For decades, Uranus and Neptune have been filed neatly into the “ice giant” drawer, shorthand for worlds built mostly from frozen water, ammonia and methane wrapped in hydrogen and helium. A new generation of interior models is now tearing up that label, suggesting their cores may be far rockier, more layered and more alien than the textbooks admit. If those results hold, the two distant planets may belong in a different planetary class altogether, and their hidden interiors could reshape how I think about worlds across the galaxy.
At stake is more than a semantic tweak. The structure of Uranus and Neptune determines how they formed, how they generate their bizarre magnetic fields and how we interpret the many exoplanets that look similar in size and mass. Their cores, it turns out, may be telling a very different story from the icy narrative astronomers once took for granted.
The fragile idea of an “ice giant”
The classic picture of Uranus and Neptune rests on a simple hierarchy: small rocky planets close to the Sun, gas giants like Jupiter and Saturn farther out, and then a pair of “ice giants” built from deep reservoirs of frozen volatiles. That scheme assumes that water, ammonia and methane ices dominate their interiors, with only a modest rocky core and a relatively thin gaseous envelope. Since Uranus and Neptune are gas planets, their equatorial radii are even defined from the center out to the edge of their gaseous layers, which reinforces the idea that what matters most is the ice and gas above any hidden rock. Yet the more closely I look at their gravity fields, heat flows and magnetic quirks, the less that tidy category seems to fit.
New modeling work argues that the longstanding label may actually conflict with what the data allow. Analyses that start from the measured mass, radius and gravitational harmonics of both worlds find that a wide range of interior structures can reproduce the observations, including configurations where rock dominates over ice. One assessment notes that, since Uranus and Neptune are gas planets, the traditional “ice giant” tag glosses over multiple viable internal arrangements and “possibilities for their hidden depths” that include rock-rich as well as ice-rich solutions, a point underscored in Since Uranus and Neptune. The category that once seemed secure now looks more like a working hypothesis than a settled fact.
Hybrid models that loosen old assumptions
The shift in thinking is driven by a new generation of interior models that deliberately avoid the rigid assumptions of earlier work. Instead of forcing Uranus and Neptune into a three-layer template of rock, ice and gas, researchers now combine detailed equations of state with flexible, data-driven profiles of density and composition. One team describes how, unlike previous research that either imposed strict layering or relied on simplified empirical profiles, their approach blends physical modeling with observational constraints to explore a much broader family of structures, as detailed in Unlike previous research. That methodological shift is crucial, because it lets the data, rather than tradition, decide how icy or rocky these planets really are.
These hybrid models do more than shuffle ingredients. They track how mixtures of hydrogen, helium, water, ammonia and silicate rock behave under the extreme pressures and temperatures inside Uranus and Neptune, then ask which combinations can match the observed gravity field and radius. In some solutions, the bulk of the mass sits in a dense, rock-rich mantle with only modest ice fractions, while in others, ices still play a leading role but are distributed in complex, partially mixed layers. By relaxing the old constraints, the models reveal that the “ice giant” picture is only one option among many, and not necessarily the most natural fit to the measurements.
What the new interior calculations actually show
To understand how radical this rethinking is, it helps to look at what the latest detailed calculations say about the deep interiors. In one set of simulations, researchers explicitly follow how mixtures of water, ammonia and methane behave as pressure and temperature rise toward the core. They show that these components phase separate under the pressure–temperature conditions in the interiors of Uranus and Neptune, which naturally produces distinct compositional layers rather than a single homogeneous “ice” mantle, as described in They show that these components. That phase separation opens the door to structures where heavy rock and lighter volatiles segregate in ways that were not captured by older, smoother models.
Those same calculations find that the density profiles compatible with the gravity data can be matched by both ice-rich and rock-rich interiors, provided the layering and temperature gradients are chosen appropriately. In some scenarios, Uranus in particular can host a large rocky region that extends over much of its radius, with ices confined to overlying shells or mixed zones. The models also explore whether the interior is fully convective or contains stable, non-convective layers that trap heat, which could help explain why Uranus emits so little internal energy compared with Neptune. The upshot is that the data do not force a single “icy” solution; instead, they admit a spectrum of possible cores and mantles that blur the line between ice giant and something more akin to a “rocky giant.”
Swiss-led evidence for rock-heavy “giants”
The most headline-grabbing twist comes from a Swiss-led study that leans hard into the rock-rich end of that spectrum. Using an innovative hybrid modeling framework, the team tested thousands of interior configurations against the observed properties of Uranus and Neptune and found that many of the best fits involve planets that are not dominated by ices at all. In their favored solutions, Uranus and Neptune may not be the icy worlds astronomers long imagined, but instead objects where a substantial fraction of the mass resides in dense, silicate and metal rich regions, as highlighted in Uranus and Neptune. That result does not eliminate ice, but it demotes it from the starring role implied by the traditional label.
Another report on the same work emphasizes that the new models open a completely new range of possibilities for the internal composition of both planets. Using their model, the researchers discovered that the potential interiors can be either water rich or rich in rock, with both categories matching the available data within uncertainties, a point captured in the phrase Completely new range of possibilities. That duality is precisely why the “ice giant” tag looks shaky: if rock-dominated and ice-dominated interiors are both viable, the name that singles out ice starts to look more like historical inertia than a reflection of what lies beneath the clouds.
