Transforming Dust to Planets

We review recent progress in understanding how nebular dust and gas are converted into the planets of the present-day solar system, focusing particularly on the “Grand Tack” and pebble accretion scenarios. The Grand Tack can explain the observed division of the solar system into two different isotopic “flavours”, which are found in both differentiated and undifferentiated meteorites. The isotopic chronology inferred for the development of these two “flavours” is consistent with expectations of gas-giant growth and nebular gas loss timescales. The Grand Tack naturally makes a small Mars and a depleted, dynamically-excited and compositionally mixed asteroid belt (as observed); it builds both Mars and the Earth rapidly, which is consistent with the isotopically-inferred growth timescale of the former, but not the latter. Pebble accretion can explain the rapid required growth of Jupiter and Saturn, and the number of Kuiper Belt binaries, but requires specific assumptions to explain the relatively protracted growth timescale of Earth. Pure pebble accretion cannot explain the mixing observed in the asteroid belt, the fast proto-Earth spin rate, or the tilt of Uranus. No current observation requires pebble accretion to have operated in the inner solar system, but the thermal and compositional consequences of pebble accretion have yet to be explored in detail.     

Feedstocks of the Terrestrial Planets

The processes of planet formation in our Solar System resulted in a final product of a small number of discreet planets and planetesimals characterized by clear compositional distinctions. A key advance on this subject was provided when nucleosynthetic isotopic variability was discovered between different meteorite groups and the terrestrial planets. This information has now been coupled with theoretical models of planetesimal growth and giant planet migration to better understand the nature of the materials accumulated into the terrestrial planets. First order conclusions include that carbonaceous chondrites appear to contribute a much smaller mass fraction to the terrestrial planets than previously suspected, that gas-driven giant planet migration could have pushed volatile-rich material into the inner Solar System, and that planetesimal formation was occurring on a sufficiently rapid time scale that global melting of asteroid-sized objects was instigated by radioactive decay of ²⁶Al. The isotopic evidence highlights the important role of enstatite chondrites, or something with their mix of nucleosynthetic components, as feedstock for the terrestrial planets. A common degree of depletion of moderately volatile elements in the terrestrial planets points to a mechanism that can effectively separate volatile and refractory elements over a spatial scale the size of the whole inner Solar System. The large variability in iron to silicon ratios between both different meteorite groups and between the terrestrial planets suggests that mechanisms that can segregate iron metal from silicate should be given greater importance in future investigations. Such processes likely include both density separation of small grains in the nebula, but also preferential impact erosion of either the mantle or core from differentiated planets/planetesimals. The latter highlights the important role for giant impacts and collisional erosion during the late stages of planet formation.     

Water Reservoirs in Small Planetary Bodies: Meteorites,Asteroids, and Comets

Asteroids and comets are the remnants of the swarm of planetesimals from which the planets ultimately formed, and they retain records of processes that operated prior to and during planet formation. They are also likely the sources of most of the water and other volatiles accreted by Earth. In this review, we discuss the nature and probable origins of asteroids and comets based on data from remote observations, in situ measurements by spacecraft, and laboratory analyses of meteorites derived from asteroids. The asteroidal parent bodies of meteorites formed \(\leq 4\) Ma after Solar System formation while there was still a gas disk present. It seems increasingly likely that the parent bodies of meteorites spectroscopically linked with the E-, S-, M- and V-type asteroids formed sunward of Jupiter’s orbit, while those associated with C- and, possibly, D-type asteroids formed further out, beyond Jupiter but probably not beyond Saturn’s orbit. Comets formed further from the Sun than any of the meteorite parent bodies, and retain much higher abundances of interstellar material. CI and CM group meteorites are probably related to the most common C-type asteroids, and based on isotopic evidence they, rather than comets, are the most likely sources of the H and N accreted by the terrestrial planets. However, comets may have been major sources of the noble gases accreted by Earth and Venus. Possible constraints that these observations can place on models of giant planet formation and migration are explored.     

The Formation of Mars: Building Blocks and Accretion Time Scale

In this review paper I address the current knowledge of the formation of Mars, focusing on its primary constituents, its formation time scale and its small mass compared to Earth and Venus. I argue that the small mass of Mars requires the terrestrial planets to have formed from a narrow annulus of material, rather than a disc extending to Jupiter. The truncation of the outer edge of the disc was most likely the result of giant planet migration, which kept Mars’ mass small. From cosmochemical constraints it is argued that Mars formed in a couple of million years and is essentially a planetary embryo that never grew to a full-fledged planet. This is in agreement with the latest dynamical models. Most of Mars’ building blocks consists of material that formed in the 2 AU to 3 AU region, and is thus more water-rich than that accreted by Earth and Venus. The putative Mars could have consisted of 0.1 % to 0.2 % by mass of water.     

