Linn Boldt-Christmas
@nplinnspace.bsky.social
🇸🇪 Exoplanet atmospheres + chemistry + formation at Uppsala University 🪐🔭 Also into astrobiology, policy, climate, equality/EDI, and scicomm ✨ (she/her) – linnboldtchristmas.wordpress.com
Absolutely floored to have received an award for "open and inclusive" #scicomm and outreach. I'm not usually one to broadcast such things, but I think highlighting this type of work is more important than ever. Please consider getting involved in your own communities! 🔭🧪
www.uu.se/institution/...
www.uu.se/institution/...
October 31, 2025 at 11:18 AM
Absolutely floored to have received an award for "open and inclusive" #scicomm and outreach. I'm not usually one to broadcast such things, but I think highlighting this type of work is more important than ever. Please consider getting involved in your own communities! 🔭🧪
www.uu.se/institution/...
www.uu.se/institution/...
RIP to the incredible Jane Goodall. I learned about her (a woman scientist! They exist?!) as a precocious child attempting to learn English by reading The Far Side. She wrote the foreword to one collection after she was featured, to her delight, in a comic. The story: screenrant.com/far-side-con...
October 1, 2025 at 6:46 PM
RIP to the incredible Jane Goodall. I learned about her (a woman scientist! They exist?!) as a precocious child attempting to learn English by reading The Far Side. She wrote the foreword to one collection after she was featured, to her delight, in a comic. The story: screenrant.com/far-side-con...
We *absolutely* should celebrate it. Do your part by enjoying the 200+ page masterpiece in its entirety on ADS! ui.adsabs.harvard.edu/abs/1925PhDT... 🔭
Special shoutout to Figure 2, an incredible piece of art that I want on my wall (and suspect could even make an excellent tattoo if executed well?)
Special shoutout to Figure 2, an incredible piece of art that I want on my wall (and suspect could even make an excellent tattoo if executed well?)
September 23, 2025 at 8:03 PM
We *absolutely* should celebrate it. Do your part by enjoying the 200+ page masterpiece in its entirety on ADS! ui.adsabs.harvard.edu/abs/1925PhDT... 🔭
Special shoutout to Figure 2, an incredible piece of art that I want on my wall (and suspect could even make an excellent tattoo if executed well?)
Special shoutout to Figure 2, an incredible piece of art that I want on my wall (and suspect could even make an excellent tattoo if executed well?)
RIP to this real one (2022–2025) who I suppose is entering forced retirement from my outreach talks after an illustrious 3-yr career 😔 🕊️ truly my MVP and I guess one I will have to replace by either making my own or hoping one gets published... I wonder how much the statistics have changed by!
September 18, 2025 at 11:56 AM
RIP to this real one (2022–2025) who I suppose is entering forced retirement from my outreach talks after an illustrious 3-yr career 😔 🕊️ truly my MVP and I guess one I will have to replace by either making my own or hoping one gets published... I wonder how much the statistics have changed by!
*So* excited about our pre-print on TSD: Transmission Spectroscopy Decomposition, a novel algorithm that basically delivers your nIR high res #exoplanet 's transmission spectrum without PCA/SYSREM. There might be a 🧵 later, so for now, I'll let this speak for itself... ⬇️
arxiv.org/abs/2509.12737 🧪🔭
arxiv.org/abs/2509.12737 🧪🔭
September 17, 2025 at 8:58 AM
*So* excited about our pre-print on TSD: Transmission Spectroscopy Decomposition, a novel algorithm that basically delivers your nIR high res #exoplanet 's transmission spectrum without PCA/SYSREM. There might be a 🧵 later, so for now, I'll let this speak for itself... ⬇️
arxiv.org/abs/2509.12737 🧪🔭
arxiv.org/abs/2509.12737 🧪🔭
Update: thesis is printed! 🥹✨
Now just the defence left…!
Now just the defence left…!
September 1, 2025 at 1:59 PM
Update: thesis is printed! 🥹✨
Now just the defence left…!
Now just the defence left…!
I'll leave up my conclusions slide (because yes, these are slides from a talk I gave at #ExoclimesVII) but as a parting note:
We were SO CLOSE to giving up on this data set. So close. And only because we happened to have some amazing sims (huge shout-out to 2nd author @spectraltypos.bsky.social)...
We were SO CLOSE to giving up on this data set. So close. And only because we happened to have some amazing sims (huge shout-out to 2nd author @spectraltypos.bsky.social)...
August 27, 2025 at 10:38 AM
I'll leave up my conclusions slide (because yes, these are slides from a talk I gave at #ExoclimesVII) but as a parting note:
We were SO CLOSE to giving up on this data set. So close. And only because we happened to have some amazing sims (huge shout-out to 2nd author @spectraltypos.bsky.social)...
