Premium subscribers glow so fiercely they outshine the stars

It’s amazing to me how much you can tell about some cosmic objects just by a single glance at an image of them. If you have a little experience, you can squeeze a lot of info out of that momentary glimpse. 

For example, may I present to you NGC 7469?

NGC 7469. Credit: ESA/Webb, NASA & CSA, L. Armus, A. S. Evans

If you have any knowledge of astronomy at all, you’ll likely recognize that as a spiral galaxy. Correct! It’s about 220 million light-years away from us, which I determined by first looking it up in an astronomical database, getting its redshift, and plugging it into Ned Wright’s Cosmology Calculator. The “light travel time”, literally how long it took the light to reach Earth from this galaxy, is listed as 0.222 Gyr (gigayears), or 222 million light-years.

OK, what else can we glean? [It might help to look at a much higher-res version, which is spectacular.]

The second most obvious thing is the intense red center to the galaxy with the six spikes radiating away from it. There are a couple of interesting things we can deduce from this, one trivial, one deep.

The deep one is this: this is an active galaxy, with a monster black hole in its core gobbling down matter and spitting out vast amounts of energy.

So how do I know this? Most spiral galaxy centers are extended fuzzy bulges, sometimes spherical (so they look circular to us) and sometimes stretched out in what’s called a bar, a lozenge or Tic Tac shaped collection of stars. These are usually obvious structures but not overly bright compared to the rest of the galaxy.

However, some galaxies have a lot of light pouring out from their cores, because (as I said above) they have hungry beasts there. As material falls into a supermassive black hole at the centers of such a galaxy it piles up in a disk, gets very hot — like, millions of degrees hot! — and blasts out so much light it can drown out all the others stars in the galaxy itself. This disk is light years across, but from a vast distance it appears as a dot in a telescope.

That explains the spikes. Those are not real, but are caused by light entering a telescope and getting diffracted around mechanisms inside. When light hits a barrier of some kind it can bend a bit around the edges, creating interference patterns (I explain this in detail back on the old SYFY blog). For telescopes like Hubble, these diffraction spikes are created when the light passes by the thin metal strips holding the secondary mirror in place, making four distinctive crosshair-like spikes.

JWST, on the other hand, has a mirror made up of 18 separate smaller hexagonal mirrors that all fit together to form a roughly 6-sided mirror. That creates six diffraction spikes, plus there’s a series of three struts that hold the secondary mirror in place. Some of these spikes overlap, creating eight spikes, though usually the horizontal one is faint and can be difficult to spot. 

That pattern is instantly recognizable, so I knew right away (the trivial deduction I mentioned earlier) that this image was taken by JWST.

Diffraction spikes are only easily seen when a source is very small in the sky, what we call a point source. An extended source, like a typical galaxy center, smears out the spikes, making them much fainter, even invisible. So that’s how I knew this galaxy was active. The point-source center makes that clear.

But that leads to another thing you can understand: JWST sees infrared light, so the light coming form the center must be powerful in that wavelength range. What emits that much infrared light?

Dust. Tiny grains of rocky or sooty material created in dying massive stars. Some galaxy centers are loaded with dust surrounding their supermassive black holes. That usually takes the shape of a giant torus (a doughnut shape) around the black hole, which sometimes hides a lot of the visible and higher energy light, but the warm dust itself glows in the infrared. There must be a lot of dust in NGC 7469!

That in turn implies that there could be a lot of star formation in the galaxy near the center — you need that to make lots of relatively rare massive stars. If you look closely, you can see a ring of light around the core: that’s a starburst ring, literally a ring-shaped region where stars are being born at a furious rate. Many of them are massive, creating lots of dust, which can fall to the center of the galaxy (though as we’ll see in a sec there are other ways to feed lots of dust into a galaxy’s core). The ring is about 1,500 light-years in radius.

The core of NGC 7469 (zoomed and contrast enhanced) with the ring arrowed. Credit: ESA/Webb, NASA & CSA, L. Armus, A. S. Evans

Even if I hadn’t known this was a JWST image right away, I could have figured it was an infrared telescope. Spiral arms are where stars are born, and in visible light (the kind we see) they tend to be clumpy where great nebulae are forming stars. These clumps are reddish due to hydrogen, and the arms themselves blue from the massive blue stars dominating the light. The arms of NGC 7469 here, though, while bluish, are also very smooth, almost smoky looking. That’s because we’re seeing the light from redder stars in visible light (displayed in the image as blue), which can move around a bit between the arms, making the arms look less distinct. You can in fact see some of these star factories to the lower left, bright in the wavelengths emitted by dust.

There’s more! The galaxy has some odd features, like the way those reddish clumps are only on one side. Another is a fuzzy loop of yellowish material around the center. That says to me that another galaxy likely interacted with NGC 7469 not too long in the past, its gravity distorting the shape of the spiral, and possibly causing gas clouds to collide and spark all that star formation to the lower left. Active galaxies commonly have suffered collisions, because the aftermath tends to dump a lot of material into the core where it can be fed to the black hole.

Sure enough, to the lower left in the image is another galaxy, IC 5283, and it is in fact interacting with it! This could explain a lot of the features seen. 

