# 3 years later, how strong is the evidence for planet 9?

I came across post in quora where somebody said that "most astronomers no longer think Planet 9 exists" and being somewhat shocked by that, I did a google search and found a fairly recent article that claims exactly that.

It can take several years to prove a negative within pretty good certainty, as opposed to a discovery which can be checked pretty quickly. Because I don't want to wait several years, I wanted to ask if the general consensus is that Planet 9 probably doesn't exist? Or is there still a respectable amount of belief among astronomers in the original evidence and any new evidence that's followed?

If it ends up not existing, I'll be sorry to hear it. It was a fun story when it lasted.

• I thought planet 9 neatly explained the tilt of the ecliptic with respect to the Sun's equator, so I too would be sorry to hear of planet 9's demise. – BillDOe Jan 27 '19 at 21:29
• @BillDOe The tilt could just as easily be explained by a similar sized planet that is no longer in orbit around the sun, so I don't think that was every a strong argument in favor. – userLTK Jan 27 '19 at 21:36
• @userTLK, granted, but I thought the explanation involved a relatively small force acting over billions of years, so if a similar planet was ejected from the solar system, it would have had to have occurred recently in an astronomical time scale. – BillDOe Jan 27 '19 at 21:40
• @BillDOe That might make a good separate question. I don't see why that would be necessary, but maybe it is. – userLTK Jan 31 '19 at 6:42
• What would make a separate question: whether planet 9 may hypothetically be responsible for the tilt of the ecliptic or if such a planet would have taken billions of years to accomplish the tilt? – BillDOe Jan 31 '19 at 20:32

## 1 Answer

It's worth noting that Siegel's article seems to extrapolate from one paper (Shankman et al., 2017) to "most scientists" without much justification, so it's not particularly clear what most scientists think anyway (and that's before we get into the question of whether the "most scientists" referred to includes scientists whose discipline is not astronomy-related). Usually in these kind of situations there's a lot of back-and-forth as new objects are found in the outer solar system and the analyses of systematic errors and biases in the various surveys are refined. If there is an actual detection of Planet 9, that would resolve things fairly definitively, the reverse situation is going to be rather more difficult to prove.

In response to the OSSOS paper (Shankman et al., 2017) Michael Brown and Konstantin Batygin released a new analysis of the clustering in the outer solar system, taking into account the OSSOS results that are the subject of Ethan Siegel's article. The paper can be downloaded from the arXiv. Brown also has a Twitter thread summarising the result. From the paper, their conclusion on the apparent absence of clustering in OSSOS:

That is, the uncertainties in the measurement of clustering from the OSSOS data are so large that OSSOS would not be capable of confidently detecting the clustering seen in the larger data set even if it were real and present in the OSSOS data. Because of the limited survey region and small number of detected objects, OSSOS observations are equally consistent with being drawn from a uniform distribution of longitudes of perihelion and with being clustered in longitude of perihelion as strongly as seen in the ensemble data. No conclusions on clustering of longitude of perihelion observed in the complete dataset can be drawn from the OSSOS data.

They do note that this does not necessarily prove the existence of Planet 9, but in the absence of Planet 9 there would still need to be an explanation for the clustering.

Batygin et al. subsequently posted a review of the Planet Nine hypothesis on the arXiv, favouring a lower planetary mass and a less eccentric orbit than originally proposed. The new parameters do not favour Planet 9 as an explanation for the solar obliquity.

On the anti-Planet 9 side of things, it's worth looking into Kavelaars et al. 2019, a subsequent paper from the same authors as the study based on the OSSOS data, who note that there seems to be an absence of objects with perihelia ($$q$$) between 50 and 75 AU that would be expected if a planet was causing the alignment, which so far doesn't seem to be explainable by detection bias.

A survey that might have detected a TNO at $$a ∼ 500\ \mathrm{au}$$ and $$q ∼ 75\ \mathrm{au}$$ is actually more likely to have detected objects with similar $$a$$ but smaller values of $$q$$. The same is true of the other $$q > 75\mathrm{au}$$ detections: the lower-$$q$$ but similar $$a$$ detections are always more likely. Thus, the lack of detections in the $$50 < q < 75\ \mathrm{au}$$ range may be indicating that there really is an absence of TNOs on orbits in this range. This strongly contradicts models of orbital evolution that include an additional planet, as the gravitational action of such an object would cause TNOs to be distributed across a range of $$q$$ values at any given moment (Figure 2; Shankman et al. 2017a; Lawler et al. 2017). Thus, if the lack of objects in the $$50\ \mathrm{au} < q < 70\ \mathrm{au}$$ range is real, the hypothesized external planet can be excluded.

It still is an active area of research, so it remains to be seen how the various studies will hold up.

• A related point is that even if the clustering is real, some models can get it to come from masses distributed over the Kuiper belt, rather than concentrated in a single planet. So I don't think we can confidently claim planet 9 exists until it is seen, and we can't confidently claim it doesn't exist until null results have ruled out its potential parameter space. Credit for discovery generally goes to the one who sees it. – Ken G Jan 28 '19 at 18:03
• I wonder if the Gaia mission will give us better data to help answer this?? – Jack R. Woods Feb 11 '19 at 18:18
• @JackR.Woods - that might be a good subject for its own question :-) – antispinwards Feb 11 '19 at 18:51