Transcription
In December 1995, a scientist gambled ten days of the world's most in-demand telescope with a bad reputation on a patch of sky that looked completely empty, and one of the most respected astronomers involved in building that very telescope personally tried to talk him out of it twice.
I'm Doctor Kirsten Banks, Astro Kirsten, if you will. I'm an astrophysicist and science communicator. And before I got my PhD, I used to show people the payoff from that gamble for a living. And I still get chills every single time I see it. By the end of this video, you're going to understand why. And you're never going to look at a dark patch of empty night sky the same way again.
Here's why this gamble was genuinely a big deal. This is Robert Williams, and in the early years of operation of the Hubble Space Telescope, he was the director of the Space Telescope Science Institute. That's the organization that runs Hubble's science operations from day to day. Nowadays, it's also the science operation center for JWST and the Nancy Grace Roman space telescopes.
Now, normally, if you want time on Hubble, you submit a proposal and it goes through a competitive peer review process. And typically, only somewhere around 15 to 20% of proposals ever get approved. That's how scarce and contested Hubble time actually is. But directors get something called discretionary time, a chunk of observing time that they can assign themselves without going through that rigorous competitive review that every other scientist has to go through. It's a very privileged position to be in.
Now, most directors use that time for obvious things that are unpredictable, like chasing a supernova that just went off. But Williams decided to spend a huge chunk of his time on this, instead, pointing Hubble at one tiny, deliberately boring patch of sky for ten straight days. But what made this a real gamble wasn't just the choice of deciding to stare at nothing. It was because Hubble, at the time, was recovering from a bad reputation. Its main mirror had been launched with a flaw in it back in 1990. It was a genuinely embarrassing and very public failure, and it had only just been fixed by a repair mission a couple of years earlier. So spending ten full days on a newly repaired, heavily criticized telescope staring at nothing was the kind of decision that could look very, very bad if it didn't pay off. And one of the most respected astronomers involved in Hubble's own creation went to Williams personally more than once to try and talk him out of it, because he believed that Hubble wouldn't be able to reveal any new populations of distant galaxies. It seemed to be a very bad idea, but Williams used his discretion and went ahead anyway.
Here's the patch of sky they pointed at. Take a look at it. Genuinely, you can't see anything there. This patch of sky is about the width of a tennis ball held 100m away. If you looked at it with your own eyes on the clearest night imaginable, with no light pollution, you would see essentially nothing there.
But here's what came back after pointing Hubble at that exact spot for almost ten days. Almost every single point of light in this image. Every blob, smudge, even every tiny speck of light is not a star. It's a galaxy, each containing hundreds of billions of stars of their own. The first count came in at roughly 1500 galaxies. Later, more detailed analysis of that same image raised the number to around 3000. At every stage of evolution, packed into that teeny tiny single patch of sky. Before Hubble saw its first light, Bahcall predicted that in a typical 20-minute exposure, Hubble would spy a modest 100 galaxies and maybe 20 stars. Williams' ten-day gamble absolutely obliterated that prediction. And if you do the math across the entire sky in every direction, you start to get a sense of how overwhelmingly big and full the universe really is.
So here's the obvious question: If there's nothing to see there, how did they see all of this? The answer is time, and a huge amount of it. Think of a camera like a bucket and light like rain. Take a quick snapshot and you only catch the heaviest raindrops, the brightest and closest things. But leave that bucket out there long enough and you start catching every last drop, even the faintest ones that would have been completely invisible in a quick glance. That's what a longer exposure does with light. To capture this image, Hubble stared at that one spot for ten consecutive days, December 18th to the 28th in 1995, taking 342 separate exposures, one after the other, stacking them into a single picture. Now, ten days doesn't sound like much, but work it out in seconds. And Hubble's shutter was open for almost 1,000,000 seconds just to catch light from that patch of sky that looked completely black to the naked eye.
Most of the light from those galaxies is almost unimaginably faint by the time it reaches us, because it's traveled for billions of years across an expanding space to get here. Which means you're not just looking at galaxies that are really far away. You're looking back in time. Some of that light left its galaxy before our planet even existed.
Now, before I did this yapping about space online, I was a tour guide at Sydney Observatory. And yapped about space in person. I'd lead groups of about 20 people around the place, talking about the universe, the history of the site, looking through telescopes old and new. My favorite one was the one that had actually been used to observe the transit of Venus back in the 1800s. You had to wear these white gloves to touch it so you would protect the brass. And there was just something about looking through that old glass that made the universe pop in a way that modern mirrored telescopes never quite matched. Then we'd head into our planetarium, either to escape the clouds on a terrible night or avoid the light pollution of Sydney. And there was one moment in every single tour that I looked forward to more than any other. Near the end of my time in the planetarium, I'd lock onto the exact coordinates of the Hubble Deep Field. And before I zoomed in, I'd build up the story. The same story that I had just told you. The gamble, the bet nobody thought would pay off. And then I'd press the zoom button exactly ten times. Not nine. Not 11, but ten. Coincidentally, one for every day Hubble spent staring at this patch of sky. And every single time, people would try to gasp, silently thinking no one noticed. But in a domed room, sound carries and nothing gets past me and my ears in there. And honestly, sometimes, even after doing it hundreds of times, I'd catch my own breath too. Because the universe, while it may be 13.8 billion years old, somehow it never gets old for me.
And here's a part that still gets me. Scientists didn't stop here. After Williams' bet paid off, they took the same method: pick a deliberately boring, empty-looking patch of sky and stare at it for as long as possible, and pushed it further again and again, each time seeing further back in time. Then, before. In 2004, Hubble aimed at a different patch of sky for the Ultra Deep Field, 800 exposures, 400 orbits around Earth, over 11 days of total exposure. The count of galaxies came back at nearly 10,000. Some of them among the most distant ever found at the time. Light from when the universe was just 800 million years old. In 2012, instead of a new observation, astronomers stacked a decade of the same data together, the Extreme Deep Field, over 2,000,000 seconds of combined exposure. Now, it actually shows fewer galaxies, about 5500, not 10,000, because it's a deeper and narrower crop of the same sky, but deeper means further back. Some of these galaxies appear as they were when the universe was under 5% of its current age. Then, by 2022, the then brand new telescope, JWST, tried something similar. Again, not the same patch of sky, but a different field entirely, centered on a galaxy cluster called SMACS 0723. But what JWST did was a generational leap. JWST needed just 12.5 hours, a fraction of Hubble's ten days, to produce the deepest, sharpest infrared image of the universe. Yet.
So think back to where we started. One scientist gambled ten days of the world's most valuable telescope on a patch of sky that looked completely empty, against the advice of one of his own field's most respected voices. And thankfully, that gamble absolutely paid off. And it fundamentally changed what we understood the universe to contain.
Next clear night you get, go outside and find a dark patch of sky. It doesn't really matter which one, but look at the space between the stars that you can see and realize, it's not empty. If this kind of thing gives you chills the way it still gives me chills, stick around a while and explore the universe more with me. And in the meantime, stay curious and keep looking up.