We were SO CLOSE to giving up on this data set. So close. And only because we happened to have some amazing sims (huge shout-out to 2nd author @spectraltypos.bsky.social)...
If you're looking at these plots and you're thinking... wait a minute, are they not off-centre...? ➡️ Yes! 😅
This is an interesting effect that we discuss in detail in the paper, but in a nutshell: not an off-set per se, and with more sims, we mostly blame tellurics (🌧️worst for H2O🌧️) and/or SYSREM
This is an interesting effect that we discuss in detail in the paper, but in a nutshell: not an off-set per se, and with more sims, we mostly blame tellurics (🌧️worst for H2O🌧️) and/or SYSREM
August 27, 2025 at 10:38 AM
If you're looking at these plots and you're thinking... wait a minute, are they not off-centre...? ➡️ Yes! 😅
This is an interesting effect that we discuss in detail in the paper, but in a nutshell: not an off-set per se, and with more sims, we mostly blame tellurics (🌧️worst for H2O🌧️) and/or SYSREM
This is an interesting effect that we discuss in detail in the paper, but in a nutshell: not an off-set per se, and with more sims, we mostly blame tellurics (🌧️worst for H2O🌧️) and/or SYSREM
So, drumroll! 🥁🥁🥁 Which template out of our four fared best? Are we sensitive to VMRs? Are we sensitive to clouds?
TLDR: definitely to clouds, less so for VMR – surprisingly so! Can this perhaps be a way for us to break the cloud-metallicity degeneracy by qualitatively combining space + ground obs?
TLDR: definitely to clouds, less so for VMR – surprisingly so! Can this perhaps be a way for us to break the cloud-metallicity degeneracy by qualitatively combining space + ground obs?
August 27, 2025 at 10:38 AM
So, drumroll! 🥁🥁🥁 Which template out of our four fared best? Are we sensitive to VMRs? Are we sensitive to clouds?
TLDR: definitely to clouds, less so for VMR – surprisingly so! Can this perhaps be a way for us to break the cloud-metallicity degeneracy by qualitatively combining space + ground obs?
TLDR: definitely to clouds, less so for VMR – surprisingly so! Can this perhaps be a way for us to break the cloud-metallicity degeneracy by qualitatively combining space + ground obs?
And trying again... we got the detections 🥹✨
We detect CO at S/N~6 and H2O at S/N~4.5, and using a "global" template (include all species from the RFs of W+24), we also get S/N~6. We also tried detecting NH3 but, as our sims predict, this is a non-detection for us (in our chosen wavelength band)
We detect CO at S/N~6 and H2O at S/N~4.5, and using a "global" template (include all species from the RFs of W+24), we also get S/N~6. We also tried detecting NH3 but, as our sims predict, this is a non-detection for us (in our chosen wavelength band)
August 27, 2025 at 10:38 AM
And trying again... we got the detections 🥹✨
We detect CO at S/N~6 and H2O at S/N~4.5, and using a "global" template (include all species from the RFs of W+24), we also get S/N~6. We also tried detecting NH3 but, as our sims predict, this is a non-detection for us (in our chosen wavelength band)
We detect CO at S/N~6 and H2O at S/N~4.5, and using a "global" template (include all species from the RFs of W+24), we also get S/N~6. We also tried detecting NH3 but, as our sims predict, this is a non-detection for us (in our chosen wavelength band)
Using the results of Welbanks+2024, who obtained two sets of results from two different retrieval frameworks (RF), we made templates using the volume mixing ratios from RF1 and RF2, both with and w/o clouds, for a total of four templates. This way, we could test what we're more sensitive to in CC.
August 27, 2025 at 10:38 AM
Using the results of Welbanks+2024, who obtained two sets of results from two different retrieval frameworks (RF), we made templates using the volume mixing ratios from RF1 and RF2, both with and w/o clouds, for a total of four templates. This way, we could test what we're more sensitive to in CC.
Two things became clear:
1. Using a ”mismatched” template will be notably punished. We knew this, but the severity was surprising (to us).
2. The detection significance for a cloudy won’t be like 😇a bit lower😇 — no, it will be basically GONE by a factor of ~4 (!!) and go from a clear yes to a no!
1. Using a ”mismatched” template will be notably punished. We knew this, but the severity was surprising (to us).
2. The detection significance for a cloudy won’t be like 😇a bit lower😇 — no, it will be basically GONE by a factor of ~4 (!!) and go from a clear yes to a no!
