Constraining Exoplanet Interiors via Tidal Deformation: Actionable Strategies

For researchers working on exoplanet interiors, tidal deformation is one of the few observables that directly probes the density distribution and rheology of a planet. Unlike radius and mass, which give bulk averages, the tidal Love number k₂ and the associated phase lag encode information about the core size, mantle stiffness, and even the presence of a subsurface ocean. But turning a measured k₂ into a structural constraint is not straightforward. This guide lays out the practical strategies—what works, what fails, and how to avoid wasting time on models that cannot be distinguished by the data.

We assume you already know the basics of tidal theory. Our focus is on the decisions that make or break an inversion: which rheological model to adopt, how to handle eccentric orbits, and when tidal deformation alone is insufficient. We draw on common scenarios from the literature and our own experience in collaborative projects, but we avoid citing specific papers—the field moves fast, and the principles here are meant to outlast any single study.

Field Context: Where Tidal Deformation Fits in Real Work

Tidal deformation studies are most valuable when other interior probes are unavailable. For transiting planets, we often have mass and radius from radial velocity and transit photometry, but these yield only a mean density. To go further—to distinguish a rocky planet with a large iron core from one with a thick water layer—we need additional information. Tidal Love numbers provide that extra dimension because they depend on the radial distribution of density and rigidity.

In practice, k₂ is measured from the planet's gravitational quadrupole moment, which manifests as a deviation in the transit light curve (via orbital precession) or in the radial velocity signal (via apsidal precession). The measurement is challenging: the signal is small, often buried in stellar noise, and requires high-cadence, long-baseline observations. Teams typically target hot Jupiters and sub-Neptunes on short-period orbits, where tidal effects are strongest. But even then, the uncertainty on k₂ is often 20–50%, which limits the precision of interior constraints.

A key point is that tidal deformation is not a standalone tool. It works best when combined with other data: atmospheric composition (from transmission spectroscopy) can hint at the presence of volatiles, while thermal evolution models constrain the interior temperature. The real power comes from joint inversions that simultaneously fit the mass, radius, k₂, and possibly the orbital eccentricity and its rate of change. This multi-observable approach reduces degeneracies, but it also introduces model dependencies that must be handled carefully.

For example, consider a typical hot Jupiter with a measured k₂ of 0.3 ± 0.1. A simple two-layer model (core + envelope) might fit that value with a core mass fraction of 0.2–0.5, depending on the assumed equation of state. Adding a third layer—say, a water layer—widens the acceptable range. Without additional constraints, the inversion is underdetermined. The strategy, then, is to use tidal deformation as a discriminant: it can rule out certain compositions (e.g., a pure hydrogen-helium envelope with no core) even if it cannot uniquely identify the correct one.

Another real-world scenario involves planets with eccentric orbits. Tidal deformation in an eccentric orbit is time-variable, producing a periodic signal that can be separated from the static component. This adds complexity but also information: the amplitude and phase of the variable deformation depend on the planet's viscosity and internal friction, which are linked to the presence of liquid layers. For planets with suspected subsurface oceans (e.g., some sub-Neptunes), the tidal signal can indicate whether the ocean is coupled to the mantle or decoupled by an ice shell.

In summary, the field context for tidal deformation is one of complementarity. It is not a silver bullet, but it is a powerful lever when used alongside other data. The next sections detail how to build a robust workflow.

Foundations Readers Confuse

Several conceptual pitfalls trip up even experienced researchers when interpreting tidal deformation. The most common is conflating the static Love number k₂ with the dynamic response. The static k₂ describes the planet's equilibrium deformation under a steady tidal potential—it depends only on the density profile and the elastic properties (shear modulus). In contrast, the dynamic response includes time-dependent effects like tidal heating and orbital evolution, which involve viscosity and anelasticity. Many papers report k₂ as if it were a static quantity, but for planets with fluid layers (e.g., a liquid core or ocean), the effective k₂ can differ from the static value because the fluid does not support shear stress. This distinction is critical when comparing models to observations.

Another confusion is the role of the tidal quality factor Q. Often invoked to parameterize dissipation, Q is not a direct observable from k₂ alone. To infer Q, you need the phase lag (or the imaginary part of k₂), which requires measuring the orbital evolution or the thermal emission. Some teams mistakenly treat k₂ as a proxy for Q, but they are independent: a planet can have a high k₂ (soft) and a low Q (efficient dissipation), or vice versa. The relationship depends on the rheology.