Not bad, eh? It’s incredible what you can tease out of an image just by looking at it. Of course, it does take a bit of experience, knowledge, and an eye for what to spot. I’ve stared at approximately a bazillion images like this, so I do have some practice. Not that I’m perfect at it! I can still make mistakes, or make incorrect assumptions, or ascribe some cause to a feature that may be more complicated, or due to something else. That happens. But then that’s why astronomers usually do more than glance at an image. It can take months or more to properly analyze observations, including using other observations from different telescopes to peel back the layers of what’s going on in the object and get a handle on how — and why — it behaves the way it does.

It’s a process. But it’s a fun one, and the end goal is one of the most wondrous there is: understanding. That’s the whole point.

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

In yesterday’s issue (BAN #1094) I wrote about the weird and surprising properties of a geometric solid called Gabriel’s Horn. Specifically, I wrote about how it has infinite surface area while only a finite volume. 

The basic stuff I said is correct, but I got the math wrong: to get pi cubic units to fill the horn, you need to start the solid at x = 1! If you let x approach 0 (so the bell spreads out), the volume does indeed go to infinity, so I was wrong there too. Still, if you start at x = 1 and let it go to infinity, the volume converges while the surface area grows without bound (for the math nerds: it increases as ln(x), which always increases, though slowly), which still means the volume is finite while the surface area is infinite.

My apologies for getting this wrong, and my thanks to BAN reader Nick Fontana for pointing this out to me.

Hey, look, if you know me, you know that I will use the thinnest of excuses to post an image of a globular cluster, because I love them so.

Yeah, well, here you go:

NGC 6723, a globular cluster of the Milky Way. Credit: ESA/Hubble & NASA, A. Sarajedini, G. Piotto

That is NGC 6723, a fantabulous cluster of hundreds of thousands of stars located near the center of our galaxy but just on the other side of it from us, about 27,000 light-years away. Many such globulars are in this part of the sky because they orbit our galaxy at great distances, and when you look toward the center you’re bound to see more of them. About 160 such globulars are known to orbit the galaxy, though there are likely many more that remain undiscovered. 

NGC 6723 is in many ways typical of the breed. It was once thought that the stars in these clusters were born all at once, which made them fantastic laboratories for studying stellar populations. Since they’re all at essentially the same distance from us we can directly compare stars’ brightnesses to see how much energy they give off, and if they’re all the same age that means the models we use to understand them are simpler.

But the universe is rarely so giving. It took a long time, but it slowly became clear that the stars in globulars are not all the same age. Star formation in many globulars happened in waves, with many stars born early on, and then a pile of them formed later, sometimes by hundreds of millions or billions of years. That’s the case with NGC 6723. How this happens, exactly, isn’t well understood.

Confusing things further, some globulars aren’t globulars at all! They’re actually the cores of fully fledged but tiny galaxies that were stripped of most of their stars and gas by multiple close encounters with our much larger Milky Way. The remaining stars form a tight, compact ball that looks just like a globular, but formed and evolved in a very different way. So what was once thought to be a simple structure is in fact a mess.

I don’t know if everyone has a favorite math thing. Given the kind of person who subscribes to my newsletter there’s a greater-than-average chance you do. I certainly have one, and because it’s my newsletter I get to tell you about it!

When I was in high school I took an intro to calculus class. One thing we learned was how to use it to determine the surface area and volume of different kinds of geometric shapes. I won’t go into the details, but suffice to say you can, from just the equation of a shape, determine those properties. For example, if you have a sphere, you can figure out, from first principles, what the equations are for its area and volume (the familiar 4 pi r² and 4/3 pi r³ ).

Another thing we learned about is what’s called a surface of revolution. It’s where you take some function and revolve it around the one of the axes in a coordinate system. Wikipedia has a somewhat high-level description of how this works, as does Wolfram (with some examples shown).

As an example, with malice aforethought, take the simple function y = 1 / x. When you graph it you get a fun little curve:

The graph of 1/x. Credit: Desmos

This is decently straightforward. At x = 1, y also equals 1. But as x gets bigger (to the right), y gets smaller, and the graphed curve gets closer and closer to the x-axis. But it never quite gets to y = 0; even at very large x, y is very small but not 0. If x = a billion, y = 1/billion, which is small but not 0.

The same is true going the other way: at small x (toward the left) y gets bigger. At x = 1/billion, y = a billion. 

Here’s the (first) fun bit. Imagine that graphed curve is a wire. Now imagine grabbing the right hand side of that graph with your right hand, and the top part with your left. Keeping your right hand still, pull the top of the curve down toward you, spinning it around the x-axis. Keep rotating it all the way around until it’s back up to the top. What do you get?

You get this:

The graph of 1/x rotated around the x-axis. Note that the whole graph has been tilted a bit to show you the third dimension, the z-axis, more clearly. The color is lightened below the x-y plane to make it easier to see depth. Credit: Desmos

It looks like a trombone bell! Or a trumpet bell, I guess, but I played trombone, and again it’s my newsletter so trombone it is. This shape is called Gabriel’s Horn, in fact, after the archangel Gabriel who blows a horn at the End of Days in the Bible (though to my knowledge it’s never specified who blows the horn).

