Astronomy 2e · The Evolution and Distribution of Galaxies
Observations of Distant Galaxies
On this page 9 sections
In 30 seconds
Light travels at a finite speed, so every telescope is a time machine: a galaxy 5 billion light-years away is seen as it was 5 billion years ago. By photographing tiny patches of sky for days, deep surveys have collected thousands of galaxies at every stage of cosmic history — and the picture is clear: galaxies change dramatically over time.
The Hubble Deep field Very long exposure of a blank patch of sky Full entry → (1995) — a ~10-day exposure of a patch of sky smaller than a grain of sand at arm's length — revealed ~3,000 galaxies where a few dozen were expected. Distant galaxies are smaller, bluer, lumpier, and bursting with star formation, while nearby galaxies are larger, redder, and more settled. This topic explains how we observe that evolution and measure galaxy distances.
Why this matters
- It's the direct evidence for galaxy evolution. High-redshift galaxies simply look different from local ones — the sky itself is the experiment.
- It connects Chapters 26 and 27: redshift gives distance (Hubble's law), distance gives Lookback time How far in the past we see an object at distance d, and deep surveys show us whole galaxies in their youth.
- It explains today's observatories. JWST observes in the infrared because young galaxies' light is redshifted there; huge telescopes and long exposures are needed because distant galaxies are faint and their light is spread thin.
- Exam staples: lookback time, the Hubble Deep Field, spectroscopic vs. photometric redshifts, Surface brightness dimming Fading of brightness per area, ∝ (1+z)⁴ Full entry →, and "Cosmic noon Peak of cosmic star formation, z ≈ 2 Full entry →."
The college version
Core Concepts
Telescopes as time machines
Distance is time. Under Hubble's law (Chapter 26), redshift z measures distance, and distance measures lookback time:
- z = 1 → ~7.7 Gyr of lookback (universe ~6 Gyr old)
- z = 3 → ~11.5 Gyr of lookback (universe ~2 Gyr old)
- z > 13 (JWST's record-holding galaxies) → seen less than ~300 Myr after the Big Bang
(Commonly taught reference values — verify against current sources.) A deep image is therefore a slice through time: nearby galaxies in the foreground, progressively younger ones behind.
Deep-field surveys: staring at nothing to see everything
A deep field is a very long exposure of a deliberately blank, unremarkable patch of sky. The Hubble Deep Field (1995) piled ~10 days of exposure onto a ~2.5-arcminute field (roughly a grain of sand at arm's length) and found thousands of galaxies — most never seen before, most distant. JWST's deep fields now reach galaxies that formed within a few hundred million years of the Big Bang. "Depth" means faintness: the longer you stare, the fainter — and farther — the galaxies you detect.
Galaxy properties change with redshift
Comparing galaxies across redshift reveals systematic changes:
- Size and structure: high-redshift galaxies are smaller and often irregular or "peculiar" — clumpy, lopsided, disturbed — while grand-design spirals and mature ellipticals are common only nearby.
- Color and star formation: distant galaxies are bluer — the signature of young, massive, short-lived stars, i.e., intense ongoing star formation. The cosmic star-formation rate peaked around z ≈ 2 (dubbed "cosmic noon", ~10 billion years ago) and has declined by roughly an order of magnitude since (commonly taught figures).
- Mergers: disturbed morphologies and close pairs are far more common at high redshift; galaxies grow hierarchically by collisions and mergers.
Measuring distances: spectroscopic vs. photometric redshifts
To place a galaxy in time you need its redshift — two very different ways to get it:
- Spectroscopic redshift Redshift from identified spectral features Full entry →: identify spectral features (e.g., the Lyman-alpha line or absorption lines) and measure how far they've shifted. Accurate to ~0.1% — but slow and expensive, so only a fraction get this treatment.
- Photometric redshift Redshift estimated from broad-band colors Full entry →: match the galaxy's brightness through several broad filters to template spectra. Quick and works for very faint galaxies, but approximate — calibrated against spectroscopic samples. Deep surveys use photometric redshifts for the bulk and spectroscopy for the brightest subset.
The K-correction and why infrared surveys matter
A high-redshift galaxy's rest-frame light (what it actually emitted) arrives at longer observed wavelengths: a galaxy whose young stars emit in the ultraviolet is seen in the optical at z ≈ 2 and in the infrared at z ≈ 8. Comparing fluxes across redshifts requires a K-correction Adjustment comparing fluxes in the same rest-frame band Full entry → — placing every measurement in the same rest-frame band. This is exactly why JWST observes in the infrared: it catches the rest-frame ultraviolet and optical light of the first galaxies.
Challenges and biases in deep surveys
Observing the faintest galaxies is hard in specific ways:
- Surface brightness dimming: a galaxy's surface brightness fades as (1+z)⁴ (commonly taught) — distant galaxies are not only faint, their light is diluted over a larger apparent area, which is why the first galaxies are so hard to find even with JWST.
- Small angular sizes and confusion: distant galaxies are tiny and crowded; sharp images are needed to separate them.
- Selection effects Survey biases toward bright/active galaxies Full entry →: surveys preferentially find the brightest, most actively star-forming galaxies — an important caveat for every "galaxies back then were X" claim.