Rheological models themselves are a source of confusion. The most common choices are the Maxwell model (elastic + viscous in series) and the Andrade model (which includes a creep term). The Maxwell model is simpler and often used for rocky mantles, but it predicts a constant Q at low frequencies, which is not physical for many materials. The Andrade model fits laboratory data better but introduces an extra parameter (the Andrade exponent) that is poorly constrained. Teams often default to Maxwell because it has fewer parameters, but this can bias the inferred interior structure. A better strategy is to test both models and see if the conclusions change—if they do, the data are insufficient to discriminate, and the uncertainty should be reported.

A third common mistake is ignoring the effect of rotation. Tidal deformation is often treated in the non-rotating limit, but for close-in planets, rotation can be fast enough to alter the shape and the Love numbers. The centrifugal potential adds a second harmonic component that mixes with the tidal potential. This is usually a small correction (∼10% for typical hot Jupiters), but it can become important when trying to measure k₂ at the 5% level. Some analysis pipelines include a rotational correction, but many do not, leading to systematic errors.

Finally, there is confusion about the information content of k₂ versus the moment of inertia (MoI). The MoI is a more direct probe of the radial density distribution, but it is rarely measurable for exoplanets (it requires knowing the precession rate, which is extremely small). Tidal Love numbers are a distant second best: they are sensitive to the density profile but also to the rigidity. For a given density profile, a stiffer planet has a lower k₂. This degeneracy can be broken if you have independent constraints on the shear modulus (e.g., from seismic data, which we do not have for exoplanets) or if you assume a composition. In practice, teams assume a mineralogical model (e.g., Earth-like mantle composition) and then vary the core size. This is reasonable but should be stated explicitly.

To avoid these pitfalls, we recommend a checklist: (1) separate static and dynamic effects; (2) treat Q as a derived quantity, not an input; (3) test at least two rheological models; (4) include rotation if the orbital period is less than a few days; (5) report the assumed shear modulus and its uncertainty.

Patterns That Usually Work

Over the past decade, a set of best practices has emerged for using tidal deformation to constrain interiors. These patterns are not foolproof, but they have proven effective across a range of planet types.

Joint inversion with mass and radius

The most reliable pattern is to perform a joint inversion of mass, radius, and k₂. This is typically done using a Markov chain Monte Carlo (MCMC) sampler that explores a parameterized interior model (e.g., layers of iron, rock, water, and gas). The likelihood includes all three observables, with their uncertainties. The key is to use a realistic equation of state for each layer—for example, the SESAME database for rock and iron, and the Saumon-Chabrier or ANEOS for hydrogen-helium. Many teams use the ExoPlex or PlanetInterior codes, which are publicly available. The joint inversion often reduces the degeneracy between core size and mantle composition, especially when k₂ is measured to better than 20%.

Using multiple tidal harmonics

For eccentric orbits, the tidal potential contains multiple harmonics (e.g., the principal frequency and its overtones). Each harmonic produces a different deformation pattern, and the ratio of their amplitudes can constrain the interior viscosity. This is a more advanced technique but has been applied to a few systems (e.g., the hot Jupiter HD 209458b). The practical challenge is that the higher harmonics are weaker and harder to detect. However, if the signal-to-noise is sufficient, this pattern can distinguish between a fully solid planet and one with a liquid core—something that the static k₂ alone cannot do.

Thermal-tidal coupling

Another pattern that works well is to combine tidal deformation with thermal evolution models. The interior temperature affects the viscosity and the shear modulus, which in turn affect k₂. Conversely, tidal heating provides an energy source that can delay cooling. By modeling the coupled system, you can place joint constraints on the interior structure and the thermal state. This is particularly useful for planets that are likely tidally locked and experiencing significant heating, such as hot Jupiters on slightly eccentric orbits. The coupling introduces additional parameters (e.g., the initial entropy, the tidal heating rate), but it also breaks degeneracies: for example, a planet with a high k₂ and a high thermal flux is more likely to have a molten interior than one with a low thermal flux.