Here’s the second fun bit: I cut off the graph at pretty small values of x and y. What if we let it go on forever?

Then we get this shape:

Gabriel’s Horn extended. Credit: Desmos 

Cooool. There are two very important things to note here. First, as x gets bigger (to the lower right), z and y get smaller, so the graph gets narrower and narrower. If we extend this out more and more, it gets closer and closer to a one-dimensional line.

Second, as x gets smaller, the graph flattens out because y and z get bigger. At very small values of x, y and z get huge, and the shape approaches being a plane. It never quite gets there, but it gets very flat.

This shape is very, very strange. You can use calculus to calculate its surface area and volume, and what you get will destroy your mind.

Look first at the narrow part. That gets close to being a line, which has no surface area and no volume, so as x gets bigger it contributes to those two properties less and less.

But at small x things get weird. The surface area grows without bound, as we say, meaning it gets infinitely big. It’s essentially an infinite plane, so its area is infinite as well. OK, fine, that’s obvious enough from looking at the graph.

But what about the volume? As X gets small, the horn itself gets closer and closer the y-z plane. At very large x, y and z are small, so there’s less and less room between the horn itself and that plane. That means there’s less and less volume the smaller x gets.

That means very small values of x also don’t contribute to the volume. Instead of the volume blowing up to infinity, it actually approaches a finite value. If you do the math, you find that volume is equal to… pi! Yup, our favorite number, 3.1415 etc. etc., in whatever units the axes are in.

Do you see the bizarreness now? Imagine this is an actual three-dimensional object, extending infinitely in all directions, and the axes are measured in, say, centimeters. It would take an infinite amount of paint to cover it, but you could fill its interior with just over three cubic centimeters!

I will give you a moment to collect the pieces of your brain that have exploded all over the room.

This is not a trick; there’s no weird gimmick. It’s math: Gabriel’s Horn has infinite area but finite volume.

This is not how we see things behave in the real world. If the area goes up the volume has to as well, usually even faster! But not in this case. Because the shape gets closer to all the axes in every direction, the volume is constrained.

But that seems paradoxical! You’d think the inside of the surface and the outside would be the same, so if you wanted to paint the inside of the surface it would take an infinite amount of paint, but if that’s so then how could it be filled with only a small amount? That paint still has to touch every part of the surface, so you need an infinite amount, yet pi cubic units fills it. 

Around and around we go. It’s definitely brain-twisty.

Now, in the real world there are other issues with this as well. If you try to paint the inside, eventually the part spread out over the y-z plane gets so narrow that a paint molecule won’t even fit, and the part along the x-axis gets so small you get the same problem. You can probably argue about quantum mechanics limiting things as well when you get small enough values, but whatever, it doesn’t matter, because we’re talking pure math here, and it doesn’t always exactly mimic the real world. It’s an idealization, where you can have infinite values and perfect lines and planes and all that.

In fact this apparent paradox caused a lot of consternation among mathematicians in the 17^th century. The nature of infinity is weird and not always easy to grasp. In this case, it really messes with our notion of area and volume, which we evolved through things we can grasp with our hands. Making the leap to pure math can really be hard when the concepts seem to contradict our more physical experience.

No wonder Gabriel (maybe) used it to announce Judgment Day. I can’t blame him.

Anyway, when we went through this math in high school my mind was well and truly blown (at least as much as the Archangel’s trombone). I was both delighted by the math and horrified by the implications, and to be honest it still freaks me out to this day. It’s a reminder to me that math describes reality, but it isn’t necessarily reality. This too has very deep and difficult philosophical implications, which I will leave to bigger brains than mine to probe, even as I enjoy the argument.

And finally, to paraphrase myself:

Math! I love this stuff.

I created the graphs at the Desmos site, where you can play with various functions and graphing. It’s pretty fun! All credit to them.

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

Premium subscribers hypersonically impact my life

The moon gets pummeled with a lot of rocks from space. Without an atmosphere like Earth’s every single bit of cosmic debris slams into the moon’s surface. Most are teeny tiny and make microcraters (also called zap pits, yes, seriously) on rocks, but bigger ones make larger impact craters on the surface. The bigger the space rock the more rare it is, so substantial crater formation isn’t terribly common.

The Lunar Reconnaissance Orbiter (or LRO) takes images of the lunar surface all the time, and has been since it entered orbit around the moon in 2009. It has found many new small craters because it sees the same spots on the moon over and over again; when a new crater appears it can be pretty obvious in a before-and-after comparison.

And sometimes they’re really obvious: sometime between April and May 2024, an asteroid probably about 20 meters across (so, about the same size as the Chelyabinsk impactor, the one that blew up over Russia in 2013) slammed into the lunar surface, carving out a crater 222 meters across and about 30-40 meters deep. That’s an area of about 10 acres, so yeah, decently big. A crater this large only forms on the moon about once per 130 years, so it was pretty cool LRO happened to be up there to see it.

