Read on Twitter
My new paper is out now! https://arxiv.org/abs/2011.07075
Quick summary: @NASAWebb will be able to *directly* detect sub-Jupiter mass exoplanets at wide separations across a broad sample of objects, an order of magnitude improvement over current instruments.
Keep reading for more info!
Quick summary: @NASAWebb will be able to *directly* detect sub-Jupiter mass exoplanets at wide separations across a broad sample of objects, an order of magnitude improvement over current instruments.
Keep reading for more info!
Before I start, it's super-important to acknowledge the help I had on this work from:
My unwavering @UoE_Astro PhD advisor Sasha Hinkley & now @ucsc postdoc mentor Andy Skemer
The statistical prowess of Mariangela Bonavita at @PhysAstroEd
The modelling maestro @AstroMarkyMark
My unwavering @UoE_Astro PhD advisor Sasha Hinkley & now @ucsc postdoc mentor Andy Skemer
The statistical prowess of Mariangela Bonavita at @PhysAstroEd
The modelling maestro @AstroMarkyMark
The coronagraphy masters @djulik @marshallperrin and @LaurentPueyo from @stsci
VLT exoplanet survey expert @ArthurVigan
and last but not least, the young moving group cartographer @jgagneastro
Now let's dig into it!
VLT exoplanet survey expert @ArthurVigan
and last but not least, the young moving group cartographer @jgagneastro
Now let's dig into it!
We know JWST will be great at directly imaging exoplanets as it:
1) Is super-sensitive due to its whopping 6.5m primary mirror
2) Has coronagraphic modes to block out stars and observe faint companion objects
3) Can observe in the infrared, where planets are brighter
1) Is super-sensitive due to its whopping 6.5m primary mirror
2) Has coronagraphic modes to block out stars and observe faint companion objects
3) Can observe in the infrared, where planets are brighter
But the question this paper aims to answer is: How great is it going to be?
Although inspired by the previous similar work of Beichman et al 2010 ( https://arxiv.org/abs/1001.0351 ), this time we were armed with the latest JWST simulation tools, and planetary evolution/atmosphere models.
Although inspired by the previous similar work of Beichman et al 2010 ( https://arxiv.org/abs/1001.0351 ), this time we were armed with the latest JWST simulation tools, and planetary evolution/atmosphere models.
To tackle this we looked at JWST performance when observing members of the young moving groups Beta Pictoris/TW Hya
These objects are awesome as they are all within ~80pc so the angular scales we explore correspond to small physical separations where planets are more common
These objects are awesome as they are all within ~80pc so the angular scales we explore correspond to small physical separations where planets are more common
and their ages are ~10Myr and ~24Myr, young enough that planetary formation has largely ended, but old enough that any planets haven't significantly cooled and are therefore more luminous and easier to detect.
As lower mass exoplanets of a given age are fainter - this is great!
As lower mass exoplanets of a given age are fainter - this is great!
So, we assembled almost 100 of these objects and performed a suite of simulations of what JWST observations would look like. LOADS more detail in the paper, but all you need to know is that we tested four different filters at ~3.6, ~4.4, ~11.4, and ~15.5 microns.
Images like those in the flowchart above are cool, but what we want to get an idea of is what our sensitivity is as a function of radial separation for each target.
Let me introduce: "The Contrast Curve"
Objects above the solid line should be detectable in our images.
Let me introduce: "The Contrast Curve"
Objects above the solid line should be detectable in our images.
Using the target stars magnitude we can convert these relative magnitude contrasts into absolute magnitude sensitivity limits, and because we picked young moving group members with known ages, we can also use evolutionary models to turn these into *mass* sensitivity limits.
Almost there! To account for the range of possible orbital inclinations, eccentricities and locations of potential companions, we utilised a population synthesis model to produce "detection probability maps" and convert our angular separations to physical ones.
The contours of these plots tell us what fraction of the time we could detect a companion of a given mass and separation. But what is truly powerful is that we can *average* these maps over subsets of our sample to assess JWST sensitivity across a broad sample of objects
When we separate our sample by young moving group, we can see that Beta Pictoris members are more sensitive to closer separations companions (as they are closer to us), whilst TW Hya members are more sensitive to lower mass planets (as they are younger).
We also see that irrespective of moving group, the longest wavelength filters are best for detecting the lowest mass objects.
This is because at a given age, lower mass exoplanets are cooler and the peak of their spectral energy distributions are further into the mid-infrared.
This is because at a given age, lower mass exoplanets are cooler and the peak of their spectral energy distributions are further into the mid-infrared.
When we separate things instead by spectral type, it's clear that M stars are the best targets irrespective of the filter we're observing in. As these objects are at similar distances, this is because the earlier type stars are brighter and impart more noise into the images.
Finally, if we compare our entire sample to an equally sized subsample from a recent real survey using the state-of-the-art ground based VLT SPHERE instrument, we can determine what specific advantage JWST will provide that we don't already have.
In the bottom panels, which show the absolute differenced detection probabilities of the surveys, it's clear that ground-based instruments will retain their advantage at the closest separations, but further out JWST offers a unique opportunity towards imaging sub-Jupiters.
Looking to the known planetary population, this region of parameter space has been largely unexplored. Judging by the contours from the best performing 15.5 micron filter, we can see that JWST could provide important constraints on the occurrence rates at wide separations.
I think I'll leave it there, thanks to whoever read this far, this was my first attempt at a tweet thread so I hope it turned out ok! If you have remaining questions feel free to send them my way. Hopefully some of you can use this paper as ammo in your upcoming JWST proposals!
