amann.bsky.social
Associate professor of Physics and Astronomy at UNC Chapel Hill studying young exoplanets and stars. Dad to one human and one cat. 🏳️🌈💗💜. Carrboro Citizen. http://andrewwmann.com
308 posts
5,132 followers
468 following
Prolific Poster
Conversation Starter
comment in response to
post
Aside from the planet, there's a lot of work in the paper on updating the parent population age. This is something out group has been working on a lot lately - how to get better ages by combining a mix of age estimates (lithium, gyro, variability, lithium, CMD, etc). 🌌
comment in response to
post
We did detect such a signal (grey dot below). But it's below our SNR cut. Meaning we don't consider it significant. Unofficially, the signal looked really unconvincing. More TESS data would help here.
comment in response to
post
As we kind of expected, the new planet is really close to an orbital resonance with the existing ones, as has been the trend for young systems.
Interestingly, it looks like one could fit another planet in between the two innermost ones. Indeed, such a planet would make a perfect set.
comment in response to
post
That's a known problem with the TESS PDCSAP (default curves), they 'overfit' and weaken some planets, especially around young stars.
comment in response to
post
Because of this, we have an active survey just looking for smaller 'missed' planets in known young systems.
One thing to highlight here is how important the light curves are for this work. We have our own search algorithm, but it would have missed the planet if not for better light curves
comment in response to
post
We also worry a lot about detection sensitivity - what is the smallest planet we can detect at a given age.
Lastly, multi-planet systems are special, in part because young planets are generally found in resonant orbits which generate TTVs -> this is the best chance at precise masses.
comment in response to
post
When you are searching for young planets, you worry a lot about statistical biases. Like, are the young planets big because it's hard to find the small ones. Turns out, no, they are really bigger, but it's the right question to ask.
comment in response to
post
but where are those grants coming from? If it's endowment, maybe, but some of that might be coming out of the XX% overhead on other grants. Even if endowment, they might have to use that to cover new gaps.
comment in response to
post
I think any institution that depends on grants is in serious trouble, regardless of size.
comment in response to
post
I should note that the paper was not meant to be 'here is how to get the best periods', more 'this is what you can expect using the most common methods' (e.g., Lomb-Scargle).
A future paper will investigate best-methods for trying to get longer-period rotations. Assuming we ever figure it out...
comment in response to
post
One of our goals is to map out completeness/reliability in order to build a really large sample of stars with rotation measurements from TESS (like millions of stars) and to help find members of young clusters and associations. Completeness and reliability are very important for such large samples.
comment in response to
post
Interestingly, combining sectors doesn't seem to help. Merging consecutive sectors, for example, actually makes it harder to recover some periods. We think this is because you are stacking systematics (e.g., scattered light) more than stacking astrophysics.
comment in response to
post
After ~14 days, periods become very unreliable. Not much better than random guessing! This makes sense since we are only using single sectors (27 days).
comment in response to
post
As you might expect, short-periods are also pretty reliable. If the LS power is high, reliability is over 90%, and over 95% if you consider half-period aliases (common). Digging deeper, many of the remaining targets are binaries (so both periods might be 'correct') or have other complications.
⭐🔭🧪🌌
comment in response to
post
To the first question. TESS periods are remarkably precise. For fast rotators the periods are good to a few percent! This method should account for both astrophysical (e.g., spot evolution and differential rotation) and measurement uncertainties.
⭐🔭🧪🌌
comment in response to
post
Our main tool is a set of stars with both K2 and TESS data. We use the K2 periods as 'ground truth'
⭐🔭🧪🌌