August 27, 2025 at 10:38 AM
Two things became clear:
1. Using a ”mismatched” template will be notably punished. We knew this, but the severity was surprising (to us).
2. The detection significance for a cloudy won’t be like 😇a bit lower😇 — no, it will be basically GONE by a factor of ~4 (!!) and go from a clear yes to a no!
1. Using a ”mismatched” template will be notably punished. We knew this, but the severity was surprising (to us).
2. The detection significance for a cloudy won’t be like 😇a bit lower😇 — no, it will be basically GONE by a factor of ~4 (!!) and go from a clear yes to a no!
But ground/high R is a smaller window of greater spectral detail… so when we thought ”pft, it’s clouds, how bad can they REALLY be?” and whipped up a quick sim… we realised, oh no, OK, we’re actually losing a LOT comparatively here.
⬇️ With and without clouds, where white = data and grey = no data!
⬇️ With and without clouds, where white = data and grey = no data!
August 27, 2025 at 10:38 AM
But ground/high R is a smaller window of greater spectral detail… so when we thought ”pft, it’s clouds, how bad can they REALLY be?” and whipped up a quick sim… we realised, oh no, OK, we’re actually losing a LOT comparatively here.
⬇️ With and without clouds, where white = data and grey = no data!
⬇️ With and without clouds, where white = data and grey = no data!
As you effectively lose contrast (shorter spectral lines), how this looks in practice is that you basically ends up with only the ”hats” of the spectrum instead. I love this fig from Gandhi+2020 (it's in my thesis!) demonstrating how high altitude cloud deck -> quenched spectral lines -> hats only 🎩
![A screenshot from my thesis!
Figure caption: A plot from Gandhi et al. (2020) [https://academic.oup.com/mnras/article/498/1/194/5899070] simulating the impact of cloud decks
at different pressures (altitudes) on transmission spectra of the 1.4 µm water feature
for exoplanet GJ 436 b, across a wider wavelength range (left) versus a shorter range
(right). The transmission spectra is shown at high spectral resolution, and the diamond
markers indicate the equivalent HST WFC3 spectra. The height of the H2O spectral
feature is given in parts per million (ppm).](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:npllrmx4ntyhfyl4smoq55vf/bafkreieep7xfpipz4odmeqqylwtpl5w4oxenua3r6ulnj4izxntxmlxzsq@jpeg)
August 27, 2025 at 10:38 AM
As you effectively lose contrast (shorter spectral lines), how this looks in practice is that you basically ends up with only the ”hats” of the spectrum instead. I love this fig from Gandhi+2020 (it's in my thesis!) demonstrating how high altitude cloud deck -> quenched spectral lines -> hats only 🎩
And what happens to a transmission spectrum when you have clouds in your #exoplanet? In a clear atmosphere, the light filters through and gets to us carrying nice strong spectral lines... but with clouds, the light is more likely to get scattered, and quenched, and we lose A LOT of information 🧪 📹 ⬇️
August 27, 2025 at 10:38 AM
And what happens to a transmission spectrum when you have clouds in your #exoplanet? In a clear atmosphere, the light filters through and gets to us carrying nice strong spectral lines... but with clouds, the light is more likely to get scattered, and quenched, and we lose A LOT of information 🧪 📹 ⬇️
But when we first tried to analyse our data……… we got THIS:
And we were like 😭 what HAPPENED?! This was supposed to be a quick paper!!!! I’m doing a PhD dang it, I’m on a timeline!!!!
…….and that’s when this paper instead became a post mortem of why this happened.
Strap in and let’s do some CSI🕵️♀️
And we were like 😭 what HAPPENED?! This was supposed to be a quick paper!!!! I’m doing a PhD dang it, I’m on a timeline!!!!
…….and that’s when this paper instead became a post mortem of why this happened.
Strap in and let’s do some CSI🕵️♀️
August 27, 2025 at 10:38 AM
But when we first tried to analyse our data……… we got THIS:
And we were like 😭 what HAPPENED?! This was supposed to be a quick paper!!!! I’m doing a PhD dang it, I’m on a timeline!!!!
…….and that’s when this paper instead became a post mortem of why this happened.
Strap in and let’s do some CSI🕵️♀️
And we were like 😭 what HAPPENED?! This was supposed to be a quick paper!!!! I’m doing a PhD dang it, I’m on a timeline!!!!
…….and that’s when this paper instead became a post mortem of why this happened.
Strap in and let’s do some CSI🕵️♀️
…folks, I’m going to be honest. This was not a cakewalk.