Formation of the Cores of the Outer Planets

The formation of the giant planets seems to be best explained by accretion of planetesimals to form massive cores, which in the case of Jupiter and Saturn were able to capture nebular gas. However, the timescale for accretion of such cores has been a problem. Accretion in the outer solar system differs qualitatively from planetary growth in the terrestrial region, as the larger embryo masses and lower orbital velocities make bodies more subject to gravitational scattering. The planetesimal swarm in the outer nebula may be seeded by earlier-formed large bodies scattered from the region near the nebular “snow line”. Such a seed body can experience rapid runaway growth undisturbed by competitors; the style of growth is not oligarchy, but monarchy.     

The Contributions of Comets to Planets, Atmospheres, and Life: Insights from Cassini-Huygens, Galileo, Giotto, and Inner Planet Missions

Comets belong to a group of small bodies generally known as icy planetesimals. Today the most primitive icy planetesimals are the Kuiper Belt objects (KBOs) occupying a roughly planar domain beyond Neptune. KBOs may be scattered inward, allowing them to collide with planets. Others may move outward, some all the way into the Oort cloud. This is a spherical distribution of comet nuclei at a mean distance of ～50,000 AU. These nuclei are occasionally perturbed into orbits that intersect the paths of the planets, again allowing collisions. The composition of the atmosphere of Jupiter—and thus possibly all outer planets—shows the effects of massive early contributions from extremely primitive icy bodies that must have been close relatives of the KBOs. Titan may itself have a composition similar to that of Oort cloud comets. The origin and early evolution of its atmosphere invites comparison with that of the early Earth. Impacts of comets must have brought water and other volatile compounds to the Earth and the other inner planets, contributing to the reservoir of key ingredients for the origin of life. The magnitude of these contributions remains unknown but should be accessible to measurements by instruments on spacecraft.     

From Disks to Planets: The Making of Planets and Their Early Atmospheres. An Introduction

This paper is an introduction to volume 56 of the Space Science Series of ISSI, “From disks to planets—the making of planets and their proto-atmospheres”, a key subject in our quest for the origins and evolutionary paths of planets, and for the causes of their diversity. Indeed, as exoplanet discoveries progressively accumulated and their characterization made spectacular progress, it became evident that the diversity of observed exoplanets can in no way be reduced to the two classes of planets that we are used to identify in the solar system, namely terrestrial planets and gas or ice giants: the exoplanet reality is just much broader. This fact is no doubt the result of the exceptional diversity of the evolutionary paths linking planetary systems as a whole as well as individual exoplanets and their proto-atmospheres to their parent circumstellar disks: this diversity and its causes are exactly what this paper explores. For each of the main phases of the formation and evolution of planetary systems and of individual planets, we summarize what we believe we understand and what are the important open questions needing further in-depth examination, and offer some suggestions on ways towards solutions.We start with the formation mechanisms of circumstellar disks, with their gas and disk components in which chemical composition plays a very important role in planet formation. We summarize how dust accretion within the disk generates planet cores, while gas accretion on these cores can lead to the diversity of their fluid envelopes. The temporal evolution of the parent disk itself, and its final dissipation, put strong constraints on how and how far planetary formation can proceed. The radiation output of the central star also plays an important role in this whole story. This early phase of planet evolution, from disk formation to dissipation, is characterized by a co-evolution of the disk and its daughter planets. During this co-evolution, planets and their protoatmospheres not only grow, but they also migrate radially as a result of their interaction with the disk, thus moving progressively from their distance of formation to their final location. The formation of planetary fluid envelopes (proto-atmospheres and oceans), is an essential product of this planet formation scenario which strongly constrains their possible evolution towards habitability. We discuss the effects of the initial conditions in the disk, of the location, size and mass of the planetary core, of the disk lifetime and of the radiation output and activity of the central star, on the formation of these envelopes and on their relative extensions with respect to the planet core. Overall, a fraction of the planets retain the primary proto-atmosphere they initially accreted from the gas disk. For those which lose it in this early evolution, outgassing of volatiles from the planetary core and mantle, together with some contributions of volatiles from colliding bodies, give them a chance to form a “secondary” atmosphere, like that of our own Earth.When the disk finally dissipates, usually before 10 Million years of age, it leaves us with the combination of a planetary system and a debris disk, each with a specific radial distribution with respect to their parent star(s). Whereas the dynamics of protoplanetary disks is dominated by gas-solid dynamical coupling, debris disks are dominated by gravitational dynamics acting on diverse families of planetesimals. Solid-body collisions between them and giant impacts on young planetary surfaces generate a new population of gas and dust in those disks. Synergies between solar system and exoplanet studies are particularly fruitful and need to be stimulated even more, because they give access to different and complementary components of debris disks: whereas the different families of planetesimals can be extensively studied in the solar system, they remain unobserved in exoplanet systems. But, in those systems, long-wavelength telescopic observations of dust provide a wealth of indirect information about the unobserved population of planetesimals. Promising progress is being currently made to observe the gas component as well, using millimetre and sub-millimetre giant radio interferometers.Within planetary systems themselves, individual planets are the assembly of a solid body and a fluid envelope, including their planetary atmosphere when there is one. Their characteristics range from terrestrial planets through sub-Neptunes and Neptunes and to gas giants, each type covering most of the orbital distances probed by present-day techniques. With the continuous progress in detection and characterization techniques and the advent of major providers of new data like the Kepler mission, the architecture of these planetary systems can be studied more and more accurately in a statistically meaningful sense and compared to the one of our own solar system, which does not appear to be an exceptional case. Finally, our understanding of exoplanets atmospheres has made spectacular advances recently using the occultation spectroscopy techniques implemented on the currently operating space and ground-based observing facilities.The powerful new observing facilities planned for the near and more distant future will make it possible to address many of the most challenging current questions of the science of exoplanets and their systems. There is little doubt that, using this new generation of facilities, we will be able to reconstruct more and more accurately the complex evolutionary paths which link stellar genesis to the possible emergence of habitable worlds.     

