Amphitrite: A Twist on Triton's Capture

Several attributes of Neptune are difficult to understand. Why does it orbit at 30.1 AU? How did its 3:2 resonance with Pluto and many other Kuiper Belt Objects (KBOs) arise? Why is it more massive than Uranus, yet have a less massive H/He atmosphere? Why does Neptune radiate 2.6 times the energy it receives from the Sun? And especially, why does it have a large satellite resembling a KBO, in a retrograde orbit? Resolution of these mysteries promises profound insights into the formation of Neptune and the entire Solar System. Explanations for the first few conundrums have already revealed much. It is the 3:2 resonance of Pluto and other KBOs that first suggested Neptune migrated outward > 10 AU [1]. The Nice model [2,3] of planetary migration further hypothesizes that all the giant planets formed in a much more compact configuration between 5 and 15 AU, that a ∼ 35 M⊕ disk of planetesimals orbited from 15-30 AU, and that all were destabilized 4 Gyr ago when Jupiter and Saturn reached a 2:1 resonance. During the destabilization, Neptune and Uranus may have switched orbits; [4] showed that the solar nebula mass distribution implied by the Nice model is more consistent with Neptune forming closer in than Uranus. It would then make sense that Neptune’s mass should exceed Uranus’s. Still unexplained, though, are Neptune’s internal structure and the capture of Triton. The internal structures of Neptune and Uranus have been modeled by [5], who match their masses, radii and gravitational moments. Uranus was successfuly modeled with 1.5 M⊕ of H/He gas and an ice density 90% of the 0 K density, consistent with temperatures in the upper ice layers only a factor of 2 higher than the adiabat with T (1 bar) = 75K. This has been explained by stable stratification in Uranus’s ice shell that inhibits convection and traps heat inside Uranus [6]; this is consistent with Uranus’s low heat flux. The magnetic dynamos of Uranus and Neptune also suggest no convection interior to r ≈ 0.6 − 0.7R in each planet [7]. [5] find two possible configurations for Neptune. The first (“Neptune 1”) has denser ice [i.e., 100% of the 0 K density], and more H/He gas (2.2 M⊕), with an H2O/H2 ratio 150 times solar. The second (“Neptune 2”) has less dense ice [i.e., 80% of the 0 K density], less H/He gas (0.9 M⊕), and a solar H2O abundance. [5] favored “Neptune 2” because its composition better matched Uranus’s, but the lower density and higher temperature of ice is difficult cannot be explained solely by stable stratification, especially in the face of Neptune’s high heat flux. Triton in mass and density and closely resembles Pluto, and its inclined, retrograde orbit clearly marks it as a captured KBO [8]. Its initial periapse after capture must have been about 7 RN, and it later circularized by tidal dissipation to its current 14 RN orbit [8]. To date the most probable explanation appears to be that Triton was part of a binary KBO that encountered Neptune [9], probably during its outward migration 4 Gyr ago. The other component of the binary carrying off orbital momentum enables Triton’s capture, but the model faces difficulties. First, the smaller body of the KBO binary is overwhelmingly favored to be captured, so Triton must be bound to a larger object, but at most < 100 KBOs larger than Triton could have existed in the planetesimals disk [10], and only a very few of them could have been bound to larger KBOs. Second, the approach velocity to Neptune of the KBO binary, v∞, must not exceed a few times the orbital velocity of Triton; for a Triton-Triton binary, typically v∞ < 0.5 km s−1 is needed for capture [9]. During Neptune’s migration, though, typically v∞ ≈ 2 − 3 km s−1 [12], and even if all Tritonsized KBOs were bound to larger KBOs, Neptune is unlikely to capture a Triton during its migration (< 0.02 probability) [10].