We combine analytical understanding of resonant dynamics in two-planet systems with machine learning techniques to train a model capable of robustly classifying stability in compact multi-planet systems over long timescales of $10^9$ orbits. Our Stability of Planetary Orbital Configurations Klassifier (SPOCK) predicts stability using physically motivated summary statistics measured in integrations of the first $10^4$ orbits, thus achieving speed-ups of up to $10^5$ over full simulations. This computationally opens up the stability constrained characterization of multi-planet systems. Our model, trained on $\approx 100,000$ three-planet systems sampled at discrete resonances, generalizes both to a sample spanning a continuous period-ratio range, as well as to a large five-planet sample with qualitatively different configurations to our training dataset. Our approach significantly outperforms previous methods based on systems' angular momentum deficit, chaos indicators, and parametrized fits to numerical integrations. We use SPOCK to constrain the free eccentricities between the inner and outer pairs of planets in the Kepler-431 system of three approximately Earth-sized planets to both be below 0.05. Our stability analysis provides significantly stronger eccentricity constraints than currently achievable through either radial velocity or transit duration measurements for small planets, and within a factor of a few of systems that exhibit transit timing variations (TTVs). Given that current exoplanet detection strategies now rarely allow for strong TTV constraints (Hadden et al., 2019), SPOCK enables a powerful complementary method for precisely characterizing compact multi-planet systems. We publicly release SPOCK for community use.

Fast radio bursts (FRBs) are, as the name implies, short and intense pulses of radiation at wavelengths of roughly one metre. FRBs have extremely high brightness temperatures, which points to a coherent source of radiation. The energy of a single burst ranges from $10^{36}$ to $10^{39}$ erg. At the high end of the energy range, FRBs have enough energy to unbind an earth-sized planet, and even at the low end, there is enough energy to vaporise and unbind the atmosphere and the oceans. We therefore propose that FRBs are signatures of an artificial terraformer, capable of eradicating life on another planet, or even destroy the planet entirely. The necessary energy can be harvested from Wolf-Rayet stars with a Dyson sphere ($\sim 10^{38}$ erg s$^{-1}$) , and the radiation can be readily produced by astrophysical masers. We refer to this mechanism as Volatile Amplification of a Destructive Emission of Radiation (VADER). We use the observational information to constrain the properties of the apparatus. We speculate that the non-repeating FRBs are low-energy pulses used to exterminate life on a single planet, but leaving it otherwise intact, and that the stronger repeating FRB is part of an effort to destroy multiple objects in the same solar system, perhaps as a preventative measure against panspermia. In this picture, the persistent synchrotron source associated with the first repeating FRB arises from the energy harvesting process. Finally we propose that Oumuamua might have resulted from a destruction of a planet in this manner.

Many previous authors have attempted to find explanations for Westeros's climate, characterized by a generally moderate, Earth-like climate punctuated by extremely long and cold winters, separated by thousands of years. One explanation that has been proposed is that the planet orbits in a Sitnikov configuration, where two equal-mass stars (or a star and a black hole) orbit each other on slightly eccentric orbits, and the planet moves along a line through the barycenter perpendicular to the primaries' orbital plane (Freistetter & Gr\"utzbauch 2018). We modify an intermediate-complexity GCM to include the effects of such an orbit and integrate it for thousands of years to determine whether such an orbit can a) be habitable and b) explain the climatic variations observed by the inhabitants of Westeros, in both double-star and star-black hole configurations. While configurations with low primary eccentricity and initial conditions that permit only small excursions from the ecliptic plane are habitable, these orbits are too stable to explain Westerosi climate. We find that while orbits with more bounded chaos are able to produce rare anomalously long and cold winters similar to Westeros's Long Night, huge variations in incident stellar flux on normal orbital timescales should render these planets uninhabitable. We note that the presence of an orbital megastructure, either around the planet or the barycenter, could block some of the sunlight during crossings of the primaries' orbital plane and preserve Westeros's habitability. While we find that bounded chaotic Sitnikov orbits are a viable explanation for Westeros's Long Night, we propose that chaotic variations of the planet's axial tilt or semimajor axis, potentially due to torques from nearby planets or stars, may be a more realistic explanation than Sitnikov orbits.

Many of the multi-planet systems discovered to date have been notable for their compactness, with neighbouring planets closer together than any in the Solar System. Interestingly, planet-hosting stars have a wide range of ages, suggesting that such compact systems can survive for extended periods of time. We have used numerical simulations to investigate how quickly systems go unstable in relation to the spacing between planets, focusing on hypothetical systems of Earth-mass planets on evenly-spaced orbits (in mutual Hill radii). In general, the further apart the planets are initially, the longer it takes for a pair of planets to undergo a close encounter. We recover the results of previous studies, showing a linear trend in the initial planet spacing between 3 and 8 mutual Hill radii and the logarithm of the stability time. Investigating thousands of simulations with spacings up to 13 mutual Hill radii reveals distinct modulations superimposed on this relationship in the vicinity of first and second-order mean motion resonances of adjacent and next-adjacent planets. We discuss the impact of this structure and the implications on the stability of compact multi-planet systems. Applying the outcomes of our simulations, we show that isolated systems of up to five Earth-mass planets can fit in the habitable zone of a Sun-like star without close encounters for at least $10^9$ orbits.

Using images from the Hubble Space Telescope Advanced Camera for Surveys, we measure the rate of cooling of white dwarfs in the globular cluster 47 Tucanae and compare it to modelled cooling curves. We examine the effects of the outer convective envelope reaching the nearly isothermal degenerate core and the release of latent heat during core crystallisation on the white dwarf cooling rates. For white dwarfs typical of 47 Tuc, the onset of these effects occur at similar times. The latent heat released during crystallisation is a small heat source. In contrast, the heat reservoir of the degenerate core is substantially larger. When the convective envelope reaches the nearly isothermal interior of the white dwarf, the star becomes brighter than it would be in the absence of this effect. Our modelled cooling curves that include this convective coupling closely match the observed luminosity function of the white dwarfs in 47 Tuc.

The requirement that planetary systems be dynamically stable is often used to vet new discoveries or set limits on unconstrained masses or orbital elements. This is typically carried out via computationally expensive N-body simulations. We show that characterizing the complicated and multi-dimensional stability boundary of tightly packed systems is amenable to machine learning methods. We find that training an XGBoost machine learning algorithm on physically motivated features yields an accurate classifier of stability in packed systems. On the stability timescale investigated ($10^7$ orbits), it is 3 orders of magnitude faster than direct N-body simulations. Optimized machine learning classifiers for dynamical stability may thus prove useful across the discipline, e.g., to characterize the exoplanet sample discovered by the upcoming Transiting Exoplanet Survey Satellite (TESS). This proof of concept motivates investing computational resources to train algorithms capable of predicting stability over longer timescales and over broader regions of phase space.