[Beginner → Intermediate] · The electromagnetic spectrum, what telescopes actually do, and why spectroscopy is the most important technique in the field.

The spectrum

Visible light is a sliver. Each band shows you different physics, and each needed its own instrument to open:

BandRevealsObserved from
RadioCold hydrogen, pulsars, the CMB, jetsGround
MicrowaveThe cosmic microwave backgroundSpace (mostly)
InfraredDust-obscured regions, cool stars, early galaxiesSpace + high dry sites
VisibleStars, galaxies — what our eyes evolved forGround
UltravioletHot young stars, accretionSpace only
X-rayBlack hole accretion, hot gas, supernova remnantsSpace only
GammaThe most energetic events in the universeSpace only

The atmosphere is opaque to most of it. Only radio and visible pass through well, with some infrared windows. That is why space telescopes exist — not to be closer, but to be above the air. Hubble is only ~540 km up; the point is the vacuum.

Why JWST is infrared and Hubble mostly isn’t: the most distant galaxies are redshifted so far that their ultraviolet and visible light has stretched into the infrared. To see the early universe you must look in the infrared — and to do that you must be cold, which is why JWST sits at L2 behind a tennis-court-sized sunshield at ~40 K.

What a telescope actually does

Two jobs, and the first is the important one:

1. Collect light. Light-gathering power scales with aperture area — the square of the diameter. A 200 mm mirror gathers 4× a 100 mm. This is why aperture dominates every other specification, and why “magnification” on a box is marketing.

2. Resolve detail. The diffraction limit: — bigger aperture, finer detail.

Ground-based telescopes rarely reach their diffraction limit, because the atmosphere blurs everything — “seeing”, typically 1–2 arcseconds. Two answers:

  • Adaptive optics — measure the distortion with a guide star (sometimes a laser-created artificial one) and deform a mirror hundreds of times a second to cancel it. Ground telescopes can now beat Hubble’s resolution in the infrared
  • Go to space

Interferometry — combine several telescopes so resolution depends on their separation rather than their size. The Event Horizon Telescope did this across the whole Earth to image a black hole’s shadow. Standard in radio; increasingly used in the optical.

Detectors

Photographic plates → CCDs was the field’s quiet revolution. A photographic emulsion captures ~2% of incoming photons; a CCD captures >90%. Overnight, every telescope became roughly 40× more powerful, and photometry became precise rather than approximate.

CMOS sensors — the technology in your phone — now compete with and often beat CCDs, and they made serious amateur astrophotography affordable.

Spectroscopy — the technique that matters most

Split light into its spectrum and a point of light becomes a physics experiment.

Three spectrum types (Kirchhoff’s laws):

  • Continuous — a hot dense object; a smooth rainbow
  • Emission lines — a hot thin gas; bright lines at specific wavelengths
  • Absorption lines — a continuous source seen through cooler gas; dark lines

Each element produces a unique set of lines — a fingerprint. So from a spectrum you read:

  • Composition — which elements, and how much
  • Temperature — from the continuum shape and which lines appear
  • Density and pressure — from line widths
  • Magnetic field — from Zeeman splitting
  • Motion — from Doppler shift

Helium was discovered in the Sun’s spectrum in 1868, 27 years before it was found on Earth. That single fact is the best argument for spectroscopy’s power: an element identified in an object nobody could ever visit.

Redshift

A Doppler shift of the whole spectrum toward longer wavelengths. Because the line pattern is known, you can measure the shift precisely even for objects too faint to see detail in.

Three distinct causes, and conflating them is a common error:

  • Doppler — genuine motion away from us
  • Cosmological — the expansion of space itself stretching light in transit. Not motion through space → note 07
  • Gravitational — light climbing out of a gravity well → note 06

Cosmological redshift, denoted , is astronomy’s clock. It’s how “distance” and “lookback time” are actually quoted for anything far away.

The modern reality: surveys and pipelines

Astronomy is now a data science. The Vera Rubin Observatory will image the entire southern sky every few nights, producing ~20 TB nightly and millions of alerts — transients, variables, moving objects.

Nobody looks at those images by eye. The work is pipelines, classification, and increasingly machine learning → AI & ML. Gaia measured positions and motions for ~2 billion stars; that catalogue alone has generated thousands of papers.

And the data is largely public, usually after a short proprietary period. Anyone can download and analyse it → note 10.

Source: [reference] — written Aug 2026.

↑ Back to Astronomy and Space