The newly formed McGetchin crater. The image is 1.3 km in width. Credit: NASA/GSFC/Intuitive Machines 

Informally named McGetchin crater (after an Apollo-era geologist), it was first seen in ratio images, where one image is divided by another to look for changes. Those are taken six months apart, so once scientists saw the new crater they found lots of other images of the area, and were able to narrow the impact date down further. They also commanded LRO to make more observations to get as much info on the crater as they could.

It yielded a lot of very interesting science [link to journal paper]. For example, debris from the crater was found as far as 100 kilometers from the impact site! Rocks and dust were thrown a long way, and that means people looking into future exploration and inhabitability of the moon need to be very thoughtful about such events (as well as dust thrown around as rockets land). Small impacts won’t throw material as far, but happen more often, so both could affect how we build lunar structures.

A before-and-after comparison of the same area on the moon ,Each image is to the same scale, and about 925 meters across. Credit: NASA/GSFC/Intuitive Machines

Usually, material is ejected at low angles to the surface, around 2°, creating a low arc of ballistically flung debris. However, there’s a ridge eight km west of the impact site, and ejected material was found at its base, meaning some shrapnel from the impact was thrown at angles of at least 13°. That’s a huge difference, and could be due to the shape of the impacting asteroid and the angle at which it impacted.

Every headline satisfies an opinion. Except ours.

Remember when the news was about what happened, not how to feel about it? 1440's Daily Digest is bringing that back. Every morning, they sift through 100+ sources to deliver a concise, unbiased briefing — no pundits, no paywalls, no politics. Just the facts, all in five minutes. For free.

Read the newsletter trusted by 4.5 million fact-seekers.

Note: A reminder that Premium subscribers don’t see ads! But I’ll also say 1440 is a pretty good news aggregator and summarizer; I enjoy reading it — and I’ll add that I never advertise anything I myself don’t actually recommend.

I’ll add that high-velocity impact physics is extremely complicated and very difficult to test experimentally (accelerating something to those speeds is really hard), so the details of how the crater formed aren’t clear. The crater itself will be carefully studied to help us understand how these structures form.

The energies involved are huge, a large fraction of a megaton (a million tons of TNT). It takes a lot of energy to carve out that much rock. Boulders over 10 meters across were thrown over 200 meters away! Yes, gravity is lower on the moon, but a rock that big is still pretty dang hefty.

You can find more details on the LRO blog and in the science paper linked above.

The crater as seen by LRO from an oblique angle from very roughly 100 km away. Credit: NASA/GSFC/Intuitive Machines 

A final thought: the crater is located near the moons’ equator, about ¾ of the way from the center to the eastern edge. That doesn’t mean it’s easily visible from Earth: Even Hubble can’t resolve craters less than about 90 meters across, so this crater would only be a few pixels wide in such an image. Unless you have an 8-meter ‘scope lying around with advanced optics this crater is too small to be seen in detail from Earth. Now, it might appear as a dimensionless dot somewhat brighter than the surrounding area, but that’s about it. I’d be curious if any amateur telescopes could spot that, actually. It might be possible.

But that puts the moon into perspective: it’s covered in craters, literally; the ground is saturated with them. The vast majority, billions if not trillions of them, are too small to resolve from Earth! We can only see the biggest ones, and even then there are too many to individually name. 

I like Earth’s atmosphere; it lets us breathe and can be a pretty color during the day. But this really brings home the fact that our atmosphere also protects us exceedingly well from invading space rocks. Or, at least, the smaller ones, of which there are many.

The images here are from the LRO images blog, which is a terrific resource and well worth putting in your RSS feed reader.

Via a follower on Bluesky I found out that the folks at Complexly (Hank and John Green’s outfit that make Crash Course videos) have edited together all 46 episodes of Crash Course Astronomy that I wrote and hosted into one giant marathon video!

So if you have eight hours 21 minutes 25 seconds with nothing to do then why not learn about the universe?

And you will almost never hear anyone say this, but… yes, do read the comments! People love astronomy, and it’s really, really uplifting to see so many folks being so positive about science. My thanks to the Complexly team who helped create this series, and for getting tens of millions of people (!!) to think about the stars. 

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

The Nancy Grace Roman Space Telescope launched on August 30, 2026, and is well on its way to its new roost at the Earth-Sun L2 point, a region of relative gravitational stability where space missions can follow Earth along its orbit around the sun.

It takes about a month or so to get there, but it hasn’t been idle along the way. Engineers have already been prepping its two main instruments for operations. The workhorse camera, called the Wide Field Instrument, or WFI, has been cooling down for a while now, which helps it see more clearly in visible light as well as in the infrared.

Even though Roman is still moving toward L2, it was able to perform a very important task: first light. This is the first time a camera is pointed at a target and photons reach its detectors. I wouldn’t go so far to say it’s a ritual for every new space telescope, but it’s close. It’s a critical moment for any instrument, and the scientists and engineers wait very nervously and excitedly for that very first image.

And NASA just released it! Here it is, Roman’s first astronomical image:

Roman’s Wide Field Instrument first light image. Credit: NASA’s Goddard Space Flight Center, Tyler Desjardins (STScI)

It’s not what you expected, is it? It looks pretty good to me, though.