Formation of Planetesimals and Accretion of the Terrestrial Planets

Planetesimals formed in the solar nebula by collisional coagulation. Dust aggregates settled toward the central plane, the larger ones growing by sweeping up smaller ones. A thin, dense layer of particles formed; shear-generated turbulence and differential motions induced by gas drag inhibited gravitational instability. Growth proceeded by collisions, producing planetesimals on a timescale of a few thousand years in the terrestrial zone. For bodies smaller than about a kilometer, motions were dominated by gas drag, and impact velocities decreased with size. At larger sizes gravitational interactions became significant, and velocities increased due to mutual perturbations. Larger bodies then grew more rapidly, this ``runaway' led to formation of tens to hundreds of lunar- to Mars-sized planetary embryos in the zone of terrestrial planets. The final accretion of these bodies into a few planets involved large impacts, and occurred on a timescale of 10⁷ to 10⁸ years. This scenario gives a reasonably consistent picture of the origin of the terrestrial planets, but does not account for the anomalously low eccentricities of the Earth and Venus. This revised version was published online in June 2006 with corrections to the Cover Date.     

Infrared Spectroscopy of Solar-System Planets

Most of our knowledge regarding planetary atmospheric composition and structure has been achieved by remote sensing spectroscopy. Planetary spectra strongly differ from one planet to another. CO₂ signatures dominate on Mars, and even more on Venus (where the thermal component is detectable down to 1 μm on the dark side). Spectroscopic monitoring of Venus, Earth and Mars allows us to map temperature fields, wind fields, clouds, aerosols, surface mineralogy (in the case of the Earth and Mars), and to study the planets’ seasonal cycles. Spectra of giant planets are dominated by H₂, CH₄ and other hydrocarbons, NH₃, PH₃ and traces of other minor compounds like CO, H₂O and CO₂. Measurements of the atmospheric composition of giant planets have been used to constrain their formation scenario.     

Planetary Radiation Budgets

The energy state of a planet depends fundamentally on its radiation budget. Measurements made from space over past decades have led to significant revisions of ground-based estimates, both of the reflected fraction (the Bond albedo) of solar radiative flux and of the emitted thermal infrared radiation flux, for the Earth as well as for the other planets. After a brief survey of methods and difficulties in accurately determining planetary radiation budgets, we note contradictions in existing tabulations of global parameters, in particular Bond albedo. For the Earth, such contradictions are unjustified, considering that global and annual means as well as the seasonal cycle of Earth Radiation Budget components have now been determined with high accuracy. The Earth's Bond albedo is close to 0.3. Net storage of energy in the Earth-ocean system is close to zero, with a well-established annual cycle of amplitude close to ±12 Wm⁻². Some contradictions remain for the other terrestrial planets. For the giant planets, modern reduced values of the Bond albedo imply reduced but still significant internal energy generation.     