Common Confusions
| Do Not Confuse | With | Difference |
|---|---|---|
| A high-z galaxy being "far away now" | It being seen as it was long ago | Lookback time: distance = past, not present |
| All distant galaxies being small | Small apparent size/selection | Many are genuinely small, but the brightest dominate surveys |
| Red galaxies being young | Red = old stars or dust | Color tells age/dust, not distance; "red and dead" ellipticals are ancient |
| Photometric redshifts being precise | Spectroscopic redshifts | Photometric z is an estimate from colors; spectroscopy confirms |
| Blue light meaning hot stars only | Blue = recent star formation (in galaxies) | AGN light can also be blue — a known contaminant |
| Galaxy types being fixed | Types evolving | Mergers, gas depletion, and feedback transform galaxies |
| Deep surveys seeing "all" galaxies | The bright tip of the iceberg | Faintness limits + surface brightness dimming hide most |
Eli explains
The same idea, in plain words
Explain it like I’m 10
Looking far away is looking back in time, because light takes time to travel. A deep photograph of a tiny, empty-looking patch of sky is like a family photo album: galaxies up close are "now," and the tiny faint ones far away are baby pictures — smaller, messier, glowing blue with new stars. Comparing baby pictures with grown-up galaxies shows how galaxies grow up: they collide, merge, and slowly settle into the quiet spirals and ellipticals we see nearby.
Worked example
A survey image contains a tiny blue smudge. Here's the reasoning chain:
- Distance. Its photometric colors suggest z ≈ 3; a spectrograph confirms it — the Lyman-alpha line (rest 121.6 nm) appears at ~480 nm, so z ≈ 2.9. Hubble's law converts this to ~11.5 billion light-years — we see the galaxy when the universe was ~2 billion years old.
- Properties. The galaxy is small (a few thousand light-years across), blue (young hot stars), and lumpy (bright star-forming clumps, no orderly spiral arms).
- Interpretation. A young galaxy mid-assembly: gas-rich clumps turning into stars at a furious rate, destined to grow by merging over the next 10 billion years. Contrast it with a nearby giant elliptical — red, smooth, old stars — and you have the galaxy-evolution story in two objects.
Every claim traces back to a measurable: a wavelength shift, a color, an angular size.
Key takeaways
- Distance = time: z = 1 → ~7.7 Gyr lookback; z = 3 → ~11.5 Gyr (universe ~2 Gyr old); JWST reaches z > 13 (commonly taught).
- Deep fields (HDF 1995, HUDF, JWST): long stares at blank sky reveal thousands of galaxies across time.
- High-z galaxies are smaller, bluer, lumpier, and star-forming; star formation peaked at z ≈ 2 ("cosmic noon") and has since declined ~10× (commonly taught).
- Spectroscopic redshifts are precise but slow; photometric redshifts are approximate but fast.
- Surface brightness dims as (1+z)⁴ — the fundamental difficulty of seeing the first galaxies (commonly taught scaling).
- K-correction + infrared surveys (JWST) are needed because rest-frame UV of high-z galaxies arrives as infrared.
- Selection effects: deep surveys see the brightest, most active galaxies — not a fair sample of all.
Check yourself
6 review questions from the chapter. Try each one, then open the answer.
Why is observing a distant galaxy equivalent to observing the past?
Show answer
Light travels at a finite speed, so light from a galaxy at distance d takes d/c years to reach us; we see it as it was that long ago (lookback time).
What did the Hubble Deep Field demonstrate?
Show answer
That a long exposure of a blank, tiny patch of sky contains thousands of galaxies at all distances/epochs — the universe is full of galaxies we never knew existed, and deep fields sample cosmic time.
How do the sizes, colors, and shapes of galaxies change with increasing redshift?
Show answer
At high redshift, galaxies are smaller, bluer (young stars), and more irregular/lumpy; mature spirals and ellipticals dominate only at low redshift, and star formation was far more intense.
Compare spectroscopic and photometric redshifts: accuracy, speed, and when each is used.
Show answer
Spectroscopic redshifts use identified spectral features — precise but slow and requiring brighter targets; photometric redshifts use broad-band colors — fast and applicable to faint galaxies, but approximate and needing spectroscopic calibration.
Why does JWST observe the earliest galaxies in the infrared?
Show answer
The rest-frame ultraviolet light of early galaxies is redshifted into the infrared by z ~ 8–13, so infrared detectors (plus K-corrections) are required to see it.
What is "cosmic noon," and what has happened to star formation since?
Show answer
Cosmic noon is the peak of cosmic star-formation rate density around z ≈ 2 (~10 Gyr ago); since then the global star-formation rate has declined by roughly an order of magnitude (commonly taught figure).
Study tools & related lessonsKey vocabulary · Related
Key vocabulary
- Lookback time
- How far in the past we see an object at distance d
- Deep field
- Very long exposure of a blank patch of sky
- Spectroscopic redshift
- Redshift from identified spectral features
- Photometric redshift
- Redshift estimated from broad-band colors
- Rest frame
- Wavelengths as emitted by the galaxy
- K-correction
- Adjustment comparing fluxes in the same rest-frame band
- Surface brightness dimming
- Fading of brightness per area, ∝ (1+z)⁴
- Cosmic noon
- Peak of cosmic star formation, z ≈ 2
- Selection effects
- Survey biases toward bright/active galaxies
Sources & references
This lesson was adapted from the open educational references above; their licenses and attributions are preserved. See Copyright & Licensing.
Educational content only. It is not medical, legal or professional advice. Found an error? Tell us.