WFI has 18 detectors, like the camera chip in your phone but, um, a lot better. They’re arranged in that arc-pattern, and total about 300 megapixels. They cover an immense area of the sky for a space telescope, about 100 times bigger than Hubble’s field of view! 

I don’t have access to the actual data so I’m not sure what part of the sky that is, the exposure time, or anything like that. Still, you can see the stars are pretty badly out of focus. But that’s fine! Part of the first light image is to see how the instrument is working, and no one expects it to be focused. That happens later; there is a focusing mechanism inside the camera that will move mirrors around to sharpen the image. It’s like when you use a magnifying glass to see something tiny; you move it back and forth until the image is crisp.

The stars aren’t just fuzzy, they have that peculiar pattern to them. That’s expected, too! It’s called the PSF, for Point Spread Function. That name means any point source, like a star, gets its light spread out by the telescope optics, and there’s a specific pattern to it. You want that PSF to be as small and tight as possible, but when it’s out of focus you can see all that structure that depends on the telescope and camera optics, including the frame that holds one of the main telescope’s mirrors in place. The PSF will be vastly improved as soon as they fiddle with the focus (remember, JWST’s PSFs were a mess before calibration as well).

The pattern can get really strange. I remember in 2003 when Spitzer Space Telescope first light image was taken, and I was shocked to see the stars were triangles! You can see that image here (go to page 13). I called a friend who was a muckety-muck with Spitzer, and she told me that was fine, they expected it. They focused it and that mission went on to be ridiculously successful.

Premium subscribers keep me in resonance

HR 8799 is a very young star only 133 light-years from us. It’s likely about 40 million years old, though its age isn’t well restrained. In 2008, astronomers announced they had found three giant planets orbiting it — the planets were seen directly in imaging, appearing as little dots near the star. In 2010, a fourth planet was found as well.

These observations were made looking at infrared light, which makes it easier. For one thing, the star itself is fainter in those wavelengths (making fainter nearby planets easier to spot), but also the planets are still hot from their formation, so they glow in infrared, making them even brighter.

Over time, more images of the system were taken, leading to one of the coolest astronomical videos ever made: you can physically see the orbital motion of all four planets!

Mind you, the innermost planet moves the most rapidly, and it still takes 40 years to orbit the star once! Seeing this motion is amazing.

All four planets are significantly more massive than Jupiter, and our understanding of how planets form has trouble explaining this system. More observations always help.

But could there be a fifth planet? JWST had been previously used to observe HR 8799^* , but no fifth planet was seen (unfortunately, its capability to see a faint planet close to its star was hampered by internal optics more than had been expected before launch). However, a different team of astronomers took those publically archived observations and reprocessed them using a new method. This used models of the optical system in JWST to correct for issues that made interference from the bright host star less of a problem, allowing them to see fainter nearby objects [link to journal paper].

And it looks like they found evidence for a fifth planet! Nominally called planet f (the first four are b, c, d, and e) it would be the innermost of the system so far detected. Based on the positions of the other planets over time, and the measured position for the candidate fifth planet, the astronomers find its orbit is likely mildly elliptical about a billion kilometers out from the star, and has a mass of roughly 1.6 times Jupiter.

The planet is only about 0.0002 times as bright as the star, so it’s remarkable that they could see it at all, with it being so close in — it’s only about 0.15 arcseconds from the star as seen from Earth, which is the same as the apparent diameter of a dime seen from 14 kilometers away! So yeah, tiny.

This graph shows the positions (squares) of the four outer planets over time, and mathematical fits to their orbits (in different colors for different planets); the fits are good near the planet but spread out a bit on the other sides of their orbits. The inner planet is in red, which is shown bigger in the inset. The octagons are where the planets were when the observations were made, and the X marks where the planets were predicted to be using another mathematical model. The y-axis is direction north/south, and the x-axis is east/west, and the tic marks are in milliarcseconds, thousandths of an arcsecond. Credit: Nguyen et al. 2026 

Mind you, this isn’t confirmed yet. More observations with JWST could do it; the planet probably takes about 20 years to orbit the star, so images taken in the coming few years may show orbital motion.

If this pans out, astronomers will have fun using its existence and physical parameters to try to figure out how this system formed. Normally five giant planets like this would be unstable around a star, but their orbits appear to be in resonance with each other, with periods (their “years”) all being simple multiple factors of each other. In this case, it’s a 1:2:4:8 chain, meaning the outermost planets has a period 8 times longer than the innermost one, the second outermost takes 4 times as long, and the second innermost takes twice as long. This helps keep the system stable; if a planet starts to wander off its path a bit, the gravity of the other planets will pull it back into line. This isn’t a coincidence; over time planets can affect each other gravitationally and shift positions, and when they fall into a resonance like this they tend to stay there. We see this with three of Jupiter’s big moons, too!

I already love it when planets are directly imaged, and if this one is confirmed it’ll show JWST is back to its design spec for this type of observation. It also makes HR 8799 an even more interesting system, and it had already set a pretty high bar for that.

Tip o’ the dew cap to astronomer Sarah Kendrew on Bluesky for this news!