I’m actually going to be v vulnerable here and I’m even going to show you a behind the scenes blooper. For those of you not familiar with the field, all you have to know is that a so-called detection map is *supposed* to look a bit like this:
I’m actually going to be v vulnerable here and I’m even going to show you a behind the scenes blooper. For those of you not familiar with the field, all you have to know is that a so-called detection map is *supposed* to look a bit like this:
August 27, 2025 at 10:38 AM
…folks, I’m going to be honest. This was not a cakewalk.
I’m actually going to be v vulnerable here and I’m even going to show you a behind the scenes blooper. For those of you not familiar with the field, all you have to know is that a so-called detection map is *supposed* to look a bit like this:
I’m actually going to be v vulnerable here and I’m even going to show you a behind the scenes blooper. For those of you not familiar with the field, all you have to know is that a so-called detection map is *supposed* to look a bit like this:
And this is not necessarily a bad thing! Just that one is sometimes better suited than the other 💪 It also means we can follow the ethos of great philosopher Hannah Montana and use *both* JWST *and* VLT instruments with overlapping wavelengths to say even *more* about a target. Like in this study!
August 27, 2025 at 10:38 AM
And this is not necessarily a bad thing! Just that one is sometimes better suited than the other 💪 It also means we can follow the ethos of great philosopher Hannah Montana and use *both* JWST *and* VLT instruments with overlapping wavelengths to say even *more* about a target. Like in this study!
This means it's super "puffy" and why you've maybe heard of it as the "popcorn planet" 🍿
But that's not all: it's also got an escaping tail, asymmetry btwn morning/evening limb, a funky orbit, lots of clouds ☁️☁️☁️ and many more traits that interest us.
So it's a weirdo... but a well-studied weirdo.
But that's not all: it's also got an escaping tail, asymmetry btwn morning/evening limb, a funky orbit, lots of clouds ☁️☁️☁️ and many more traits that interest us.
So it's a weirdo... but a well-studied weirdo.
August 27, 2025 at 10:38 AM
This means it's super "puffy" and why you've maybe heard of it as the "popcorn planet" 🍿
But that's not all: it's also got an escaping tail, asymmetry btwn morning/evening limb, a funky orbit, lots of clouds ☁️☁️☁️ and many more traits that interest us.
So it's a weirdo... but a well-studied weirdo.
But that's not all: it's also got an escaping tail, asymmetry btwn morning/evening limb, a funky orbit, lots of clouds ☁️☁️☁️ and many more traits that interest us.
So it's a weirdo... but a well-studied weirdo.
According to www.color-blindness.com/coblis-color... (which is a fantastic tool that I highly recommend everyone uses before submitting a paper or giving a presentation), it actually works surprisingly well – more than many others I've seen! I'm impressed.
June 29, 2025 at 6:15 AM
According to www.color-blindness.com/coblis-color... (which is a fantastic tool that I highly recommend everyone uses before submitting a paper or giving a presentation), it actually works surprisingly well – more than many others I've seen! I'm impressed.
🔭🧪 A comparison of a smaller region by DESI vs Rubin — shoddily lined up by eye, taken as screenshots from the compressed photos in the post below, and STILL the difference is THIS obvious
🤯 @vrubinobs.bsky.social is truly going to be revolutionary!
🤯 @vrubinobs.bsky.social is truly going to be revolutionary!
June 24, 2025 at 3:58 AM
🔭🧪 A comparison of a smaller region by DESI vs Rubin — shoddily lined up by eye, taken as screenshots from the compressed photos in the post below, and STILL the difference is THIS obvious
🤯 @vrubinobs.bsky.social is truly going to be revolutionary!
🤯 @vrubinobs.bsky.social is truly going to be revolutionary!
🔭🪐 Happy to report that I passed my ”pre-defence seminar” where my department assesses whether I am in fact ready for my doctoral defence or not… guess I have to actually submit my thesis now because I’m now officially defending it in September! 😅🥲😭
May 23, 2025 at 2:04 PM
🔭🪐 Happy to report that I passed my ”pre-defence seminar” where my department assesses whether I am in fact ready for my doctoral defence or not… guess I have to actually submit my thesis now because I’m now officially defending it in September! 😅🥲😭
My take, based on 10+ years of watching everyone from teenagers to Emeritus professors lugging around the same kind of faded Jansport backpack at every university campus I’ve ever been to
May 11, 2025 at 1:07 PM
My take, based on 10+ years of watching everyone from teenagers to Emeritus professors lugging around the same kind of faded Jansport backpack at every university campus I’ve ever been to
Me: ”wait, that’s a great idea — why haven’t I thought to do this??”
Me: ”…oh yeah…”
Me: ”…oh yeah…”
April 28, 2025 at 9:16 PM
Me: ”wait, that’s a great idea — why haven’t I thought to do this??”
Me: ”…oh yeah…”
Me: ”…oh yeah…”