Hierarchical modeling

A practical workflow that many teams adopt is hierarchical modeling: start with a simple two-layer model to get a rough constraint on the core size, then add layers (e.g., a water layer or an atmosphere) and check if the fit improves significantly using a Bayesian information criterion (BIC) or similar. This avoids overfitting while still exploring complexity. The hierarchy should be chosen based on the planet's likely composition (e.g., for a sub-Neptune, a three-layer model with rock, water, and H/He is natural). The key is to report the evidence for each model, not just the best fit.

Validation with synthetic data

Before applying the inversion to real data, it is wise to test the pipeline on synthetic data. Generate a mock planet with known interior parameters, compute its k₂ (including realistic noise), and then run the inversion to see if you recover the input. This reveals biases in the model or the sampling. For example, if the inversion systematically overestimates the core size when using a Maxwell rheology, you know to include a correction or switch to Andrade. Many teams skip this step, but it is one of the most cost-effective ways to improve reliability.

These patterns are not exhaustive, but they represent the consensus from successful studies. The common thread is to combine multiple data sources, test model assumptions, and validate the methodology.

Anti-Patterns and Why Teams Revert

Not every approach to tidal deformation works. Some patterns are so problematic that teams often revert to simpler methods after trying them. Here are the most common anti-patterns.

Overinterpreting a single k₂ measurement

The most frequent mistake is to take a measured k₂ at face value and draw strong conclusions about the interior. For example, a k₂ of 0.3 might be interpreted as evidence for a large core, but the same value could arise from a small core with a stiff mantle or a large core with a soft mantle. Without additional constraints, the inference is non-unique. Teams that fall into this trap often have to walk back their claims when more data arrive. The remedy is to always report the range of models consistent with the measurement, not just the best fit.

Using a constant Q model

Many early studies assumed a constant tidal quality factor Q for the planet, but this is physically unrealistic. Laboratory experiments and geophysical observations show that Q depends on frequency, temperature, and composition. Using a constant Q can lead to errors in the predicted orbital evolution and the inferred interior viscosity. Teams that try this often find that their models cannot simultaneously fit the k₂ and the orbital decay rate (if measured). The better approach is to use a frequency-dependent rheology, even if it adds parameters.

Ignoring the star's contribution

The tidal deformation of the planet is not the only source of the observed signal. The star itself can be deformed by the planet's gravity, and the stellar tide can contaminate the measurement of k₂. This is especially problematic for planets around active stars, where starspots and faculae introduce photometric variability. Some teams subtract the stellar contribution using a model of the star's tidal response, but this model itself depends on uncertain stellar parameters. A common anti-pattern is to assume the star is a point mass, which is only valid if the planet is much less massive than the star. For hot Jupiters around Sun-like stars, the stellar tide is small but not negligible. The safest approach is to include the stellar tide in the model and marginalize over the stellar parameters.

Fitting too many parameters

With a limited dataset (e.g., one k₂ value with 20% error), it is tempting to fit a complex interior model with many layers. This leads to overfitting and large uncertainties. Teams often revert to simpler models after realizing that the data cannot constrain the extra parameters. The rule of thumb is to have at most one free parameter per independent observable. For a typical case with mass, radius, and k₂, you can constrain at most three parameters (e.g., core mass fraction, mantle composition, and water layer thickness). Adding more parameters without additional data is counterproductive.

Neglecting orbital evolution

Tidal deformation is not static: it causes orbital evolution (eccentricity damping, semi-major axis decay). Some studies treat k₂ as a snapshot and ignore the fact that the planet's orbit is changing. This can lead to inconsistencies if the orbital parameters are measured at different epochs. For example, a planet with a measured eccentricity of 0.01 and a k₂ of 0.5 might be undergoing rapid circularization, meaning the eccentricity was higher in the past. Ignoring this time dependence can bias the inferred interior properties. The anti-pattern is to treat the system as time-invariant; the better practice is to model the coupled orbital-thermal evolution.

Teams that avoid these anti-patterns tend to produce more robust results. If you find yourself tempted by any of these, step back and consider whether the data truly support the complexity.

Maintenance, Drift, and Long-Term Costs

Once you have a tidal deformation model and an inferred interior structure, the work is not over. Models drift as new data arrive, and maintaining consistency over time requires deliberate effort.