Solar Variability and Climate Impact on Terrestrial Planets

Some possible factors of climate changes and of long term climate evolution are discussed with regard of the three terrestrial planets, Earth, Venus and Mars. Two positive feedback mechanisms involving liquid water, i.e., the albedo mechanism and the greenhouse effect of water vapour, are described. These feedback mechanisms respond to small external forcings, such as resulting from solar or astronomical constants variability, which might thus result in large influences on climatic changes on Earth. On Venus, reactions of the atmosphere with surface minerals play an important role in the climate system, but the involved time scales are much larger. On Mars, climate is changing through variations of the polar axis inclination over time scales of ～10⁵–10⁶ years. Growing evidence also exists that a major climatic change happened on Mars some 3.5 to 3.8 Gigayears ago, leading to the disappearance of liquid water on the planet surface by eliminating most of the CO₂ atmosphere greenhouse power. This change might be due to a large surge of the solar wind, or to atmospheric erosion by large bodies impacts. Indeed, except for their thermospheric temperature response, there is currently little evidence for an effect of long-term solar variability on the climate of Venus and Mars. This fact is possibly due to the absence of liquid water on these terrestrial planets.     

Vesta and Ceres: Crossing the History of the Solar System

The evolution of the Solar System can be schematically divided into three different phases: the Solar Nebula, the Primordial Solar System and the Modern Solar System. These three periods were characterized by very different conditions, both from the point of view of the physical conditions and from that of the processes there were acting through them. Across the Solar Nebula phase, planetesimals and planetary embryos were forming and differentiating due to the decay of short-lived radionuclides. At the same time, giant planets formed their cores and accreted the nebular gas to reach their present masses. After the gas dispersal, the Primordial Solar System began its evolution. In the inner Solar System, planetary embryos formed the terrestrial planets and, in combination with the gravitational perturbations of the giant planets, depleted the residual population of planetesimals. In the outer Solar System, giant planets underwent a violent, chaotic phase of orbital rearrangement which caused the Late Heavy Bombardment. Then the rapid and fierce evolution of the young Solar System left place to the more regular secular evolution of the Modern Solar System. Vesta, through its connection with HED meteorites, and plausibly Ceres too were between the first bodies to form in the history of the Solar System. Here we discuss the timescale of their formation and evolution and how they would have been affected by their passage through the different phases of the history of the Solar System, in order to draw a reference framework to interpret the data that Dawn mission will supply on them.     

Formation of the Outer Planets

Models of the origins of gas giant planets and ‘ice’ giant planets are discussed and related to formation theories of both smaller objects (terrestrial planets) and larger bodies (stars). The most detailed models of planetary formation are based upon observations of our own Solar System, of young stars and their environments, and of extrasolar planets. Stars form from the collapse, and sometimes fragmentation, of molecular cloud cores. Terrestrial planets are formed within disks around young stars via the accumulation of small dust grains into larger and larger bodies until the planetary orbits become well enough separated that the configuration is stable for the lifetime of the system. Uranus and Neptune almost certainly formed via a bottom-up (terrestrial planet-like) mechanism; such a mechanism is also the most likely origin scenario for Saturn and Jupiter.     

Water and Volatile Inventories of Mercury,Venus, the Moon,and Mars

We review the geochemical observations of water, \(\mbox{D}/\mbox{H}\) and volatile element abundances of the inner Solar System bodies, Mercury, Venus, the Moon, and Mars. We focus primarily on the inventories of water in these bodies, but also consider other volatiles when they can inform us about water. For Mercury, we have no data for internal water, but the reducing nature of the surface of Mercury would suggest that some hydrogen may be retained in its core. We evaluate the current knowledge and understanding of venusian water and volatiles and conclude that the venusian mantle was likely endowed with as much water as Earth of which it retains a small but non-negligible fraction. Estimates of the abundance of the Moon’s internal water vary from Earth-like to one to two orders of magnitude more depleted. Cl, K, and Zn isotope anomalies for lunar samples argue that the giant impact left a unique geochemical fingerprint on the Moon, but not the Earth. For Mars, an early magma ocean likely generated a thick crust; this combined with a lack of crustal recycling mechanisms would have led to early isolation of the Martian mantle from later delivery of water and volatiles from surface reservoirs or late accretion. The abundance estimates of Martian mantle water are similar to those of the terrestrial mantle, suggesting some similarities in the water and volatile inventories for the terrestrial planets and the Moon.     

