What the James Webb Space Telescope Has Actually Found

The James Webb Space Telescope is routinely described as Hubble's replacement. It is not. The two are different instruments built to answer different questions, and Hubble is still working.

The distinction that matters is wavelength.

Why infrared changes everything

Hubble observes mainly in visible and ultraviolet light, with some near-infrared. JWST is built almost entirely for infrared. That single choice drives every other design decision, and it buys three things Hubble cannot deliver.

Seeing the earliest galaxies. The universe is expanding, so light from distant objects is stretched to longer wavelengths on its way to us. Light emitted as visible or ultraviolet by a galaxy in the first few hundred million years arrives here as infrared. Hubble physically cannot see it; the light is no longer in its range. JWST can.

Seeing through dust. Stars form inside dense molecular clouds, which visible light cannot penetrate. Infrared passes through far more readily. Regions Hubble renders as opaque dark columns, JWST shows as nurseries full of forming stars — the two telescopes' images of the Pillars of Creation make the comparison immediately obvious.

Reading exoplanet atmospheres. Molecules leave their strongest absorption fingerprints in the infrared. Water, carbon dioxide, methane and sulfur dioxide are all far more detectable there.

What that required

Infrared is heat. A warm telescope glows in exactly the band it is trying to observe, so JWST had to be made extremely cold — under 50 K for the main instruments, and around 7 K for the mid-infrared instrument MIRI, which needs its own cryocooler.

That is why JWST does not orbit Earth. It sits at the second Sun-Earth Lagrange point, 1.5 million kilometres away, four times further than the Moon, where Earth and Sun stay in roughly the same direction. That geometry lets a single five-layer sunshield the size of a tennis court block heat from Sun, Earth and Moon simultaneously. The sunshield's sun-facing side runs at about 85 °C; the far side at about −233 °C.

The primary mirror is 6.5 metres across, against Hubble's 2.4, giving it roughly six times the light-collecting area. No rocket fairing could accommodate a 6.5-metre mirror, so it was built as 18 gold-coated beryllium hexagons that folded for launch and unfolded in space, with the sunshield unfurling separately.

The deployment involved 344 single-point failures — mechanisms with no backup, any one of which could have ended the mission. All of them worked. And unlike Hubble, which astronauts serviced five times, JWST is far beyond any possibility of repair. Everything had to work the first time, permanently.

Hubble and JWST compared

They are complementary rather than competing. Hubble retains ultraviolet capability JWST lacks entirely, and there is no planned successor for that. Observations combining both are common.

The galaxies that should not be there

JWST's most significant early result was also its most awkward.

It found galaxies at very high redshift — seen as they were within the first few hundred million years after the Big Bang — that appeared far more massive and more mature than models of galaxy formation predicted. Some looked like they had assembled billions of stars' worth of mass in a period when they should still have been assembling.

The reporting around this was frequently overheated, with headlines announcing that JWST had broken cosmology. It had not, and the more careful reading is worth understanding.

Redshifts initially estimated from photometry — colours through filters — are less reliable than spectroscopic measurements, and follow-up spectroscopy revised some candidates downward. Separately, several of the most extreme objects turned out to be "little red dots": compact sources whose light appears to be dominated by an accreting black hole rather than by starlight, meaning the inferred stellar masses were substantial overestimates.

What survives is still significant. Early galaxies do appear to have formed faster and more efficiently than standard models expected, and the abundance of massive black holes early on has sharpened the question of how supermassive black holes grew so quickly. This is a real tension requiring adjustment to galaxy formation models — not a refutation of the Big Bang, which rests on entirely independent evidence from the cosmic microwave background and light-element abundances.

Exoplanet atmospheres

JWST has delivered on its atmospheric promise. It made the first unambiguous detection of carbon dioxide in an exoplanet atmosphere (WASP-39b), and found sulfur dioxide there — evidence of photochemistry driven by the host star's light, a process previously known only from theory.

The TRAPPIST-1 system, with seven Earth-sized planets around a nearby red dwarf, has been a priority target. Results for the innermost planets point toward thin atmospheres or none, consistent with stellar flares having stripped them. That is not the answer anyone was hoping for, but it is exactly the kind of measurement that turns habitability from speculation into a testable claim — and the outer planets in the system remain under investigation.

Closer to home

JWST's solar system work is easy to overlook. It has imaged Neptune's rings with a clarity not achieved since Voyager 2's 1989 flyby, watched storms on Jupiter, detected carbon dioxide on Europa's surface in a form suggesting it originated in the subsurface ocean, and resolved the nested debris belts around Fomalhaut that revised a well-known planetary system.

What it cannot do

JWST has real limits, and they are worth stating plainly. It has no ultraviolet capability. Its field of view is small, so it is a precision instrument rather than a survey machine — the Vera Rubin Observatory and Euclid cover that role. Observing time is heavily oversubscribed, with far more proposals than can be accommodated.

And it has a finite life. Its propellant reserve for station-keeping at L2 was expected to last around ten years, though an unusually efficient launch left far more margin than planned and current estimates run to roughly twenty. When the fuel runs out, the mission ends. There is no servicing option.

That makes the science being done now a limited window — which is a good argument for paying attention to it while it is open.