Updating with new observations

As more transits are observed, the measurement of k₂ improves. Initially, the uncertainty might be 30%, but after several years of TESS or PLATO data, it could drop to 5%. When that happens, the interior model that fit the old data may no longer be valid. The cost is that you need to re-run the inversion, which can be computationally expensive if you use MCMC. A practical strategy is to archive the posterior samples from the previous inversion and use them as a prior for the new one. This speeds up convergence and ensures continuity.

Model drift from improved equations of state

The equations of state used for planetary materials are constantly being refined. For example, the high-pressure behavior of water and rock is an active area of research. If you published a model in 2020 using one EOS, a 2025 update might shift the inferred core size by 10%. This is not a flaw in the method; it is a natural consequence of scientific progress. To manage this, document the exact EOS version used and re-evaluate periodically. Some teams maintain a living document that tracks updates and their impact on published results.

Computational costs

Running joint inversions with full MCMC and multiple rheological models is computationally intensive. A typical run might take days on a cluster. As the number of planets studied grows, the total cost becomes significant. Teams often have to prioritize which planets to analyze in detail. One way to reduce costs is to use surrogate models (e.g., neural networks) that approximate the forward model. These can speed up the inversion by orders of magnitude, but they introduce their own errors. The trade-off is between accuracy and speed.

Team expertise turnover

The people who built the original model may move on to other projects. New team members need to understand the assumptions and the codebase. Poor documentation can lead to mistakes: for example, a new graduate student might use the wrong rheological model because the code comments are ambiguous. The long-term cost is that the model becomes a black box that no one fully understands. To mitigate this, invest in clear documentation, version control, and automated tests. It is also helpful to write a short paper or technical note describing the methodology, even if it is not published in a journal.

Data archive decay

The raw data (light curves, radial velocity measurements) are often stored in public archives, but the processed data products (e.g., the derived k₂ values) may not be. If the original analysis code is lost, it becomes difficult to reproduce the results. This is a known problem in exoplanet science. The solution is to deposit the final k₂ measurements and the posterior samples in a repository like Zenodo or the NASA Exoplanet Archive. This ensures that future researchers can build on your work without having to redo the entire analysis.

Maintenance is an ongoing cost that is often underestimated. Planning for it from the start saves time and frustration later.

When Not to Use This Approach

Tidal deformation is not always the right tool. There are clear situations where the effort is not justified by the return.

Planets with long orbital periods

For planets with orbital periods longer than about 10 days, the tidal signal is extremely small. The equilibrium tide amplitude scales as (R_p/a)³, so for a planet at 0.1 AU, the amplitude is about 1000 times smaller than for a hot Jupiter at 0.01 AU. Measuring k₂ for such planets is currently impossible with existing instruments, and even with future missions like PLATO, it will be challenging. If you are studying a warm Jupiter or a temperate super-Earth, tidal deformation is not a viable constraint. Focus on other methods like transmission spectroscopy or astrometry.

Planets with large uncertainties in mass or radius

If the mass and radius are poorly known (e.g., >20% uncertainty), adding k₂ will not help much. The degeneracies are too large. In such cases, it is better to wait for better data rather than fitting a complex model that will yield uninformative posteriors. A quick test: if the mass uncertainty alone leads to a factor of two range in mean density, then k₂ will not break the degeneracy.

When the star is too active

Stellar activity introduces noise that can mimic or mask the tidal signal. If the star has a high rotation rate or frequent flares, the photometric variability may swamp the tiny tidal deformation signal. In these cases, even a long baseline of observations may not yield a reliable k₂. Some teams attempt to correct for activity using Gaussian processes, but the correction itself introduces uncertainty. If the activity level is high, it is often better to exclude the planet from tidal analysis until better data are available (e.g., from a quieter star).

When the planet is not tidally locked

The standard tidal theory assumes that the planet is in a pseudo-synchronous rotation state, which is the equilibrium for a planet on an eccentric orbit. If the planet is not tidally locked (e.g., because it is very young or has a massive moon), the deformation pattern is more complex and the static Love number approximation breaks down. In such cases, you would need to model the full spin-orbit evolution, which adds many free parameters. Unless you have independent constraints on the rotation state, it is best to avoid tidal deformation analysis.

When the goal is to test a specific hypothesis

Sometimes the question is not