^* In fact, in 2025 JWST images showed direct evidence of carbon dioxide in exoplanet atmospheres, the first time this had ever been done (CO₂ had been detected before, but by using spectra). [link to journal paper]

Every headline satisfies an opinion. Except ours.

Remember when the news was about what happened, not how to feel about it? 1440's Daily Digest is bringing that back. Every morning, they sift through 100+ sources to deliver a concise, unbiased briefing — no pundits, no paywalls, no politics. Just the facts, all in five minutes. For free.

Read the newsletter trusted by 4.5 million fact-seekers.

Reminder: Premium subscribers don’t see ads! Click here to sign up. I’ll also note I only put up ads for products I actually approve of; 1440 Media has good synopses of the news and I like it.

As I wrote last week in Issue 1085, I moderated a Carnegie-sponsored panel on the search for exoplanets and alien life, using a Star Trek theme. The panelists were three Carnegie scientists (Mike Wong, Maggie Thompson, and Nicole Wallack) doing cutting-edge research in the field, and it was a really fun night.

Me, Maggie Thompson, Nicole Wallack, and Mike Wong talking about lava worlds in Trek and in reality. Credit: Katy Cain / Carnegie Earth and Planets Laboratory

As you can see, we all wore Trek shirts (and we’re all nerds). I have a science officer shirt somewhere in the house but I don’t think I ever unpacked it after we moved to Virginia, so instead I’m wearing my “woke” Trek shirt (showing the Enterprise with a rainbow coming out the back). I love that shirt.

If you want to watch, there’s a video of the night now on YouTube.

One other thing. After the panel was over and the audience was mostly leaving, Mike Wong called me over to meet someone. I figured it was a scientist he worked with, but as I stuck out my hand Mike, from behind me, says, “This is Marc Okrand. He invented the Klingon language.”

I may have had a slight brain cloud for a moment upon hearing that. Holy moly. So I said what any dork in my position would have said: “Well then, qapla’!” He smiled and laughed, at which point my brain said, “You just made the guy who invented Klingonese laugh” and that’s another thing I will always remember about that evening.

Today is a good day to smile. Credit: Phil Plait

The evening was part of the Carnegie Neighborhood Lecture Series, which has talks monthly or so in Washington, DC at the Earth and Planets Laboratory (I gave a lecture based on my book Under Alien Skies for this series in October 2024, too). If you live in or near DC, you should check them out! They’re free, and lots of fun. My thanks to my friend and lab director Mike Walter for inviting me to participate, to everyone who came, and especially everyone who made this night happen. Let’s keep boldly going.

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

Premium subscribers show there are ten sides to every argument

Saturn is generally acknowledged as the most beautiful planet in the solar system (excepting Earth), but it’s also one of the weirdest. It has gigantic rings! A fleet of bizarre moons! A very strange and borderline eerie hexagon at its north pole!

And now it only gets weirder: using images taken by Hubble Space Telescope, astronomers have spotted a polygonal structure at Saturn’s south pole. And it’s not any ordinary shape, like a square or a pentagon. It’s a decagon.

Yeah, that’s ten sides. The jagged inner region is where Hubble couldn’t see due ot the orientation of Saturn. Credit: NASA, ESA, A. Sánchez-Lavega (Basque Country University, EHU)

The edges aren’t perfectly distinct, but there are ten sides there [link to journal paper]. So, what the heck is going on?

Physics! And fluid dynamics!

These structures are due to a phenomenon called Rossby waves, and come about naturally when you have two adjacent fluids moving at different speeds on a rotating planet. As the outer one moves around the inner one, they interact and create regular polygonal shapes. The number of sides (and the length of each side) depends on a lot of factors, like the relative speed, how fast the planet is spinning, and so on. 

This seems counterintuitive, but there’s a great video showing this exact thing occurring in a lab setting. They put two fluids on a rotating table and mounted the camera above the table and rotating with it, and you can see the hexagon (six-sided figure) forming:

See? And remember, gas is a fluid; anything that flows is a fluid, and that includes liquids and gasses. Incidentally, Jupiter has similar structures, with five immense cyclones circling a central cyclone at its south pole. The underlying physics is similar. Neptune might have a hexagon at one of its poles as well.

The decagon at Saturn’s austral pole is persistent, in that it’s existed for several years at least. It was seen in Hubble images starting in 2023, and was still there in late 2025. It may be older than that, too. The Cassini spacecraft saw a partially formed polygon in 2004, but it didn’t last, so this feature didn’t exist 20 years ago.

Spotting it from Earth is hard because a) it’s at the pole which we can’t see from above, and 2) Saturn has an axial tilt similar to Earth’s, and it’s been austral winter since 2012, with the pole pointing away from the sun (and us). It was only in 2023 that the pole swung back into view and Hubble could see it.

Every headline satisfies an opinion. Except ours.

Remember when the news was about what happened, not how to feel about it? 1440's Daily Digest is bringing that back. Every morning, they sift through 100+ sources to deliver a concise, unbiased briefing — no pundits, no paywalls, no politics. Just the facts, all in five minutes. For free.

Read the newsletter trusted by 4.5 million fact-seekers.