Magnetic fields and the case for layered, exotic oceans
One of the strangest features of Uranus and Neptune is their magnetic fields, which are wildly tilted and offset from the planets’ centers. Any credible interior model has to explain how such fields are generated. For years, scientists thought Uranus and Neptune hosted deep oceans of water mixed with ammonia and other volatiles, and that these conductive fluids powered the dynamo. More recent work refines that picture, suggesting that the planets may hide complex oceans or shells of superionic water and other exotic phases that do not mix easily with surrounding materials, behaving together like oil and water, as described in Are Uranus and Neptune. Those stratified layers could help explain why the magnetic fields are so oddly configured and variable.
The new rock-rich models dovetail with that magnetic story by naturally producing layers of electrically conductive material at intermediate depths. One analysis notes that the models created such conductive zones, which can account for the magnetic fields of Uranus and Neptune without requiring a simple, global water ocean, as outlined in The models also created layers. In that view, the dynamo may operate in a relatively thin shell of exotic fluid perched above a dense rocky interior, a configuration that would be hard to reconcile with a simple “ice giant” caricature but fits comfortably within the new, more flexible framework.
Machine learning, hydrocarbons and the missing “diamond rain”
Another line of evidence challenging the old picture comes from simulations that track how complex mixtures of carbon, hydrogen and other elements behave at depth. One researcher used machine learning to simulate the behavior of exactly 540 atoms under the extreme conditions thought to exist inside Uranus and Neptune. However, last year, using machine learning, he simulated the behavior of 540 atoms and unexpectedly discovered that large hydrocarbon molecules form instead of the neat diamond crystals that earlier theories predicted, as reported in However. That result suggests that, rather than a simple “diamond rain” layer, the planets may host thick strata of complex hydrocarbons that alter both their density and their electrical properties.
As pressure and depth increase in those simulations, the hydrocarbon rich layer does not behave like a clean, crystalline solid but instead forms a more disordered, possibly fluid region. That has two important implications. First, it changes how much mass can be packed into a given radius, which feeds back into the interior models that try to match the gravity data. Second, it provides another potential source of conductive material for the magnetic dynamo, one that is compatible with a rock-heavy interior capped by volatile rich shells. The more I incorporate these machine learning results into the broader picture, the less plausible it seems that a single “ice” component can capture the diversity of phases and layers inside these planets.
From “ice giants” to “rocky giants” and what that means
As the modeling and simulations accumulate, some astronomers have started to argue that Uranus and Neptune the “ice giants” might be better described as something else entirely. One summary notes that astronomers have long called Uranus and Neptune the ice giants because models suggested their interiors were dominated by frozen volatiles, but that the new research points toward planets that could instead be called “rocky giants,” as discussed in Astronomers. The proposed rebranding is not just cosmetic. It signals a deeper shift in how planetary scientists think about the building blocks and formation histories of these worlds.
However, new research from the University of Zurich (UZH) and the National Centre of Competence in Research argues that the best fitting models are agnostic about whether ice or rock dominates, and that both categories remain viable within current uncertainties, as emphasized in However. That caution matters. It means that while the “rocky giant” label captures an intriguing possibility, the real lesson is that Uranus and Neptune occupy a broader compositional space than the old taxonomy allowed. Any new classification will have to accommodate that ambiguity rather than simply swapping one rigid label for another.
Why exoplanet science is watching closely
The debate over what to call Uranus and Neptune is not just an internal bookkeeping issue for Solar System specialists. Many of the exoplanets discovered so far fall into a similar mass and radius range, and they are often lumped into categories like “mini Neptunes” or “sub Neptunes” based on the assumption that they resemble our own ice giants. New research challenges Uranus and Neptune’s classification as ice giants, with some models suggesting most of their interiors could be rock rich, a point that has been widely discussed in New. If our local benchmarks are misclassified, then the shorthand astronomers use for distant worlds may also need a rethink.
That uncertainty cuts both ways. On one hand, it complicates efforts to infer exoplanet compositions from limited data, because a given mass and radius could correspond to a wide range of rock to ice ratios. On the other hand, it enriches the menu of possible worlds, suggesting that planets in the Uranus and Neptune regime might span a continuum from water dominated to rock dominated interiors. For mission planners and theorists, that diversity is a feature, not a bug. It means that a future flagship mission to Uranus or Neptune would not just tidy up a classification dispute, it would calibrate the entire exoplanet field by anchoring our models to a real, well measured example of whatever class these enigmatic worlds truly belong to.
What we still do not know about their hidden cores
Despite the flurry of new models, the cores of Uranus and Neptune remain out of direct reach, and key questions are still unresolved. The agnostic approaches that allow both ice rich and rock rich solutions highlight how much freedom remains in the interior structure, even when gravity, radius and magnetic data are all taken into account. One analysis notes that this agnostic approach did not fully pin down what Uranus and Neptune are really like, and that, for example, the rock to water ratio remains poorly constrained without new measurements, a limitation spelled out in For example, the rock. Without in situ probes or orbiters dedicated to mapping their gravity and magnetic fields in detail, the models will remain underdetermined.
That uncertainty extends to the very center of each planet. Are their cores compact, Earth sized balls of rock and metal, or more diffuse regions where rock, ice and gas intermingle in a gradual transition? Do stable layers suppress convection and trap heat, or are the interiors vigorously mixed? The latest studies sketch out plausible answers, but they also underline how much is still unverified based on available sources. Until a mission flies past the hazy blue clouds and listens closely to the gravitational and magnetic whispers from within, the true nature of Uranus and Neptune’s cores will remain an open question, and their proper place in the planetary family tree will stay up for debate.
More from MorningOverview