greek) volcanism.

Figure 5: Contrast between the surface of the cantaloupe morphologies (west in the image) and smooth cryovolcanic terrains (east). The geographic boundary between the two geological domains is shown as a white dashed line, approximately drawn based on visual examination of the image. Base-map by Dr. Paul Schenk at the Lunar and Planetary Institute, Houston, Texas: image selection, radiometric calibration, geographic registration and photometric correction, and final mosaic assembly. Image data from Voyager 2 (NASA, JPL). Credit: Graziella Caprarelli

The contrast between the Cantaloupe and Smooth Terrains is shown in Fig. 5, with smooth surfaces clearly identifiable to the east of the dashed line, and the rough landscapes (resembling the texture of a cantaloupe skin, hence their name) to the west. The eastern lands are higher in elevation than the western regions (although this is not apparent from the mosaiced image shown). 

This, coupled with evidence that vestiges of the western morphologies are still (feebly) visible east of the boundary, suggests that the smooth terrains overlay cantaloupe-like units, and are therefore younger. Some volcanic features are identified: pancake-shaped calderas (such as the feature named ‘Leviathan’ in the figure), commonly known as “paterae”, are probably due to resurgent volcanism; these calderas are seen to be surrounded by flat topographies, possibly lava flow fields; linear chains of depressions termed “catenae” (sing.: “catena” [Lat.]) may represent dikes or linear fissures from which molten material erupts. 

Other features (‘Tuonela’ and ‘Ruach’ in the figure) known as “planitiae” (sing.: “planitia” [Lat.], meaning “large plain”) are more difficult to explain, although their extremely flat appearance suggests they were formed by solidification of very fluid material, similar to hot lava. 

Endogenic (i.e., caused by processes originating from the interior of a planet or, as in this case, a satellite) features can be also identified in the enigmatic Cantaloupe Terrain. Here we can observe ovoidal or kidney-shaped depressions, approximately 25-30 km in extent, known as “cavi (sing.: “cavus” [Lat.]) believed to be diapirs, i.e., intrusions of ductile and buoyant material that ascends through hard and brittle overlaying rock. The logical implication of this interpretation is that Triton’s geology is stratified, with materials of different compositions and rheological characteristics (rheology is the study of the behaviour of materials when they are subject to deforming forces) laying on top of each other. Modelling the rheology of the materials likely to form the strata, further suggests that the top layer may be about 20 km thick (Schenk and Jackson, 1993, Geology 21, 299-301). 

Figure 6: Detail of sulci. Point A is the intersection of two sulci, at coordinates: 331.44°E, 30.09°N. Also marked a cavus (refer to text for definitions). Features and possible interpretation are described in the text of the article. Scale bar: 50 km (lower right). Base map as in Fig. 5. Credit: Graziella Caprarelli

Systems of linear features hundreds of km long are also associated with endogenic processes and intersect across the entire surface of Triton. The ridges are typically 15-20 km wide. In some cases, sets of parallel ridges are separated by a continuous axial depression, resembling furrows (Fig. 6), and are thus aptly termed “sulci” (sing.: “sulcus” [Lat.]). They are likely tectonic in origin, and, at least in some cases, might be linked to volcanic activity, although the initial hypothesis (Smith et al., 1989, Science 246, 1422-1449) of them being analogous to terrestrial grabens (long linear crustal depressions bound by normal faults on each side) has been discarded in favour of models combining tidal stresses associated to orbital decay, and to changes in the body radius caused by progressive cooling of the satellite (Collins and Schenk, 1994, LPSC XXV, 277-278). 

Figure 7: Bright polar region. Some maculae and plume deposits are shown. North is up. Base map as in Fig. 5. Credit: Graziella Caprarelli

South of the Cantaloupe Terrain, a vast bright region (named Uhlanga) was imaged by Voyager 2 (Fig. 7). Some of the features identified in the northern domains can be seen here as well, but much more subdued, which suggests the morphologies are covered by thick ice. Furthermore, because the boundary of the bright terrains cuts across the cantaloupe / smooth plains boundary, it is likely that polar terrains form the youngest region on Triton. Enigmatic black features surrounded by very bright material termed “maculae” (sing.: “macula”, [Lat.], meaning “spot”) may be planitiae lying under thick ice cover. Two very clearly visible maculae areas (‘Zin’ and ‘Akupara’) located between 60 and 70° E longitude, suggest the ice cover may be thinning here, possibly in response to seasonal variations. Ubiquitous dark streaks are probably solid residue from the gas jets described above.