_{A quick reminder that premium subscribers don’t see ads (or this note). Sign up here!}

So how do we get a Hubble image from above then? We don’t, not really. The image is actually a mosaic made from several observations, all of which show the pole from a shallow angle at different parts of Saturn’s day (as shown in the top part of the image above). Over time the whole decagon was seen, and astronomers could create a mosaic of it unrolled, like it had been peeled off Saturn and laid flat (astrophotographer Damian Peach did this in a cool image of Jupiter as an example). From there you can use geometry and map projection to reconstruct the image as if it’s seen from above. I’ve done similar processing myself to look at circular structures and be able to analyze them more easily.

The decagon is huge, too. Each side is nearly 17,000 kilometers long — Earth is about 13,000 km in diameter, for comparison, so our entire planet could pass easily through one side of this structure without touching the vertices.

The astronomers point out that there’s a big anticyclonic storm near the decagon (not seen in this image) that may play a role in its formation, though how isn’t clear. While the overall physics of these kinds of features is understood, the details are incredibly complex. 

The bizarre hexagon at Saturn’s north pole, imaged by Cassini. Credit: Cassini / NASA, JPL-Caltech, Space Science Institute / image processing Mark McCaughrean, CC BY-SA

I remember back in 2012 when the hexagon at Saturn’s north pole was clearly seen in Cassini images, and being really amazed. Of course some people cried “aliens” — some people always do, even without, y’know, evidence — but in my opinion that shows a lack of imagination. It’s easy to simply say aliens did something. The hard part is showing how nature might be able to do it.

And here we have exactly that. Nature is clever, more clever than we are, in that it can still surprise us by creating structures we may not have expected. But once we see them, study them, we can apply our scientific knowledge to them and explain them. That’s why science will always be cooler than pseudoscience.

Did you ever wonder how real scientists are searching for alien life? What they look for in exoplanets, and how they can determine what’s going on in a world a quadrillion kilometers away?

I’m moderating a panel of real astrobiology scientists called “Strange New Worlds: The Science of Star Trek”, hosted by the Carnegie Science Earth and Planets Laboratory on Wednesday, September 9, 2026. Panelists include Mike Wong, Maggie Thompson, and Nicole Wallack. We’ll be talking about alien planets in the context of Star Trek, comparing the fact and fiction while nerding out about both.

The panel will be at 6:30 p.m. at the Greenwalt Lecture Hall in Washington DC. It’s free, but you need to register so they get a head count. There will be a live stream, too.

All the info you need is on their web page for the event. I hope y’all can come!

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

The Nancy Grace Roman Space Telescope launched into space yesterday (August 30, 2026) at 11:26 UTC aboard a Falcon Heavy rocket! It’s headed for the Earth-Sun L2 point, a gravitational stable region of space about 1.5 million kilometers from Earth, where it can remain without much need for station-keeping. It’ll get there in a month or so, then go through an extended checkout phase before routine operations start. 

I wrote about Roman’s capabilities here on BAN in Issue #968, and just wrote an article for my column in Scientific American about its coronagraphic instrument that will get direct observations of exoplanets. Roman is much like Hubble, except it has a far wider field of view, and can see an area on the sky 100 times larger than Hubble! What takes Hubble many hours Roman can do in one shot, though at about half the resolution (so it can’t see details quite as well, but still way better than most other telescopes). Its main camera, the Wide-Field Instrument (WFI), has a wider wavelength range, too, capable of detecting light from blue out to near infrared at about 2.3 microns (roughly three times longer than the reddest color your eye can see). This overlaps with JWST, too, so that’s cool. That helps tie observations together from different missions.

Moments after Roman was deployed: the top of the observatory is visible with a radio dish it’ll use to send data to Earth. The struts are part of the structure that held it in place during launch. Credit: NASA

I hope for very big things from this machine, including incredibly dramatic images of the sky that will be impossible to display at full resolution due to their immense size (the WFI is a 300 megapixel camera, after all!). But even the sub-images will be amazing, and the science — looking to characterize dark energy, among many many other goals — will be fantastic.

But that’s months away now. Calibrating and testing the cameras will take a while — I’ve been through this with Hubble, and I know the folks on the Roman team will be very busy for the next few months. Congrats to all of them for a successful launch, and lets hope the rest of the mission goes as well. 

For updates, NASA has a blog set up with current news.

Speaking of space telescopes…

Every now and again, you just need a dynamite image of a stunning spiral galaxy from Hubble Space Telescope.

I am here to serve:

NGC 3137, courtesy of Hubble. Credit: ESA/Hubble & NASA, D. Thilker and the PHANGS-HST Team

Woof. That’s NGC 3137, a spiral galaxy about 50 million light-years from us. If we saw it face-on it would likely look circular, with its broad disk of wide-flung arms. However, by happenstance we see it from a much lower angle, closer to edge-on. This foreshortens the disk, making it look more like an oval, and makes the arms a bit harder to see. 

