Space telescope science: multiwavelength astronomy, gravitational-wave observatories

Intuition Beginner

Earth's atmosphere blocks most of the light arriving from space. Only visible light, a slice of infrared, and some radio waves reach the ground. X-rays, gamma rays, most ultraviolet light, and the bulk of infrared light are absorbed or scattered long before they reach the surface. To study the cosmos in those hidden bands, telescopes must be lifted above the atmosphere, into orbit or deep space.

The Hubble Space Telescope, launched in 1990 and serviced by astronauts five times, observes from about 540 kilometres up. Freed from the atmosphere's blurring turbulence, it returns razor-sharp images in visible and ultraviolet light. Its deep-field exposures revealed thousands of galaxies in patches of sky no larger than a grain of sand held at arm's length, and its measurements pinned down the accelerating expansion of the universe.

The Chandra X-ray Observatory and the Fermi Gamma-ray Space Telescope probe the most violent phenomena: black holes tearing matter apart, exploding stars, and colliding galaxies. The James Webb Space Telescope, launched in 2021, is the largest space telescope ever flown. Operating in infrared light, it detects the faint heat glow of the first galaxies that formed just after the Big Bang.

In 2015 an entirely new channel opened. The LIGO detectors recorded gravitational waves, ripples in the fabric of spacetime itself, produced by two black holes merging more than a billion light-years away. For the first time, astronomers could observe the universe not by the light it emits but by the trembling of space.

Visual Beginner

Spectral band	Wavelength range	Reaches the ground?	Flagship mission
Radio	cm to m	Yes	Ground arrays: VLA, ALMA
Sub-millimetre	0.1 to 1 mm	Partial, high dry sites	Herschel, SPHEREx
Infrared	1 to 28 µm	Almost none	IRAS, Spitzer, JWST
Optical / UV	0.1 to 0.7 µm	Yes, but blurred	Hubble (HST)
X-ray	0.01 to 10 nm	No	Chandra, XMM-Newton, NuSTAR
Gamma-ray	below 0.01 nm	No	Fermi, Swift, INTEGRAL

Frequency band	Frequency	Observatory	Example source
Nanohertz	10^-9 to 10^-7 Hz	Pulsar timing arrays (NANOGrav, EPTA, PPTA)	Supermassive black-hole binaries
Millihertz	10^-4 to 10^-1 Hz	LISA (space, planned)	Merging massive black holes
Audio	10 to 1000 Hz	LIGO, Virgo, KAGRA	Stellar-mass black-hole and neutron-star mergers

Worked example Beginner

Example 1: A bigger mirror sees finer detail

A telescope's ability to distinguish fine detail is set by diffraction, the spreading of light around the edge of its mirror. The smallest resolvable angle is

$$\theta \approx 1.22\frac{\lambda }{D}$$

Here $\lambda$ is the wavelength of light and $D$ is the mirror diameter. Larger mirrors resolve finer detail and gather more light, so they reveal fainter and more distant objects.

For Hubble, $D=2.4$ m at visible wavelengths gives a resolution near $0.05$ arcseconds, sharp enough to separate a football from hundreds of kilometres away. The James Webb Space Telescope reaches comparable sharpness in infrared light with its $6.5$ m mirror.

Example 2: Why telescopes go to space

From the ground, Earth's atmosphere absorbs X-rays and gamma rays entirely, blocks most infrared and ultraviolet light, and blurs visible light with turbulent eddies. A telescope in orbit escapes all of these problems at once.

Hubble's celebrated sharpness comes not from an exceptional mirror but from its position above the atmosphere. An infrared telescope in space likewise sees through the cosmic dust that hides newborn stars and distant galaxies from any ground-based instrument.

Example 3: How small is a gravitational wave

A passing gravitational wave stretches space in one direction and squeezes it in the perpendicular direction. The strain $h$ measures this fractional change in length. LIGO's arms are $4$ km long, and a typical merger signal has strain $h\approx 10^{-21}$.

The resulting change in arm length is $4000\times 10^{-21}=4\times 10^{-18}$ m, about ten thousand times smaller than a proton. Detecting such a signal requires extraordinary isolation from vibration and the most precise length measurement ever made.

Check your understanding Beginner

Formal definition Intermediate+

The electromagnetic spectrum and atmospheric windows

Earth's atmosphere transmits only three windows: the optical window (roughly 300 to 700 nm), parts of the near-infrared, and the radio window (roughly 1 cm to 10 m). Ultraviolet light below 300 nm is absorbed by ozone; infrared is blocked by water vapour and carbon dioxide; X-rays and gamma rays are stopped high in the atmosphere. Space astronomy opened every other band to observation, and flagship missions define each field. On the ground, the Very Large Array (VLA) and the Atacama Large Millimeter/submillimeter Array (ALMA) cover radio and sub-millimetre wavelengths. In orbit, IRAS, Spitzer, and JWST observe the infrared; Hubble covers optical and ultraviolet; Chandra, XMM-Newton, and NuSTAR observe X-rays; and Fermi's Large Area Telescope (LAT) and Swift cover gamma-rays. Each band reveals different physics: radio maps cold gas and synchrotron emission from relativistic electrons; infrared penetrates dust to show star formation; X-rays expose million-degree gas and matter spiralling into black holes; gamma-rays trace cosmic rays and the most violent explosions.

Angular resolution, aperture, and the diffraction limit

A telescope's angular resolution, the smallest angle it can distinguish, is bounded by diffraction. For a circular aperture of diameter $D$ observing at wavelength $\lambda$, the Rayleigh criterion gives

$${\theta }_{\min }\approx 1.22\frac{\lambda }{D}$$

in radians. Resolution improves with larger apertures and shorter wavelengths. Collecting area scales as $A\propto D^2$, so a bigger mirror also gathers more photons and reaches fainter sources in less time. Hubble's 2.4 m mirror resolves about 0.05 arcsec at visible wavelengths; JWST's 6.5 m mirror resolves about 0.1 arcsec in the mid-infrared, where the longer wavelength partly offsets the larger aperture. Ground-based 8 to 10 m telescopes, corrected by adaptive optics, now rival or exceed Hubble's resolution in the near-infrared, but only over small fields around bright reference stars.

Spectroscopy

Most astrophysics lives in spectra, not images. A spectrograph spreads light by wavelength to reveal emission and absorption lines that encode temperature, density, composition, and velocity through the Doppler shift. Echelle spectrographs use high-order gratings to achieve resolving powers $R=\lambda /\Delta \lambda$ of tens of thousands, measuring exoplanet and stellar radial velocities at the metre-per-second level. Integral field units (IFUs) record a spectrum for every spatial element, producing data cubes over targets such as galaxies and nebulae. Grisms, gratings mounted on prisms, give lower-resolution spectroscopy over wide fields for surveys. JWST's NIRSpec and MIRI, Hubble's COS and STIS, and the X-ray microcalorimeter on XRISM illustrate the range from ultraviolet to X-ray energies.

Adaptive optics and interferometry

Adaptive optics corrects atmospheric turbulence in real time. A guide star, natural or laser-generated, drives a deformable mirror that flattens the incoming wavefront hundreds of times per second, restoring near-diffraction-limited resolution from the ground at infrared wavelengths. Interferometry goes further by combining light from separated telescopes. The resolution is then set not by a single dish diameter but by the baseline $B$ between telescopes, $\theta \sim \lambda /B$. ALMA combines up to 66 antennas across baselines of kilometres to image star formation at sub-millimetre wavelengths. Very Long Baseline Interferometry (VLBI) links radio dishes across continents, and the Event Horizon Telescope added the South Pole to reach an effective baseline the size of Earth, resolving the shadows of the supermassive black holes in M87 and in the Galactic Centre.

Gravitational-wave detection

Gravitational waves are propagating distortions of spacetime predicted by general relativity. A passing wave changes the relative length of two perpendicular arms. A laser interferometer such as LIGO splits a laser along two 4 km vacuum arms, bounces the beams between suspended mirrors, and recombines them; a gravitational wave shifts the interference pattern. The challenge is sensitivity: the strains to be measured are of order $h\sim 10^{-21}$. LIGO, Virgo (3 km arms in Italy), and KAGRA (Japan) form a network whose timing triangulates sources. LISA, a planned space mission with arm lengths of millions of kilometres, will target the millihertz band inaccessible from the ground, where supermassive black-hole binaries and compact stellar binaries radiate.

Multi-messenger astronomy

On 17 August 2017 the network detected GW170817, a gravitational-wave signal from two neutron stars merging, followed 1.7 seconds later by a short gamma-ray burst (GRB 170817A) and, over the following days, an optical and infrared kilonova and X-ray and radio afterglows. This single event, observed across gravitational waves and the entire electromagnetic spectrum, confirmed that neutron-star mergers forge heavy elements through r-process nucleosynthesis and validated the long-suspected link between mergers and short gamma-ray bursts. Multi-messenger astronomy, combining gravitational waves, electromagnetic radiation, and for some sources neutrinos, has become a distinct observational paradigm.

Key result: angular resolution and the matched-filter signal-to-noise ratio Intermediate+

Resolution: single aperture versus interferometer

For a filled circular aperture, the Rayleigh diffraction limit is ${\theta }_{\min }=1.22\lambda /D$. The number of resolution elements across a source of angular size $\Theta$ scales as $(\Theta /{\theta }_{\min }{)}^2\propto D^2/{\lambda }^2$. For an interferometer, the fringe angular scale is set instead by the baseline, $\theta \sim \lambda /B$, while the point-source sensitivity is governed by the total collecting area of the individual elements. This separation of resolution from collecting area is the central advantage of interferometry: ALMA, VLBI, and the Event Horizon Telescope achieve resolutions far beyond any single dish by trading field of view and surface-brightness sensitivity for angular detail.

Matched filtering and gravitational-wave detectability

A gravitational-wave detector records strain $s(t)=h(t;\Theta )+n(t)$, where $h$ is a signal from source parameters $\Theta$ and $n$ is detector noise. Assuming stationary Gaussian noise with one-sided power spectral density $S_n(f)$, the optimal linear statistic for a known waveform is the matched filter, whose signal-to-noise ratio satisfies

$${\rho }^2=4{\int }_0^{\infty }\frac{|\tilde{h}(f){|}^2}{S_n(f)}df$$

where the tilde denotes the Fourier transform. Detection requires $\rho$ above threshold, roughly 8 in a single detector and 12 to 25 in a coincident network. Because the true source parameters are unknown, analyses search a bank of template waveforms spanning plausible masses and spins. The templates themselves come from post-Newtonian theory, effective-one-body models, and full numerical relativity simulations. The inspiral of a binary black hole spends many cycles in band, building $\rho$ coherently and allowing extraction of the component masses, spins, and luminosity distance.

Exercises Intermediate+

Advanced results Master

Telescope optical design

Space telescopes favour designs that deliver wide, well-corrected fields and low scattering. The Ritchey-Chretien variant of the Cassegrain, with hyperbolic primary and secondary mirrors, corrects coma as well as spherical aberration and is used by Hubble and many ground observatories. Off-axis designs, in which the secondary does not block the primary, eliminate diffraction spikes and reduce stray light. Three-mirror anastigmats correct the three Seidel aberrations (spherical, coma, astigmatism) over wide fields and underlie survey telescopes. JWST uses a three-mirror anastigmat supplemented by a fine-steering mirror, with a 6.5 m primary assembled from 18 hexagonal beryllium segments each carrying independent actuators for phasing.

Thermal control and the JWST L2 architecture

Infrared detectors are themselves sources of thermal emission, so an infrared telescope must be colder than the sources it observes. JWST operates near the Sun-Earth L2 Lagrange point, about 1.5 million kilometres from Earth, where it keeps a fixed orientation relative to the Sun. A five-layer Kapton sunshield the size of a tennis court blocks solar radiation and radiatively cools the telescope to roughly 40 K; the mid-infrared instrument reaches about 7 K with a mechanical cryocooler. This passive thermal architecture avoids carrying consumable cryogen and yields a mission lifetime governed by station-keeping propellant and instrument health rather than coolant mass.

Detector technologies

Photon detection underlies every instrument. Silicon CCDs, used from the near-ultraviolet to about 1 µm, achieve read noise below a few electrons and near-unity quantum efficiency. For the infrared, HgCdTe arrays (H2RG devices, up to 2048 by 2048 pixels and larger) operate between roughly 30 and 140 K with low dark current. Transition-edge sensors (TES), held at the superconducting transition, measure individual photon energies with resolving power of order 10 at X-ray energies and underlie microcalorimeters such as XRISM's Resolve. More exotic technologies include microwave kinetic inductance detectors (MKID) and superconducting nanowire single-photon detectors (SNSPD), which combine time-tagging with energy resolution and are central to proposed ultraviolet and optical missions. Gamma-ray instruments instead rely on pair-conversion trackers (silicon strip and tungsten, as in Fermi-LAT) or scintillators.

Sensitivity and the radiometer equation

The noise-equivalent flux of a detector follows from the radiometer equation. For a system with noise-equivalent antenna temperature $T_{\mathrm{sys}}$, bandwidth $\Delta \nu$, and integration time $t$, the rms noise on a measurement scales as

$$\sigma \propto \frac{T_{\mathrm{sys}}}{\sqrt{\Delta \nu t}}$$

Sensitivity therefore improves only with the square root of observing time, so reaching twice as faint takes four times longer. This drives the demand for larger collecting area: because area enters linearly, a bigger aperture lowers the flux limit and shortens integrations. Background-limited observations, as in the far-infrared or from the ground, add further penalties from atmospheric and telescope emission.

Gravitational-wave data analysis: template banks and waveform models

Detection and parameter estimation rest on comparing the data against a bank of template waveforms. For compact binaries, the inspiral is described by the post-Newtonian expansion in small velocities; the late inspiral, merger, and ringdown require numerical relativity, the direct solution of the Einstein equations on supercomputers. Effective-one-body models and surrogate waveforms stitch these regimes into fast, accurate templates spanning the space of masses and spins. The template bank must be dense enough that the mismatch between any astrophysical signal and its nearest template is small, yet sparse enough to remain computable in a low-latency search. Detected events are then reanalysed with full Bayesian inference to extract posterior distributions for masses, spins, distance, inclination, and sky position.

LIGO noise sources and squeezed light

LIGO's sensitivity is bounded by four noise families. Seismic noise dominates below about 10 Hz and is suppressed by multi-stage pendulum suspensions. Thermal noise, from Brownian motion in the mirror coatings and suspensions, dominates the mid-band. At high frequency the limit is quantum shot noise, the photon-counting statistics of the laser. Radiation-pressure noise, the recoil of photons against the mirrors, dominates the low-frequency quantum regime. Shot noise and radiation-pressure noise form a trade-off: raising laser power reduces shot noise but increases radiation-pressure noise, defining the standard quantum limit. Injecting squeezed vacuum states into the dark port of the interferometer reshapes this trade-off, pushing sensitivity below the standard quantum limit in a chosen band, a technique now routine in both LIGO and Virgo.

LISA and the millihertz window

LISA will consist of three spacecraft in a triangular constellation with arms of roughly 2.5 million kilometres, trailing Earth around the Sun. Each spacecraft houses free-falling test masses shielded by drag-free control, and inter-spacecraft laser links measure their separations through time-delay interferometry, which cancels laser frequency noise without needing optical cavities between spacecraft. The millihertz band is rich: mergers of massive black holes at cosmological distances, extreme-mass-ratio inspirals of stellar remnants into massive black holes, and tens of thousands of compact Galactic binaries. Many sources will be detectable years before merger, enabling electromagnetic follow-up.

Pulsar timing arrays and the stochastic background

At nanohertz frequencies, where no human-built detector is conceivable, pulsar timing arrays probe gravitational waves by monitoring the arrival times of radio pulses from millisecond pulsars. A gravitational wave perturbs the Earth-pulsar path, inducing correlated timing residuals. The Hellings-Downs angular correlation predicts how residuals correlate across pairs of pulsars, the signature that distinguishes a gravitational-wave background from instrumental or astrophysical noise. NANOGrav, the European Pulsar Timing Array (EPTA), the Parkes Pulsar Timing Array (PPTA), and the combined International Pulsar Timing Array (IPTA) reported evidence for a stochastic background in 2023, consistent with a population of supermassive black-hole binaries, though exotic origins remain under study.

The Event Horizon Telescope and black-hole shadows

The Event Horizon Telescope (EHT) is a global VLBI array operating at 1.3 mm wavelength. By combining dishes from Hawaii, Chile, Europe, the South Pole, and elsewhere, it reaches an effective baseline of Earth's diameter and a resolution of roughly 20 microarcseconds. In 2019 the EHT released the first image of a black-hole shadow, the bright photon ring surrounding the event horizon of the 6.5 billion solar mass black hole at the centre of the galaxy M87. A subsequent image of Sagittarius A*, the 4 million solar mass black hole at the centre of the Milky Way, tested general relativity in the strong-field regime and constrained the black hole's spin and accretion flow.

Future flagship missions

The coming decades will extend multiwavelength and gravitational-wave astronomy. The Nancy Grace Roman Space Telescope will survey the infrared sky with a field of view about a hundred times Hubble's, probing dark energy and microlensing exoplanets. Proposed large ultraviolet, optical, and infrared missions (the decadal-survey-endorsed Habitable Worlds Observatory, descended from the LUVOIR and HabEx concepts) aim to image Earth-like exoplanets. Lynx, a candidate X-ray flagship, would pair high resolution with high throughput. ARIEL will conduct a transit spectroscopy survey of exoplanet atmospheres, and THESEUS would target gamma-ray bursts and the early universe. On the gravitational-wave side, third-generation ground detectors (Cosmic Explorer and the Einstein Telescope) and LISA will expand the observable volume by orders of magnitude.

Time-domain astronomy and the virtual observatory

Modern astronomy is increasingly synoptic, repeatedly scanning the sky to catch transient and variable phenomena. The Zwicky Transient Facility (ZTF) and, at far greater scale, the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) generate millions of alerts per night, routed through automated brokers that classify candidates and trigger follow-up. This time-domain flood demands scalable pipelines for image subtraction, machine-learning classification, and cross-matching across catalogues. The Virtual Observatory framework and federated data archives let researchers query multiwavelength and multi-messenger data uniformly, turning a constellation of instruments into a single distributed observatory.

Connections Master

Connections to general relativity and quantum optics

Gravitational-wave astronomy is applied general relativity: the waveforms encode the strong-field, dynamical behaviour of spacetime, and detections test relativity in regimes inaccessible to laboratory or solar-system experiments. The squeezing used by LIGO is pure quantum optics, engineering the quantum state of light to reduce measurement uncertainty. Both show how astrophysical instrumentation becomes a laboratory for fundamental physics, with implications for tests of gravity and for quantum measurement theory.

Connections to signal processing and statistics

Matched filtering, the workhorse of gravitational-wave detection, is identical in form to the matched filters used in radar, sonar, and digital communications. Template-bank placement, Bayesian model selection, and false-alarm rate control draw directly from statistical signal processing. Time-domain astronomy likewise relies on change detection, autoregressive modelling, and anomaly detection that overlap with financial and industrial time-series analysis.

Connections to engineering and precision metrology

Space telescopes and gravitational-wave detectors are exercises in extreme precision engineering. JWST's deployment involved hundreds of single-point failure modes and required vibration isolation, cryogenic mechanisms, and wavefront sensing. LIGO suspends 40 kg mirrors as multi-stage pendulums in ultra-high vacuum and measures displacements smaller than an atomic nucleus. These instruments drive advances in optics, materials, vibration control, and laser stabilisation that feed back into manufacturing, navigation, and geodesy.

Connections to cosmology and the early universe

Multiwavelength and multi-messenger data anchor cosmology. The cosmic microwave background, mapped by COBE, WMAP, and Planck from space, fixes the composition and age of the universe. JWST's infrared reach probes the first galaxies during cosmic dawn. Gravitational-wave standard sirens, sources whose luminosity distance is measured directly from the waveform, offer an independent route to the Hubble constant, potentially resolving the tension between early- and late-universe measurements.

Historical and philosophical context Master

From light to multiwavelength astronomy

For most of history, astronomy meant visible light. The opening of the radio window after the Second World War, when Karl Jansky's serendipitous detection of cosmic radio noise was followed by Martin Ryle and others building interferometers, revealed a violent universe invisible to the eye. X-ray astronomy began with rocket flights in the 1960s; the Uhuru satellite in 1970 discovered X-ray binaries and the first stellar-mass black-hole candidate, Cygnus X-1. A succession of Great Observatories, Hubble, Compton, Chandra, and Spitzer, then covered the spectrum from gamma-ray to infrared, complemented by European missions and by all-sky surveyors such as IRAS, ROSAT, and Fermi. JWST continues this lineage, and each new band has overturned prior assumptions about what the universe contains.

The long road to gravitational waves

Einstein predicted gravitational waves in 1916 but doubted they would ever be detected. Joseph Weber's resonant-bar experiments of the 1960s claimed detections that could not be reproduced, seeding decades of scepticism. Rainer Weiss proposed laser interferometry in 1972, and Kip Thorne and Ronald Drever shaped the concept into LIGO, funded in 1992 and built by a collaboration of over a thousand scientists. After an advanced upgrade, LIGO detected GW150914 on 14 September 2015, during engineering runs before the first observing run had even begun. Weiss, Thorne, and Barry Barish received the 2017 Nobel Prize in Physics, and later that year GW170817 inaugurated multi-messenger astronomy with gravitational waves.

New windows, new questions

Each new window rewrites the map of the unknown. Gravitational waves reveal the dark side of the universe, black holes and the mergers of compact objects, almost invisible to light. Multi-messenger events tie those signals to luminous counterparts, turning transient detections into rich physical experiments. The broader lesson of multiwavelength and gravitational-wave astronomy is that the universe is mostly invisible to unaided human senses, and that building new instruments is itself a form of discovery: what we can perceive shapes what we think exists.

Bibliography Master

1. Carroll, B.W. & Ostlie, D.A. (2017). An Introduction to Modern Astrophysics (2nd ed.). Pearson. Ch. 6, telescopes; Ch. 7, detectors and instruments.

2. Harwit, M. (1987). Astronomy: From the Earth to the Universe. Cambridge University Press. Multiwavelength and space astronomy.

3. Abbott, B.P. et al. (LIGO Scientific Collaboration & Virgo Collaboration) (2016). "Observation of gravitational waves from a binary black hole merger." Physical Review Letters 116, 061102.

4. Abbott, B.P. et al. (2017). "Multi-messenger observations of a binary neutron star merger." The Astrophysical Journal Letters 848, L12.

5. Gardner, J.P. et al. (2006). "The James Webb Space Telescope." Space Science Reviews 123, 485-606.

6. Event Horizon Telescope Collaboration (2019). "First M87 Event Horizon Telescope results. I. The shadow of the supermassive black hole." The Astrophysical Journal Letters 875, L1.

7. Rieke, G.H. (2003). Detection of Light: From the Ultraviolet to the Submillimeter (2nd ed.). Cambridge University Press.

8. Maggiore, M. (2008). Gravitational Waves, Vol. 1: Theory and Experiments. Oxford University Press.

9. Creighton, J.D.E. & Anderson, W.G. (2011). Gravitational-Wave Physics and Astronomy: An Introduction to Theory, Experiment and Data Analysis. Wiley.

10. Kellermann, K.I. & Moran, J.M. (2001). "The development of high-resolution imaging in radio astronomy." Annual Review of Astronomy and Astrophysics 39, 457-501.

11. Saulson, P.R. (2017). Fundamentals of Interferometric Gravitational Wave Detectors (2nd ed.). World Scientific.