Planetesimals and Satellitesimals: Formation of the Satellite Systems

The origin of the regular satellites ties directly to planetary formation in that the satellites form in gas and dust disks around the giant planets and may be viewed as mini-solar systems, involving a number of closely related underlying physical processes. The regular satellites of Jupiter and Saturn share a number of remarkable similarities that taken together make a compelling case for a deep-seated order and structure governing their origin. Furthermore, the similarities in the mass ratio of the largest satellites to their primaries, the specific angular momenta, and the bulk compositions of the two satellite systems are significant and in need of explanation. Yet, the differences are also striking. We advance a common framework for the origin of the regular satellites of Jupiter and Saturn and discuss the accretion of satellites in gaseous, circumplanetary disks. Following giant planet formation, planetesimals in the planet’s feeding zone undergo a brief period of intense collisional grinding. Mass delivery to the circumplanetary disk via ablation of planetesimal fragments has implications for a host of satellite observations, tying the history of planetesimals to that of satellitesimals and ultimately that of the satellites themselves. By contrast, irregular satellites are objects captured during the final stages of planetary formation or the early evolution of the Solar System; their distinct origin is reflected in their physical properties, which has implications for the subsequent evolution of the satellites systems.     

Water and the Interior Structure of Terrestrial Planets and Icy Bodies

Water content and the internal evolution of terrestrial planets and icy bodies are closely linked. The distribution of water in planetary systems is controlled by the temperature structure in the protoplanetary disk and dynamics and migration of planetesimals and planetary embryos. This results in the formation of planetesimals and planetary embryos with a great variety of compositions, water contents and degrees of oxidation. The internal evolution and especially the formation time of planetesimals relative to the timescale of radiogenic heating by short-lived ²⁶Al decay may govern the amount of hydrous silicates and leftover rock–ice mixtures available in the late stages of their evolution. In turn, water content may affect the early internal evolution of the planetesimals and in particular metal-silicate separation processes. Moreover, water content may contribute to an increase of oxygen fugacity and thus affect the concentrations of siderophile elements within the silicate reservoirs of Solar System objects. Finally, the water content strongly influences the differentiation rate of the icy moons, controls their internal evolution and governs the alteration processes occurring in their deep interiors.     

Origin, Internal Structure and Evolution of 4 Vesta

Asteroid 4 Vesta is the only preserved intact example of a large, differentiated protoplanet like those believed to be the building blocks of terrestrial planet accretion. Vesta accreted rapidly from the solar nebula in the inner asteroid belt and likely melted due to heat released due to the decay of ²⁶Al. Analyses of meteorites from the howardite-eucrite-diogenite (HED) suite, which have been both spectroscopically and dynamically linked to Vesta, lead to a model of the asteroid with a basaltic crust that overlies a depleted peridotitic mantle and an iron core. Vesta??s crust may become more mafic with depth and might have been intruded by plutons arising from mantle melting. Constraints on the asteroid??s moments of inertia from the long-wavelength gravity field, pole position and rotation, informed by bulk composition estimates, allow tradeoffs between mantle density and core size; cores of up to half the planetary radius can be consistent with plausible mantle compositions. The asteroid??s present surface is expected to consist of widespread volcanic terrain, modified extensively by impacts that exposed the underlying crust or possibly the mantle. Hemispheric heterogeneity has been observed by poorly resolved imaging of the surface that suggests the possibility of a physiographic dichotomy as occurs on other terrestrial planets. Vesta might have had an early magma ocean but details of the early thermal structure are far from clear owing to model uncertainties and paradoxical observations from the HEDs. Petrological analysis of the eucrites coupled with thermal evolution modeling recognizes two possible mechanisms of silicate-metal differentiation leading to the formation of the basaltic achondrites: equilibrium partial melting or crystallization of residual liquid from the cooling magma ocean. A firmer understanding the plethora of complex physical and chemical processes that contribute to melting and crystallization will ultimately be required to distinguish among these possibilities. The most prominent physiographic feature on Vesta is the massive south polar basin, whose formation likely re-oriented the body axis of the asteroid??s rotation. The large impact represents the likely major mechanism of ejection of fragments that became the HEDs. Observations from the Dawn mission hold the promise of revolutionizing our understanding of 4 Vesta, and by extension, the nature of collisional, melting and differentiation processes in the nascent solar system.     