Digging below the surface

Simplified model of Triton’s interior. Depth of core and thicknesses of ice and putative ocean are approximated from published values, as reported in the article. Credit: Graziella Caprarelli

The ridges (described above) are hundreds of km in extent, and a major feature of the Cantaloupe Terrain which, purely based on stratigraphic relations among the regions (as discussed in the previous section) is considered to be the oldest observable geological unit on Triton. This implies that the ridges themselves have persisted millions of years after their formation and, therefore, that the bedrock  must be strong enough to support these structures. This observation narrows the field of the type of ice forming the Tritonian rocks to water or ammonia-water ice (Prockter et al., 2005, Geophys Res Lett 32, L14202). 

The scarcity of large impact craters observed on the surface of Triton points to periodic resurfacing by endogenic processes, such as cryovolcanism. It is therefore plausible that molten material advects (i.e., ascends) from the warm interior as a consequence of convection in the ice shells mantling the core (refer to previous feature for a description of Triton’s interior). Another clue of endogenic activity is the typical reddish colour of the Tritonian veneer. This is attributed to photochemical reactions caused by UV irradiation of CH4 ice (Cruikshank et al., 1985, In: Ices in the Solar System, pp. 817-827). Long term exposure to UV rays, however, turns the red coloration to black. Hence, for the red colour to persist, methane ice must be replenished on a regular basis, presumably through emissions from the interior. 

At present, all we know about the innermost structure of Triton is based on models and comparisons with other solar system bodies (e.g., our Moon, Mars, Mercury, Pluto). It is beyond doubt that Triton’s core must be composed of denser material than ice, and it is therefore absolutely reasonable to hypothesise a rocky core, perhaps with a small fraction of metals. The core is mantled by less dense material, and thus far the best explanation is that this mantle comprises two concentric shells of (mostly) water ice. Some of the models predict the presence of an ammonia-rich ocean between the two shells (for more details, refer to the previous feature). We have done all we could with models though, and we need to collect data to acquire new knowledge about Triton’s interior. Some of these data will be hopefully acquired by the Trident mission. 

Open Questions and the Trident Mission

I hope that this brief encounter with Triton makes you even more curious about this extraordinary world and makes you want to discover more than what I could present in the limited space of two feature articles (this and the previous one). It should be clear by now that many more questions remain unanswered, for example: 

  • Is Triton a typical Kuiper Belt object, or is it peculiar, and if so, why?  

  • What is its internal structure and composition, and is there an ocean in its interior? If so, does or could this ocean host life?

  • What lies on the undiscovered 60% of Triton’s surface?

  • What is the structure of Triton’s atmosphere? Could we gain a better understanding of its chemical composition and thermal structure? And, 

  • What is the influence of Neptune’s magnetosphere on Triton’s ionosphere (not discussed in these feature articles)? 

The proposed Trident mission aims to address specifically the question of whether Triton has an ocean, why Triton’s surface is so young, and the unusual intensity (for a satellite) of Triton’s ionosphere (Prockter et al., 2019, LPSC L, #3188). The Trident mission team is proposing to obtain gravity and atmospheric data by radio occultation and plasma spectrometry, and to use data acquired by a magnetometer, narrow- and wide-angle camera imagery, and a high resolution infrared spectrometer, to gain a more detailed and complete picture of Triton’s surface and its interior. More information about Trident can be found here.

This concludes our Triton adventure. From the next feature we will be moving toward the Sun. Next stop: Jupiter’s innermost moon, the volcanic world of Io. 


Graziella Caprarelli was generally considered a polymath at school, which made it impossible for her to receive consistent career advice from teachers and family. She knew, however, that she wanted to spend her life wandering around the globe looking for adventure in far and exotic lands. She therefore decided to apply to study at the Faculty of Mathematical, Physical and Natural Sciences at Sapienza University in Rome (Italy), and in due course, having obtained an MS in Geological Sciences and a Ph.D. in Earth Sciences, she left her country of origin and moved abroad. To date, she has lived in four continents, flew over frozen lands, sailed the seas, and visited Buddhist temples in the Himalayas. Her scientist husband publicly declared that he finds her to be “fascinating”, though one should take this personal opinion with a grain of salt. Her professional resume can be found here. When she grows up she wants to live on the Moon.

Connect with ArjaRoxe on Twitter