It’s a natural question to ask, after exploring the Solar System, why is there such a large gap between the sizes of Earth and Neptune? With the discovery of exoplanets, most of the gap has been filled, but there is one region, around 1.5 to 2 Earth radii, where an actual “gap” in the distribution of discovered exoplanets exists. Why are there surprisingly few planets of such sizes? It’s an interesting question for researchers since the gap is found in the distribution, and we’ll explore some of the explanations in this article.
Looking At the Data
Contrary to what its name suggests, if you look at the frequency distribution of planet sizes (shown on the graph above), you would notice that there are still quite a number of planets with radii of 1.5 to 2 Earth radii, around where that “gap” should be. The value doesn’t drop to near zero, and the “gap” is instead closer to a “valley”, where the value in the distribution is “unusually low” compared to any “normal-looking” smooth curve over the data distribution. This is important as this phenomenon does not say that planet formation within that size range is forbidden. It just says that you are less likely to see a planet of that size than we might expect from simple statistical models. This message is essential as we explore the potential reasons for why this “anomaly” arises in the data distribution.
Mass Loss by Photoevaporation
One of the proposed reasons why this “small radius planet gap” exists is because of the photoevaporation phenomenon. Gas particles are already moving very fast at the temperatures we are experiencing today, and if a high-energy radiation particle strikes the gas particle, it could cause the gas particle to gain enough kinetic energy and thus speed to escape from the planet’s gravitational pull.
With the X-ray radiation or extreme ultraviolet radiation (EUV), most likely emitted by the star, the gas particles can get activated just enough to escape the gravitational pull of planets with radii originally in that gap. This often results in a shrinkage of planet sizes until the radiation is not strong enough to allow more gas to escape (remember the escape velocity increases with decreasing distance from the center of mass). But for more massive planets, the gravitational pull is stronger and thus it is not as easy to cause gases to escape them. Therefore, their masses are often kept on the higher end of the spectrum.
Core-Powered Mass Loss
Other than photoevaporation, it is possible that the temperature left over from the original formation of the cores of the planets could provide enough energy for these gas particles to escape. In particular, the cores of planets during formation are usually at their hottest, and they cool down after that point. In fact, as the core of older planets (billions of years old) is still cooling down gradually, it could still shed more and more of its atmosphere with the radiation it emits internally from its heat. [1]
Giant Impacts
Moreover, giant impacts can cause a planet to lose significant amounts of atmosphere. With the power to disrupt the entire structure of the planet, these impacts can fling large amounts of gas particles out of the planet’s atmosphere, sometimes even causing the planet’s size to shrink significantly, especially if the thick atmosphere contributes a lot to the planet’s radius. [2]
What Do These Mechanisms Have in Common?
Both of the mechanisms above, relying on loss of atmospheric mass, rely on an assumption from the distribution of planets, particularly in the borderline region between those of rocky planets and gaseous planets. For this to work, most of the planets in that size region must have very thick atmospheres, analogous to the gaseous planets. However, some very large super-Earths have been found. For example, Kepler-10c has a radius of about 2.3 times that of Earth, on the upper end of that gap, but it is believed to be a terrestrial planet with high density instead of a (relatively) small gas giant.
This could help provide some insights about how super-Earths (large terrestrial planets) and mini-Neptunes (small gas giants) are distributed in these borderline regions, where a planet could exist in either form if all we know is its radius. But before we do all the measurements, we should explore some alternate mechanisms causing this valley in the distribution to make sure that no other factors are affecting the data.
Formation-based Theories
Another explanation for this phenomenon could be that the planet formation process is biased towards producing planets on the lower or the upper end of the valley, but not within the valley. It’s possible that planetary formation just doesn’t favor planet formation of this kind of size. If planet formation in that size range means having to stuff a very thick atmosphere onto a rocky planet, there may simply not be enough gas for the smaller planets to accrete after the larger planets have “taken” it during the formation of the planetary system.
Evaluating the Theories
How do researchers investigate the effects of each of these mechanisms on the distribution of planet sizes? They turn to the age of the system. The three mechanisms mentioned above occur in different timescales. The formation-time deficit in the gases left around to accrete onto the small planets take place right after the star has formed. The photoevaporation mechanism is slower, on the timescales of tens of millions of years. The core-powered mass loss lasts even longer, taking billions of years. [1] By looking at the distribution of planet radii in stars of different ages, we can get some information about the timescale of the actual mechanism involved, thus enabling us to identify the dominating mechanism.
In 2023, a study conducted using data in the Praesepe and Hyades open clusters provided evidence that core-powered mass loss might have been the dominating factor. [3] In particular, they analyzed the hot sub-Neptune planets, planets with very thick atmospheres that could have a radius inside that valley. The planetary systems were split into two groups in the two star clusters, one with ages of 600 to 800 million years, and the other one with ages of 3 to 9 billion years. They found that the occurrence rates of these hot sub-Neptunes were more common in the younger cluster than in the older cluster, while there isn’t much difference between that of the young TESS FGK stars and the younger cluster. This provides evidence that the decrease in planet sizes occur on timescales more similar to that of core-powered mass loss than photoevaporation.
This suggests that the radius valley involves very long-term processes
Source: https://arxiv.org/pdf/2311.18709, shared under the CC BY 4.0 license https://creativecommons.org/licenses/by/4.0/
But another study in 2024 provided evidence that planetary systems with planets in that valley have very different planet size distributions. [4] In particular, many planetary systems contain planetary configurations known as “peas-in-a-pod”, in which multiple planets have similar sizes to each other. The relative deficit of this kind of configuration in systems containing planets in the gap suggests that the involved systems had a more violent history, suggesting that dynamical mechanisms like giant impacts might play a role in making the radius valley.
Conclusion
In this article, we’ve explained what the small planet radius gap is, the potential explanations for it, and how the effects of these mechanisms can be studied by looking at the data.
References
- (2023, November 15). NASA Data Reveals Possible Reason Some Exoplanets Are Shrinking. Retrieved January 31, 2025, from https://www.jpl.nasa.gov/news/nasa-data-reveals-possible-reason-some-exoplanets-are-shrinking/
- Durham University. (2020, July 15). Supercomputer reveals atmospheric impact of gigantic planetary collisions. Retrieved January 31, 2025, from https://phys.org/news/2020-07-supercomputer-reveals-atmospheric-impact-gigantic.html#google_vignette
- J. L. Christiansen et al. (2023, November 30). Scaling K2 VII: Evidence for a high occurrence rate of hot sub-Neptunes at intermediate ages. Retrieved January 31, 2025, from https://arxiv.org/pdf/2311.18709
- Q. Chance, S. Ballard. (2024, October 3). Evidence that Planets in the Radius Gap Do Not Resemble Their Neighbors. Retrieved January 31, 2025, from https://arxiv.org/pdf/2410.02150