Formation of Giant Planets– An Attempt in Matching Observational Constraints

We present models of giant planet formation, taking into account migration and disk viscous evolution. We show that migration can significantly reduce the formation timescale bringing it in good agreement with typical observed disk lifetimes. We then present a model that produces a planet whose current location, core mass and total mass are comparable with the one of Jupiter. For this model, we calculate the enrichments in volatiles and compare them with the one measured by the Galileo probe. We show that our models can reproduce both the measured atmosphere enrichments and the constraints derived by Guillot et al. (2004), if we assume the accretion of planetesimals with ices/rocks ratio equal to 4, and that a substantial amount of CO₂ was present in vapor phase in the solar nebula, in agreement with ISM measurements.     

Dynamo Models for Planets Other Than Earth

Observations from planetary spacecraft missions have demonstrated a spectrum of dynamo behaviour in planets. From currently active dynamos, to remanent crustal fields from past dynamo action, to no observed magnetization, the planets and moons in our solar system offer magnetic clues to their interior structure and evolution. Here we review numerical dynamo simulations for planets other than Earth. For the terrestrial planets and satellites, we discuss specific magnetic field oddities that dynamo models attempt to explain. For the giant planets, we discuss both non-magnetic and magnetic convection models and their ability to reproduce observations of surface zonal flows and magnetic field morphology. Future improvements to numerical models and new missions to collect planetary magnetic data will continue to improve our understanding of the magnetic field generation process inside planets.     

Formation and Composition of Planetesimals

The composition of planetesimals depends upon the epoch and the location of their formation in the solar nebula. Meteorites produced in the hot inner nebula contain refractory compounds. Volatiles were present in icy planetesimals and cometesimals produced in the cold outer nebula. However, the mechanism responsible for their trapping is still controversial. We argue for a general scenario valid in all regions of the turbulent nebula where water condensed as a crystalline ice (Hersant et al., 2004). Volatiles were trapped in the form of clathrate hydrates in the continuously cooling nebula. The epoch of clathration of a given species depends upon the temperature and the pressure required for the stability of the clathrate hydrate. The efficiency of the mechanism depends upon the local amount of ice available. This scenario is the only one so far which proposes a quantitative interpretation of the non detection of N₂ in several comets of the Oort cloud (Iro et al., 2003). It may explain the large variation of the CO abundance observed in comets and predicts an Ar/O ratio much less than the upper limit of 0.1 times the solar ratio estimated on C/2001 A2 (Weaver et al., 2002). Under the assumption that the amount of water ice present at 5 AU was higher than the value corresponding to the solar O/H ratio by a factor 2.2 at least, the clathration scenario reproduces the quasi uniform enrichment with respect to solar of the Ar, Kr, Xe, C, N and S elements measured in Jupiter by the Galileo probe. The interpretation of the non-uniform enrichment in C, N and S in Saturn requires that ice was less abundant at 10 AU than at 5 AU so that CO and N₂ were not clathrated in the feeding zone of the planet while CH₄, NH₃ and H₂S were. As a result, the 14N/15N ratio in Saturn should be intermediate between that in Jupiter and the terrestrial ratio. Ar and Kr should be solar while Xe should be enriched by a factor 17. The enrichments in C, N and S in Uranus and Neptune suggest that available ice was able to form clathrates of CH₄, CO and the NH₃ hydrate, but not the clathrate of N₂. The enrichment of oxygen by a factor 440 in Neptune inferred by Lodders and Fegley (1994) from the detection of CO in the troposphere of the planet is higher by at least a factor 2.5 than the lower limit of O/H required for the clathration of CO and CH4 and for the hydration of NH₃. If CO detected by Encrenaz et al. (2004) in Uranus originates from the interior of the planet, the O/H ratio in the envelope must be around of order of 260 times the solar ratio, then also consistent with the trapping of detected volatiles by clathration. It is predicted that Ar and Kr are solar in the two planets while Xe would be enriched by a factor 30 to 70. Observational tests of the validity of the clathration scenario are proposed.