Hacker News Reader

Comment by semaphoreP | Hacker News Reader

This title is a bit imprecise. They detected four planets with lower bound on their masses to be down to 1.7 Earth masses. Because these planets don't transit, there are no direct measurements from their radius. They can use mass-radius relations to infer the radius of these planets, but the key finding is their masses (actually lower bounds on their masses).

View on HN

Can you say they _don't_ transit? They say the planets were detected by analyzing star wobbles, which doesn't necessarily mean that the planets don't transit, just that it's not how they were detected.

Well it doesn't transit from our point of view, so yes, I think he can say that as the observer.

"In astronomy, a transit or astronomical transit is the phenomenon of at least one celestial body appearing to move across the face of another celestial body, hiding a small part of it, as seen by an observer at some particular vantage point."[0]

[0]https://en.wikipedia.org/wiki/Transit_(astronomy)

I understand that. I guess I don't know how small we can reliably view transits. We can certainly see large planet transits. Perhaps planets of this size are too small to view at this distance/ with this star type with our current technology. Just because we haven't observed a transit doesn't mean the transit isn't happening. But again, I don't really know this specific circumstance.

I guess you're right. I know there aren't any observed transits, but I also don't know what the current constraints from monitoring the star's brightness for transit is (the star is actually so bright that it becomes hard to monitor for planet transits).

However, we have a good prior on the inclination of these planets, because we know the inclination of the dust disk around the star (https://www.scientificamerican.com/article/tau-ceti-s-dust-b...), and it is likely the planets are at a similar inclination. Because the disk isn't edge on, the planets also likely aren't, and won't transit.

Since they are watching the star wobble, they probably have exact inclinations.

Seems like it would have to be. If you've worked out from the wobble what the mass and orbit are it would stand to reason you've either calculated or at least stated an assumption regarding the inclination.

You can put a lower bound on both, but inclination introduces an unknown upper bound.

But shouldn't they have _some_ upper bound ? I guess for example that it's unlikely they are larger than 100 Earth masses since they call them Earth-sized.

Measuring the star wobble (i.e. doppler shift) doesn't give inclinations.

"Wobble" is perhaps a imprecise term. What is actually measured is the doppler shift of a spectral line in the star. In other words, you are measuring the velocity of the star in the radial direction (towards and away from the Earth). By measuring how the velocity changes over time, you can get a orbital period for the planet. By measuring the magnitude of the velocity change, you can get a lower bound on the mass of the planet. It is only a lower bound as depending on the orbital inclination, some of the movement will be in the perpendicular direction (back and forth on the sky). We are unable to measure this movement precisely enough to detect (in most cases).

Yeah, but the Kepler observatory has looked for exoplanet transits for many years now and found over 1000.

You have to assume it has examined this close neighbor thoroughly.

Kepler has only focused on two tiny patches of the sky in its lifetime IIRC.

Actually this star is not in the Kepler Field, and it is also too bright for Kepler. Even most ground based telescopes looking for transits probably haven't bothered looking at it, due to its brightness.

Thanks, I had no idea Kepler was under these constraints.

I'd expect it's easier to measure at the brighter stars. Maybe calibrating the instrument for the weaker stars makes it "overload" for a really bright ones?

You're in for a treat, "Kepler 2.0" launches next year: https://tess.gsfc.nasa.gov/overview.html

Reply on news.ycombinator.com