Almost thirty years have passed since the discovery of Pegasi 51 b, the first known planet in orbit around a star other than the Sun. Now, astrophysicists have identified more than 5,000 exoplanets, and at least 800 star systems that host multiple planets. Does this look like the solar system? To answer this question, we must look at the structure of the systems, i.e. the location of the orbits and the physical properties of their planets. The main trend observed is known as peas in a pod (literally: “peas in a pod”): the system’s planets have similar masses and diameters, and the spacing between their orbits is relatively uniform. But until now, astronomers have lacked the precise tools to describe all of the exoplanet systems. With that in mind, Lokesh Mishra, of the University of Bern, Switzerland, and his colleagues proposed categorizing them into four categories, according to the way the planets organize themselves there.
In this context, a system would be said to be “similar” if the masses of all the planets are comparable to it, and “ordered” if the planets are all closer to the star because they are light and “anti-order”. . Unlike that. Finally, if no trend appears in the overall distribution of a planetary system, it is considered “mixed”. The solar system, for example, with four small inner planets and many gas giants on the periphery, belongs to the class of ordered systems.
Of the known planetary systems, 59% are similar, 37% are ordered, 5% are mixed, and none are antipodal. However, these data are not statistically significant. In fact, to test this new framework, the researchers applied it to systems containing at least four planets of known mass, or just 41 planetary systems. Therefore, the researchers compared this classification with the classification of numerically simulated systems. Using a model of planet formation (called the “Berne model”), the team randomly built a large number of artificial systems. Among them, the proportion of similar systems rises to 80% while the proportion of ordered systems drops to 1.5%. Mixed and untidy structures complete the picture in equal parts.
The low number of observations alone is not sufficient to justify such discrepancies. In fact, ordered systems, which are assumed to be rare in simulations, account for more than a third of known systems, and conversely, no anti-ordered structure has been observed to date. How do we explain this difference between observations and simulations? On the one hand, there are many observational biases that limit knowledge of exoplanetary systems. For example, a planet becomes more difficult to detect the lower its mass and the farther it is from its star, which does not favor observing anticommand structures. However, the lack of a single planet is enough to distort the rating. For example, even though it’s ordered, the solar system would be jumbled if Neptune had a mass similar to Earth’s. On the other hand, nothing guarantees representation of planet formation models. However, in both observations and simulations, systems with similar structures are the majority, which reinforces the trend peas in a pod.
The main interest of this new framework is that it allows the structure of a system to be compared to its other properties and to infer correlations. For example, in simulations, the majority of planets composed of less than 1% water belong to similar systems while in anti-order systems, planets often contain more than 40% water. In the same way, systems containing at least one planet located in the habitable zone (at a distance from the star consistent with the possibility of liquid water on the surface) are for the most part of the same architecture.
Lokesh Mishra and colleagues next sought to determine whether correlations between the initial conditions of system formation and its eventual structure emerge in relation to their new study framework. Currently, theories of planet formation (as well as simulations) are based primarily on the core accretion model. According to this model, when a planetary system is formed, the solid particles in the disk of dust and gases orbiting the newly formed star coalesce to form the planetesimals. These objects will be able, if they reach a sufficient mass, to attract the surrounding gas towards themselves until they form an atmosphere or become gas giants. It is difficult to determine the influence of the initial conditions in the protoplanetary disk on the final structure of the system. During the first million years, the protoplanets regularly changed their orbits under the influence of the gravitational interaction exerted by one on the other. Thus, a young planetary system is the scene of many violent and unpredictable events such as collisions, eruptions, or even the fall of protoplanets onto the star! All these seemingly chaotic events are of a nature to modify the structure of the planetary system in the future.
However, interesting associations are clearly visible in the researchers’ findings. In particular, the amount of solid matter in the protodisk appears to be partly decisive. If it is less than the cut-off value of the order of Jupiter’s mass, then for the most part the system will adopt a similar structure, otherwise the other three classes will be distinguished. According to the researchers, too little solid matter in the disk would not allow the formation of protoplanets massive enough to become gaseous planets, the first players in the chaos that reigns around young stars.
Thus, the formation of similar systems will be directly related to the initial conditions, while the other three types of structures will be determined by random events. The team also noted that in simulations, a lower incidence of violent events is associated with a mixed architecture, and as the frequency of such events increases, the systems become anti-order and then order.
This new framework has yet to convince the scientific community before it can be adopted as a standard, especially since so many other classifications of planetary systems can be imagined. For the time being, if he shows his interest in highlighting the interesting connections between the structure of a planetary system and the properties of the planets that compose it, his conclusions are still drawn from the study of artificial systems whose representation cannot be guaranteed. One solution to assess the reliability of the highlighted associations is to apply the same tools to other models of planetary formations.