I found this interesting: it’s one of two large spiral members of a small group of galaxies (the other spiral being NGC 3175). The group has a lot of smaller satellite galaxies too… and if this sounds familiar, it’s because our own Local Group of galaxies is similar, with the Milky Way and Andromeda being the lead members, and a bunch of smaller galaxies tagging along. Nomenclaturally speaking (a word I just made up), there isn’t really a dividing line between a group and a cluster of galaxies; they’re just somewhat informally designated; a group is a few bigger galaxies with perhaps lots of smaller ones, but a cluster is larger, with dozens or hundreds (or thousands!) or big galaxies.

Anyway, I was really drawn in by this image. NGC 3137 has a lot of star-forming nebulae, hydrogen gas clouds seen as reddish blobs strewn along the spiral arms. There are also a lot of clumps of blue stars — these are young, massive stars that can be many thousands of times more luminous than the sun, so they shine brightly and are easy to see even though they are relatively rare. They don’t live long, so they tend to be near the nebulae in which they were born.

Note that the center of the galaxy, called the bulge, is yellowish; star formation ceased there long ago, and all the blue stars have long since exploded as supernovae. All that’s left are stars that are yellower, oranger, and redder.

These colors are interesting, and are so vivid I was suspicious about how the image was made. Sure enough, instead of just using three color filters to mimic natural light (like how our eyes see), this image uses six! What does that mean?

Each filter lets through a specific range of colors of light. Then, after some processing of the individual images to clean them up, a visual artist combines them. Each filter is assigned a color to display, and in this case what you see as violet is actually light from an ultraviolet filter, blue is actually blue, teal is blue-green, green in the image is actually yellow-green light, and near-infrared is displayed as red. In addition, they used a filter that lets through a very narrow slice of red light where hydrogen and nitrogen atoms glow brightly, and those are abundant in gas clouds. That’s also colored red in the image, so these gaseous star-forming factories really stand out.

Back in the day (the late 90s, mostly) I tried my hand at making color Hubble images, and I never got the color balance right. On the other hand, I’m no expert in using that kind of software (like Photoshop). Many other people are, and I appreciate not just the beauty of their images but the amount of work that goes into it as well. That’s one reason I like to give some details on it when I can.

The deep sky is wondrous and gorgeous, and it only gets better when you understand what you’re seeing. I’m happy to oblige.

You can email me at thebadastronomer@gmail.com (though replies can take a while), and all my social media outlets are gathered together at about.me. Also, if you don’t already, please subscribe to this newsletter! And feel free to tell a friend or nine, too. Thanks!

Note: Because the lunar eclipse is something almost everyone can enjoy, I’m sending this part of the newsletter to all free subscribers as well. I hope you get a chance to look!

On Thursday night/Friday morning, August 27-28, the moon will slide almost but not quite entirely into Earth’s shadow, creating a lunar eclipse. It’s technically a partial eclipse, since the moon won’t be completely dark, but it’s very close to full, with about 93% of the moon’s diameter plunged into shadow. 

It’s a bit weird to think that Earth has a shadow, since it’s in space; we’re used to a shadow being cast onto something. But think of it this way: there is a region of space behind Earth, opposite the direction to the sun, where, if you were there, you’d see Earth blocking the sun. That’s the shadow.

It’s shaped like a looooong cone, and at the distance of the moon is about 1.5° in width, bigger than the moon’s apparent 0.5° size. The moon doesn’t pass through this area every month because its orbit is tilted to the plane of Earth’s orbit around the sun by about 5°, so usually it misses. But sometimes things line up just right and it slides right through Earth’s shadow.

From the moon’s point of view Earth blocks the sun, and from our point of view we see the moon grow dark. Because the shadow is cone-shaped, as the moon enters the shadow we see a slight circular arc cut into it. That grows and grows, like a cookie with a bigger and bigger bite taken out of it. If the moon’s trajectory is just so it’ll pass all the way into the shadow. This time, though, it only traverses the outer part, with just a little bit of its surface outside the shadow. At maximum eclipse it’ll look like there’s a thin crescent still lit. 

Unlike solar eclipses, which happen on Earth in a narrow path so you have to be at the right place at the right time, a lunar eclipse happens on the moon, so wherever you are, if the moon is up in the sky during the eclipse you can watch it. This week’s event can be seen by all the Americas, essentially all of Africa, Europe, and the Middle East. 

You can go to the terrific TimeAndDate.com website and enter your location to get your own personal viewing times. For the east coast of the US the moon starts to enter the dark part of Earth’s shadow just after 22:30. Maximum eclipse is at 00:13 (just after midnight), and the moon leaves the shadow at 01:52. That’s great timing for me; it means I don’t have to stay up super late to watch at least some of it. The farther east you are the later it all happens on your clock. For west coasters the eclipse starts just before or after moonrise, depending on your exact location.

I like lunar eclipses; they’re slow and ponderous, so you can go outside, take a look, then go back in, then head back out 15 minutes later to see the progress. It’s also fun to take a look through binoculars or a telescope to see the shadowed regions only dimly illuminated by sunlight that’s passed through Earth’s air… which is also why the moon can look red during a lunar eclipse. Speaking of which, I have way more info about lunar eclipses in general in my Crash Course Astronomy eclipse episode.

I hope you have clear skies!