Source of inspiration: On July 31, 2026, the NSF-DOE Vera C. Rubin Observatory released its first full scientific release — an image of the COSMOS field covering approximately 45 square degrees (roughly 200 full Moons), containing over 500,000 galaxies and 50,000 Milky Way stars. This is the first major image from LSST Camera — the largest digital camera in the history of astronomy.
When Galileo pointed his telescope at Jupiter, he saw four dots next to the planet's disk. When Hubble looks into a dark patch, astrophysicists wait for a single exposure lasting several days. All of this is desktop astronomy: one instrument, one object, one moment in time.
Vera C. Rubin Observatory works differently. This is not a telescope, but a movie camera for the Universe. Every 40 seconds, its LSST Camera produces a 3.2-gigapixel frame. In one night there are hundreds of such frames, totaling 20 TB of raw data per night. Over 10 years, the Legacy Survey of Space and Time (LSST) will cover the southern sky ~825 times, making **30 trillion measurements** — ten times more than all the words ever spoken by humanity.
And on July 31, 2026, this machine truly "opened its eyes" for the first time on the COSMOS field — one of the most studied patches of astronomy, where the Hubble Space Telescope once showed us galaxies in deep field. Only now — in a single 15-second frame, without countless orbits and exposures.
LSST Camera is not just a "big camera." This is an engineering masterpiece from SLAC National Accelerator Laboratory, hand-assembled over 20 years. Key numbers:
| Parameter | Value |
|---|---|
| Full resolution | 3,200 megapixels |
| Number of science CCDs | 189 units |
| Assembly (rafts) | 21 rafts + 4 corner (for guiding and wavefront) |
| Pixel size | 0.2 arcseconds (≈ 10 microns) |
| CCD frame | 4000 × 4000 pixels |
| Full frame readout time | 2 seconds |
| Shutter | two shutters, opening/closing in 1 second |
| Filter change | 2 minutes |
| Filters | u, g, r, i, z, y — 6 total, 5 in the carousel simultaneously |
Curious detail: the sensors came from two different manufacturers — e2v (England) and ITL (USA). Each raft has only one vendor. Differences in characteristics (for example, readout noise of 5.4 e⁻ for e2v versus 6.21 e⁻ for ITL) are accounted for at the ISR (Instrument Signature Removal) stage — this is a software layer that "subtracts" individual sensor defects from the data before scientific analysis.
Twenty-one rafts in the focal plane is roughly like assembling a mosaic from 21 X-ray images, where there are 0.5 mm gaps between pieces, and each piece must be calibrated separately. Only the mosaic size is 64 centimeters in diameter.
The COSMOS field (Cosmic Evolution Survey) is "a patch of sky with history." Since the 2000s, it's been mapped by Hubble, Spitzer, XMM-Newton, Subaru, VLA and a dozen other instruments. Here lies the COSMOS-Web deep field — the largest deep space image obtained by Webb (1.5 million galaxies in the frame).
But Rubin looks at COSMOS not like Webb (pointedly, deeply), but broadly and frequently. This is completely different science:
In one of those Hubble releases, it showed a galaxy cluster on COSMOS lensing a background quasar — Rubin found dozens of such systems in a single frame.
The observatory was designed for four specific tasks:
The nature of dark matter and dark energy. Over 10 years, Rubin will accumulate ~10 million galaxies with spectroscopic redshifts (through joint programs with DESI and 4MOST). Combining weak gravitational lensing (from galaxy shapes) and Type Ia supernovae (standard candles), astronomers will be able to measure the equation of state of dark energy w with accuracy < 1%. After DESI (w ≈ −0.8 in 2024), this will be the major update to our map of the Universe's expansion.
Solar System inventory. Rubin should increase the number of known trans-Neptunian objects by ~10 times. A catalog of ~5 million asteroids, ~100,000 Kuiper belt objects, and by estimates, the discovery of several new dwarf planets like Pluto is expected. This is critically important: only understanding the structure of the outer Solar System can tell us whether "Planet 9" exists.
Milky Way map. 50,000 stars in the first frame is only the beginning. Over 10 years, LSST will measure parallaxes and proper motions of ~10 billion stars in the Galaxy, building a 6D map (3 coordinates + 3 velocities) — this will be "Gaia on steroids."
Variable sky. Alert stream — ~10 million events per night (more than the combined alert stream of all other sky surveys combined). Each alert is a JSON packet sized ~1 KB, distributed within 60 seconds after exposure through a broker system (Antares, Fink, Lasair, ALeRCE, BABAMUL, Lasair). The alert pipeline in the first weeks of operation has already detected several interstellar object candidates and suspiciously fast transients.
There are no perfect instruments. Rubin has several serious challenges:
Vera Cooper Rubin (1928–2016) was an American astronomer who in the 1970s, working at Carnegie Institution, measured galaxy rotation curves and discovered that stars on the periphery rotate too fast — as if pulled by invisible mass. This became one of the main observational proofs of the existence of dark matter. For 40 years no one listened to her work, and then she received the National Medal of Science and became the second woman in history to be awarded the Gold Medal of the Royal Astronomical Society.
When the observatory was renamed in her honor in 2020, many astrophysicists wept. This was a rare event for academia: a person whose discovery was marginalized for decades was elevated as an icon.
The Rubin image is not just a pretty picture. This is a test frame from an instrument that will photograph the Universe every 40 seconds for 10 years. During this time it will:
This is the first time we have an instrument whose main scientific task is to find what we don't yet know.
And on July 31, 2026, it took its first step — showed us 500,000 galaxies in a frame and said: "Hello, I'm Rubin. Now I'll be photographing the Universe every 40 seconds. The next 10 years are mine."
Sources: