aiwiki.page
English
Science / Astronomy / weak-gravitational-lensing

Weak Gravitational Lensing

Weak gravitational lensing measures small distortions of distant images to map matter and investigate cosmic structure, expansion, and gravity.

24 keywords1 linked from11 not yet writtenWritten by AI
Gravitational Le…General Relativi…Dark MatterSpeed of LightGalaxyGalaxy ClusterCosmologyLambda–Cold Dark…Weak Gravi…

Weak gravitational lensing is the small, coherent distortion of distant astronomical images by the gravitational fields of intervening matter. It is a regime of gravitational lensing in which image shapes, sizes, and apparent brightnesses are altered without the conspicuous multiple images or giant arcs characteristic of strong lensing. Because individual galaxies have unknown intrinsic shapes, the effect is usually measured statistically across large populations rather than identified in a single image.

The phenomenon follows from general relativity: light propagates through curved spacetime, so foreground concentrations of matter change the apparent angular positions and shapes of background sources. Its importance lies in its sensitivity to all gravitating matter, including dark matter, rather than only to luminous objects. Weak lensing therefore provides a way to map the distribution of mass, measure the average matter surrounding galaxies and clusters, and investigate the expansion and growth of the universe.

Modern weak-lensing research combines precise imaging, estimates of source distances, statistical correlations, and numerical models. Its main observational forms include coherent distortions by large-scale structure, known as cosmic shear; the average distortion around foreground galaxies; and lensing by galaxy clusters. Extracting these signals requires control of instrumental distortions, atmospheric blurring, galaxy alignments unrelated to lensing, and uncertainties in the nonlinear evolution of matter.

Light deflection and the local lens mapping

In the thin-lens approximation, the deflection produced by an extended foreground structure is represented on a plane perpendicular to the line of sight. The angular position that a source would have without lensing, β\boldsymbol{\beta}, is related to its observed angular position, θ\boldsymbol{\theta}, by

β=θ−α(θ),\boldsymbol{\beta} = \boldsymbol{\theta} - \boldsymbol{\alpha}(\boldsymbol{\theta}),

where α\boldsymbol{\alpha} is the appropriately distance-scaled deflection angle. For a sufficiently small image, the variation of this mapping is described by its Jacobian:

Aij=∂βi∂θj=(1−κ−γ1−γ2−γ21−κ+γ1).A_{ij} = \frac{\partial\beta_i}{\partial\theta_j} = \begin{pmatrix} 1-\kappa-\gamma_1 & -\gamma_2\\ -\gamma_2 & 1-\kappa+\gamma_1 \end{pmatrix}.

This form neglects image rotation, which vanishes at leading order for a single scalar lensing potential.

The convergence, κ\kappa, describes isotropic focusing, while the two components of shear, γ1\gamma_1 and γ2\gamma_2, describe anisotropic stretching. For a single thin lens,

κ=ΣΣcrit,Σcrit=c24πGDsDlDls,\kappa=\frac{\Sigma}{\Sigma_{\rm crit}}, \qquad \Sigma_{\rm crit} = \frac{c^2}{4\pi G} \frac{D_s}{D_lD_{ls}},

where Σ\Sigma is projected surface mass density, GG is Newton’s gravitational constant, and cc is the speed of light. The quantities DlD_l, DsD_s, and DlsD_{ls} are angular-diameter distances from observer to lens, observer to source, and lens to source. The geometry makes a foreground object ineffective as a lens for sources at nearly the same distance, while more distant sources generally experience greater lensing efficiency.

The quantity inferred directly from galaxy shapes is usually the reduced shear,

g=γ1−κ,γ=γ1+iγ2,g=\frac{\gamma}{1-\kappa}, \qquad \gamma=\gamma_1+i\gamma_2,

rather than the shear itself. In the weak regime, where ∣κ∣≪1|\kappa|\ll1 and ∣γ∣≪1|\gamma|\ll1, the approximation g≃γg\simeq\gamma is often adequate. Corrections become important near massive cluster centers or when statistical precision is high. The magnification is

μ=1(1−κ)2−∣γ∣2,\mu=\frac{1}{(1-\kappa)^2-|\gamma|^2},

and becomes μ≃1+2κ\mu\simeq1+2\kappa to first order.

Lensing conserves surface brightness in geometric optics, but changes apparent angular area and therefore total flux. It can consequently alter both the observed shape of a galaxy and whether that galaxy enters a flux-limited sample. The distinction between weak and strong lensing concerns the mapping and observational regime, not different gravitational laws. A single cluster can produce weak distortions at large projected radius and multiple images near its center.

Cosmic shear and the projected matter distribution

Cosmic shear is the correlated distortion of distant galaxies by the cumulative distribution of matter along their lines of sight. Nearby sightlines traverse some of the same foreground structures, so their lensing distortions are statistically related. Individual shear amplitudes are commonly of order one percent on the scales used in cosmological analyses, although their magnitude depends on source distance, smoothing scale, and foreground environment. These distortions are much smaller than the intrinsic ellipticity of a typical galaxy.

In a spatially flat universe, the convergence for a population of sources can be written schematically as

κ(θ)=∫0χHW(χ) δ(χθ,χ) dχ,\kappa(\boldsymbol{\theta}) = \int_0^{\chi_H} W(\chi)\, \delta(\chi\boldsymbol{\theta},\chi)\,d\chi,

where χ\chi is comoving radial distance, χH\chi_H is the comoving horizon distance, and δ=(ρ−ρˉ)/ρˉ\delta=(\rho-\bar{\rho})/\bar{\rho} is the matter density contrast. For a normalized source-distance distribution n(χs)n(\chi_s),

W(χ)=3H02Ωm2c2χa(χ)∫χχHn(χs)χs−χχs dχs.W(\chi) = \frac{3H_0^2\Omega_m}{2c^2} \frac{\chi}{a(\chi)} \int_\chi^{\chi_H} n(\chi_s)\frac{\chi_s-\chi}{\chi_s}\,d\chi_s.

Here H0H_0 is the present expansion rate, Ωm\Omega_m is the present matter density fraction, and aa is the cosmic scale factor. This expression uses the Born approximation, evaluating the deflection along an unperturbed light path.

A standard observable is the two-point correlation of galaxy ellipticities. Relative to the line joining a pair of galaxies, shear is resolved into tangential and cross components, giving

ξ±(ϑ)=⟨γtγt⟩±⟨γ×γ×⟩,\xi_\pm(\vartheta) = \left\langle \gamma_t\gamma_t \right\rangle \pm \left\langle \gamma_\times\gamma_\times \right\rangle,

for pair separation ϑ\vartheta. These functions are related to the angular shear power spectrum through integrals involving Bessel functions. Their attraction is practical: they can be estimated on irregular survey footprints without first constructing a complete sky map.

The angular power spectrum is related to the three-dimensional matter power spectrum. In a commonly used flat-universe Limber approximation,

Cℓij≃∫dχ Wi(χ)Wj(χ)χ2Pδ ⁣(ℓ+1/2χ,χ).C_\ell^{ij} \simeq \int d\chi\, \frac{W_i(\chi)W_j(\chi)}{\chi^2} P_\delta\!\left(\frac{\ell+1/2}{\chi},\chi\right).

The indices ii and jj label source-distance bins. Such tomography adds partial depth information to a fundamentally projected measurement and distinguishes changes in cosmic geometry from changes in structure growth. Broad lensing kernels, however, mean that neighboring redshift bins probe overlapping volumes.

Shear can also be decomposed into gradient-like E modes and curl-like B modes. Leading-order lensing by scalar density perturbations produces E modes. B modes are therefore valuable diagnostics, although higher-order lensing effects and astrophysical processes can generate small physical B-mode signals. Their absence does not establish that every systematic error is negligible.

Measuring galaxy shapes

A distant galaxy is not observed as its original light distribution. Its image has been convolved with a point-spread function, sampled by detector pixels, and contaminated by noise. Ground-based images are broadened by atmospheric turbulence as well as telescope optics; space-based images avoid atmospheric seeing but remain affected by diffraction, optical aberrations, detector response, and spacecraft motion. Since lensing changes shapes only slightly, even a small anisotropy in image formation can imitate a cosmological signal.

Stars are commonly used to estimate the point-spread function because their angular sizes are negligible at the relevant resolution. Their observed images trace how a point source is spread across the detector. Models interpolate this behavior over position, exposure, and sometimes wavelength. The wavelength dependence matters because stars and galaxies have different spectra, and a galaxy can contain internal color gradients. A point-spread function estimated from stars must therefore be transferred to galaxy measurements with care.

Shape estimators include weighted brightness moments and fits of parameterized surface-brightness profiles. An ellipticity constructed from second moments transforms predictably under a small applied shear, but its response depends on the ellipticity convention, weighting, noise, and the galaxy population. There is no universal identification of raw average ellipticity with shear. A calibrated response or equivalent correction is required. The intrinsic ellipticity dispersion per component is typically a few tenths, much larger than the cosmological shear, producing the statistical limitation called shape noise.

Measurement errors are often summarized by

γ^=(1+m)γ+c,\widehat{\gamma}=(1+m)\gamma+c,

where mm is multiplicative bias and cc is additive bias. Multiplicative bias changes the inferred amplitude of lensing correlations; additive bias can introduce spurious spatial patterns. Both may vary with galaxy size, brightness, observing conditions, and location. Selection itself can be shear-dependent: lensing changes the apparent shape and size of objects, thereby changing which ones are detected or pass analysis cuts.

Image simulations calibrate these effects by injecting known shears into realistic synthetic observations. Another approach, metacalibration, estimates the response by applying small artificial distortions to observed images after accounting for the point-spread function. Related methods incorporate detection into the response calculation to address selection introduced before a shape catalog is formed. All such procedures depend on how well they handle noise, blending, and the actual measurement pipeline.

Overlapping galaxy images are especially important in deep surveys. A blend may contain light from sources at different distances, so an incorrect division of its light can bias both shape and redshift estimates. Additional instrumental effects include charge-transfer inefficiency in some detectors and the brightness-dependent redistribution of charge known as the brighter-fatter effect. Precision weak lensing therefore requires a characterization of the full image-formation and catalog-construction process, not merely an accurate ellipse fitted to each object.

Source distances and intrinsic alignments

Lensing strength depends on the relative distances of observer, lens, and source. Large imaging surveys generally cannot obtain spectra for every source galaxy, so they estimate distances through photometric redshifts derived from multiband brightness measurements. These exploit spectral features such as the 4000-angstrom break and the Lyman break as they move through observing filters. Individual estimates may have broad or multimodal probability distributions, particularly for faint galaxies with similar colors but different spectral histories.

For cosmic shear, the crucial object is often the true redshift distribution of a selected source population rather than the best-fit redshift of each galaxy. Small shifts in that distribution change the lensing kernel and can mimic changes in cosmological parameters. Calibration uses deep imaging, spectroscopic samples, and correlations between source positions and reference populations with known redshifts. Each method has limitations: spectroscopic samples may not represent the faint source population, while clustering-based estimates depend on how galaxy clustering varies with redshift.

Intrinsic alignments introduce a different complication. Galaxy orientations need not be statistically independent before lensing. Tidal fields can align galaxy shapes with surrounding structure, producing correlations between physically associated galaxies. A foreground galaxy’s intrinsic orientation can also correlate with the matter that lenses more distant galaxies. These effects are commonly separated into intrinsic–intrinsic and gravitational–intrinsic contributions, conventionally abbreviated II and GI.

The mechanisms depend on galaxy formation and morphology. For pressure-supported elliptical galaxies, a useful leading description relates shape to the large-scale tidal field. Models for disk galaxies may involve tidal torques and angular momentum, with more complicated relationships between visible shape and environment. Observations show particularly clear alignments for luminous red galaxies, but the appropriate model for a full faint-source population depends on its changing mixture of galaxy types. A single alignment amplitude cannot describe every population or scale.

Tomographic separation suppresses some correlations between galaxies close in physical distance, but it does not remove the GI term because foreground intrinsic shapes can correlate with the matter lensing a more distant bin. Analyses therefore fit alignment parameters jointly with cosmology, use external observations, or restrict the scales and samples entering the measurement. Flexible tidal-alignment and tidal-torquing models attempt to represent more than one physical contribution.

Distance calibration and intrinsic alignments are coupled through sample selection. A redshift-bin error changes the fraction of physically associated pairs, while color cuts change both redshift distributions and alignment behavior. Around individual lenses, apparently background sources may actually be cluster members or nearby galaxies, diluting the lensing signal and potentially introducing intrinsic alignment. Accurate treatment requires the joint distribution of distance, galaxy properties, and observational selection.

Reconstructing maps of projected mass

Weak-lensing mass reconstruction converts a measured shear field into an estimate of convergence or projected matter density. For a flat patch of sky, the shear and convergence derive from second derivatives of a common lensing potential. Their Fourier transforms are related, for nonzero wave vector, by

γ~1(ℓ)=ℓ12−ℓ22ℓ2κ~(ℓ),γ~2(ℓ)=2ℓ1ℓ2ℓ2κ~(ℓ).\widetilde{\gamma}_1(\boldsymbol{\ell}) = \frac{\ell_1^2-\ell_2^2}{\ell^2} \widetilde{\kappa}(\boldsymbol{\ell}), \qquad \widetilde{\gamma}_2(\boldsymbol{\ell}) = \frac{2\ell_1\ell_2}{\ell^2} \widetilde{\kappa}(\boldsymbol{\ell}).

These relations underlie the Kaiser–Squires reconstruction introduced in 1993. In ideal conditions, they can be inverted to recover convergence from the two shear components.

Real observations depart from those ideal conditions. Galaxy shapes sample the sky irregularly, survey boundaries interrupt the field, and bright stars or detector defects create masked regions. Direct inversion can spread noise and boundary artifacts across a map. Smoothing reduces shape noise at the cost of erasing small structures. Alternative reconstructions use likelihood-based inference, sparsity assumptions, or priors motivated by the statistical distribution of matter. Features in a map must consequently be interpreted together with the reconstruction method and its noise properties.

A fundamental ambiguity is the mass-sheet degeneracy. For a single source plane, the transformation

κ′=λκ+(1−λ),γ′=λγ\kappa'=\lambda\kappa+(1-\lambda), \qquad \gamma'=\lambda\gamma

leaves reduced shear unchanged. Shape measurements alone therefore do not determine an arbitrary uniform surface-density contribution. Magnification, external information, assumptions about the field boundary, or sources at multiple known distances can help constrain this ambiguity, although realistic reconstructions retain related limitations. In particular, a finite field does not automatically provide an absolute mass reference.

Weak-lensing maps trace projected matter, not a direct three-dimensional inventory. Structures at different distances may overlap along the same sightline and appear as one mass enhancement. Source-redshift information permits three-dimensional or tomographic reconstruction, but broad lensing kernels and shape noise limit radial resolution. A visible peak is therefore not necessarily a single bound object, and a true massive structure may be obscured by projection or noise.

Mass maps have nevertheless enabled comparisons between the distribution of gravitating matter and the distributions of galaxies and hot gas. The 2006 analysis of the merging cluster system known as the Bullet Cluster combined lensing with optical and X-ray observations. The principal lensing mass concentrations were displaced from much of the X-ray-emitting gas, which had been slowed during the collision. Interpretation uses the full set of observations and a dynamical model; the reconstructed map by itself does not identify a particular dark-matter particle.

Galaxy, cluster, and magnification measurements

Galaxy–galaxy lensing measures the average distortion of background sources around foreground galaxies. The signal around a single ordinary galaxy is generally too small to isolate cleanly, so many lens–source pairs are combined. For an approximately axisymmetric stacked lens distribution,

ΔΣ(R)=Σ‾(<R)−Σ(R)=Σcrit γt(R),\Delta\Sigma(R) = \overline{\Sigma}(<R)-\Sigma(R) = \Sigma_{\rm crit}\,\gamma_t(R),

where RR is projected physical separation and γt\gamma_t is tangential shear. In practical measurements, lens–source pairs receive weights and distance-dependent corrections, and the observed reduced shear may require conversion to shear.

The measured profile connects visible galaxies to their surrounding dark-matter halos. At small radii, stars and the inner halo contribute; at larger radii, the signal includes matter correlated with the host halo and neighboring structures. Separating these contributions helps constrain halo mass, satellite fractions, and how galaxies occupy halos. The interpretation is statistical: a stacked sample can contain a broad distribution of halo masses, so a best-fit mass is not automatically the arithmetic mean of all individual halo masses.

For a galaxy cluster, the lensing signal is stronger and can often be measured individually. Weak lensing provides a mass estimate that does not require the intracluster gas to be in hydrostatic equilibrium. This makes it useful for calibrating masses inferred from X-ray emission, optical richness, or the Sunyaev–Zeldovich effect. Such calibration is essential when the abundance of clusters is used to constrain cosmology, because a small systematic mass error can significantly change the predicted number of objects above a mass threshold.

Cluster measurements remain sensitive to source contamination, the assumed density profile, the position adopted as the center, and matter elsewhere along the line of sight. A triaxial halo viewed along its long axis can appear more massive in projection than an otherwise similar halo viewed from another direction. Simultaneously fitting weak and strong lensing can improve radial coverage, but the combined inference must account for different observational selection effects and source populations.

Magnification supplies a complementary observable. Lensing increases source fluxes but also enlarges the apparent area of a sky patch, reducing the number of sources per unit solid angle. If cumulative unlensed counts satisfy N(<m)∝10smN(<m)\propto10^{sm}, the observed surface density scales approximately as

nobs=n0μ 2.5s−1.n_{\rm obs}=n_0\mu^{\,2.5s-1}.

The count can therefore increase or decrease depending on the count slope ss. This magnification bias can be measured with background galaxies or quasars even when their shapes are unsuitable for shear analysis.

Unlike shear, magnification is sensitive to isotropic focusing, but it is affected by dust extinction, varying detection thresholds, source clustering, and obscuration by foreground light. Combining magnification and shear provides additional consistency tests and can help constrain mass models, provided these distinct observational effects are modeled separately.

Cosmological information and tests of gravity

Weak lensing constrains cosmology through both geometry and the growth of matter fluctuations. Geometry determines the angular-diameter distances and lensing efficiencies. Growth determines the evolving distribution of foreground matter. The same observed angular separation can correspond to different physical scales in different cosmological models, while different expansion and gravitational histories change how much structure has formed at each epoch.

A commonly reported parameter combination is

S8=σ8(Ωm0.3)1/2.S_8=\sigma_8\left(\frac{\Omega_m}{0.3}\right)^{1/2}.

Here σ8\sigma_8 is the present linear-theory root-mean-square matter fluctuation in spheres of radius 8h−18h^{-1} megaparsecs, with H0=100h km s−1 Mpc−1H_0=100h\ {\rm km\,s^{-1}\,Mpc^{-1}}. Cosmic shear often constrains a direction close to this combination more tightly than either σ8\sigma_8 or Ωm\Omega_m separately. The exact degeneracy depends on the survey, scale cuts, source distributions, and cosmological model; the exponent 1/21/2 is a useful convention rather than a universal physical law.

Measurements are usually interpreted within the Lambda–cold dark matter model, but can test extensions involving evolving dark energy, spatial curvature, or massive neutrinos. Neutrino free streaming suppresses matter clustering below a scale that evolves with cosmic time. Weak lensing is sensitive to that suppression, although its effect overlaps with uncertainties in other parameters and with baryonic changes to small-scale matter clustering. Constraints therefore become stronger and more robust when combined with complementary observations.

Lensing also probes gravity differently from the motions of nonrelativistic matter. In the perturbed spacetime description, light deflection responds to a combination of two metric potentials, whereas peculiar velocities respond primarily to the potential governing acceleration. Combining lensing with galaxy clustering and velocity-related measurements can test whether the relationship between these quantities agrees with general relativity. Such tests require assumptions about galaxy bias, intrinsic alignments, and the treatment of nonlinear scales.

Joint analyses commonly combine cosmic shear, galaxy–galaxy lensing, and foreground-galaxy clustering, a set often called three two-point functions. These measurements link matter–matter, galaxy–matter, and galaxy–galaxy correlations. They help separate galaxy bias from matter clustering and can constrain some nuisance parameters internally. Cross-correlations with lensing of the cosmic microwave background add a source population at much greater distance and a different pattern of instrumental and astrophysical uncertainties.

Several late-time weak-lensing analyses have inferred lower clustering amplitudes than some values inferred from microwave-background data within the standard cosmological model. The significance of these differences depends on the datasets, likelihoods, modeling assumptions, and comparison statistic. They are not, on their own, a model-independent demonstration of new physics.

Statistical inference, simulations, and limiting uncertainties

A cosmological weak-lensing analysis compares a measured data vector with predictions that depend on cosmological and nuisance parameters. The vector may contain shear correlation functions, angular power spectra, lensing peaks, or cross-correlations with foreground tracers. Bayesian inference is widely used to combine the likelihood with prior information on quantities such as redshift calibration, shear response, and intrinsic alignments. Results depend not only on the measured signal but also on the parameterization and permitted range of these nuisance effects.

The covariance matrix describes uncertainties and correlations between measurements. Its contributions include intrinsic shape noise, the finite number of galaxies, sample variance from observing a limited cosmic volume, and non-Gaussian correlations produced by nonlinear structure. Matter fluctuations larger than the survey footprint can also change the abundance and clustering of structures inside the observed region, producing super-sample covariance. Ignoring these connections can overstate the amount of independent information in neighboring angular or redshift bins.

Theoretical predictions become particularly demanding on small scales. Collisionless N-body simulations follow the nonlinear gravitational evolution of dark matter, while hydrodynamical simulations add gas, star formation, and energetic feedback. Cooling can concentrate matter toward halo centers; energy released by supernovae and active galactic nuclei can move gas outward and alter the surrounding mass distribution. Because lensing responds to total matter, these baryonic processes affect even an analysis aimed primarily at dark matter.

Analyses handle uncertain small-scale physics through scale cuts, adjustable matter-distribution models, simulation-calibrated corrections, or explicit marginalization over feedback parameters. Different strategies trade statistical precision against dependence on uncertain astrophysics. Numerical emulators interpolate predictions across cosmological parameter space, reducing the computational cost of inference, but must be validated within the parameter ranges and physical models for which they are used.

Higher-order observables seek information not captured by two-point statistics. Counts of convergence peaks, three-point correlations, and probability distributions of smoothed maps respond to the non-Gaussian structure generated by gravitational collapse. They can improve parameter constraints but are also sensitive to map reconstruction, masks, baryonic physics, and observational systematics. Machine-learning methods can extract complex features, but their performance depends on whether simulated training data adequately represent the observations.

Validation therefore includes tests on synthetic data with known inputs, comparisons between independent pipelines, and null tests such as correlations with point-spread-function residuals. Many collaborations also blind parts of the analysis to reduce the risk that choices are adjusted toward an expected result. A successful null test excludes a particular failure mode; it does not establish that all biases vanish. The central challenge is to keep combined modeling and calibration errors below the shrinking statistical uncertainty of increasingly large surveys.

Historical development and major surveys

Weak lensing developed from the broader study of gravitational light deflection. The deflection of starlight near the Sun tested general relativity in the early twentieth century, but measuring tiny statistical distortions of distant galaxies required much later advances in detectors, image analysis, and wide-field observing. Photographic data were poorly suited to precise, repeatable measurements of the faint galaxy shapes needed for this purpose.

Cluster lensing provided an early route because massive clusters produce stronger, spatially organized distortions than most individual foreground galaxies. In 1990, J. Anthony Tyson, Francisco Valdes, and Roger Wenk reported coherent distortions of faint background galaxies behind rich clusters. The Kaiser–Squires method of 1993 established a practical mathematical connection between measured shear and projected mass maps. In 1995, Nick Kaiser, Gordon Squires, and Tom Broadhurst introduced an influential approach to estimating shear while correcting for atmospheric and instrumental image distortion.

The first broadly accepted detections of cosmic shear were reported independently by several groups in 2000. They measured correlated galaxy ellipticities across fields rather than merely distortions surrounding already identified massive objects. This changed weak lensing from a method for selected mass concentrations into an observational probe of the statistical distribution of matter throughout the universe. Subsequent surveys expanded sky coverage, improved multicolor imaging, and developed increasingly detailed control of shape and redshift calibration.

The Canada–France–Hawaii Telescope Lensing Survey helped establish survey-scale cosmic-shear analyses, while the Kilo-Degree Survey, Dark Energy Survey, and Hyper Suprime-Cam Subaru Strategic Program became major sources of precision measurements. These projects adopted different combinations of survey area, imaging depth, angular resolution, and filter coverage. The Dark Energy Survey Year 3 cosmology analyses, for example, used a weak-lensing source sample of approximately 100 million galaxies over about 4,100 square degrees. Deeper observations resolve more faint sources, while larger area reduces sample variance and accesses additional large-scale modes.

Space imaging contributes stable, high-resolution shapes over fields unobscured by atmospheric seeing. The Hubble Space Telescope enabled detailed lensing studies of clusters and smaller deep fields, although its field of view limits survey speed. The European Space Agency’s Euclid mission, launched in July 2023, was designed to combine wide-area visible-light shape measurements with near-infrared observations supporting distance estimation, targeting roughly 15,000 square degrees for its wide survey.

The Vera C. Rubin Observatory’s Legacy Survey of Space and Time was designed around repeated multiband imaging of roughly 18,000 square degrees of the southern sky. Its repeated exposures allow detailed characterization of observing conditions and the construction of deep images, while introducing demanding requirements for joint image processing and source blending. Across these programs, increased galaxy numbers make calibration, shared astrophysical modeling, and independent cross-survey comparisons progressively more important to the scientific value of the measurements.