Gravitational Lensing: A Natural Cosmic Telescope
Linas JuozėnasShare
Knowledge Ark · Universe · Chapter 10 / Article 06
A telescope made of gravity.
A distant galaxy can appear stretched into an arc, repeated across the sky, or magnified into view. The foreground mass that changes its appearance also leaves clues about the universe we cannot see directly.
What if the view itself is the evidence?
An unusual arc in a telescope image may be a familiar kind of galaxy seen through an unfamiliar route. Before reaching us, its light has crossed a region where gravity changes the relationship between an object’s position and its apparent place on the sky.
Understanding that relationship opens two investigations at once: what is the distant source like, and what lies between it and us? Gravitational lensing connects the two.
How can one object appear twice?
In general relativity, matter and energy shape spacetime, and freely traveling light follows paths called null geodesics. Locally, light still follows the straightest available route. Across a curved region, however, that route need not look straight in a diagram connecting the source and observer.[2]
A foreground galaxy can allow light from one background source to reach us along different paths. We assign each arriving ray a direction on the sky, so the source can appear at several positions. These are different images of the same object.[3]
An Einstein ring is a special alignment: the source lies almost directly behind a suitably symmetric lens. Less exact alignment or a more complicated mass distribution produces other patterns, including partial arcs and separate images.[4]
Three ways the effect becomes visible
The labels strong lensing, weak lensing, and microlensing describe related observational situations. Strong lenses produce conspicuous image structures; weak distortions emerge statistically; microlensing is often recognized through changing brightness when the individual images are too close together to resolve.[4]
The categories can overlap. A galaxy cluster may produce strong arcs near its center while weakly distorting a much wider field of background galaxies. Together, those observations help constrain its mass across different scales.[5]
What does a natural telescope actually improve?
Lensing can increase an object’s apparent area and total received light. The stretching is often stronger in one direction, so it can reveal finer source detail along that direction. A quoted magnification is therefore not a uniform improvement in resolution across the image.[3]
Magnification preserves surface brightness. For a transparent lens in the usual geometric-optics treatment, lensing does not increase the light received per unit apparent area from the same source. The larger image can deliver more total flux. This comparison assumes the same observing band and source redshift; cosmological dimming still applies.[2]
Hubble’s cluster telescopes
The Frontier Fields program observed six massive galaxy clusters and adjacent deep fields. Cluster magnification brought faint background galaxies into reach, while the neighboring fields supplied deep views without the same cluster-lens arrangement. Together, they expanded the study of distant galaxy populations.[1]
Webb and the Cosmic Gems
A 2024 study used JWST to resolve five compact star clusters in the highly magnified Cosmic Gems arc. The galaxy’s estimated redshift, about 10.2, came from photometry in that study. Reconstructing the clusters’ intrinsic sizes and brightnesses required models of the foreground lens.[6]
The magnification itself must be estimated. Different cluster models can agree on a broad mass profile while predicting different magnifications at a particular image position. That uncertainty carries into claims about how faint, small, or luminous the original source is.[7]
Reconstructing the lens and the source
A lens model must explain the observed image positions and shapes with one consistent arrangement of foreground mass and background light. The familiar equations below describe an idealized starting point; real galaxies and clusters require extended mass distributions.
The lens equation and Einstein angle
In the thin-lens approximation, the angular mapping is β = θ − (Dls/Ds)α̂(θ). Here β is the unlensed source direction, θ an image direction, and α̂ the physical deflection. These angles are two-dimensional vectors on the sky.[8]
For a point mass in the weak-field approximation, α̂ ≈ 4GM/(bc²), where b is the impact parameter, G the gravitational constant, M the mass, and c the speed of light. The point-lens Einstein angle is θE = √[(4GM/c²)Dls/(DlDs)], in radians. “Strong lensing” can still occur in this weak gravitational field: the name concerns multiple imaging.[8]
Dl, Ds, and Dls are angular-diameter distances from observer to lens, observer to source, and lens to source. In an expanding universe, Dls is generally not Ds − Dl; it is calculated using the adopted cosmology.[9]
An excellent fit does not always identify a unique mass distribution. The mass-sheet degeneracy illustrates the problem: for a single source plane, a coordinated change in the lens mass and reconstructed source scale can preserve image positions and shapes while changing the inferred magnification and time-delay distance. Measurements such as stellar motions provide additional constraints.[10]
Reading the faint pattern of weak lensing
Most background galaxies do not turn into dramatic arcs. Their shapes change subtly, and their original shapes are unknown. By combining many galaxies, astronomers estimate a coherent distortion pattern called shear. The statistical measurement depends on reduced shear, g = γ/(1 − κ), where γ describes shear and κ describes lensing convergence. In the sufficiently weak limit, g is approximately γ.[11]
Around a cluster, the average tangential distortion constrains its projected mass profile. Unlike some methods based on gas temperature or galaxy motions, lensing does not require the cluster to be in equilibrium. It still requires calibrated measurements, source distances, and attention to matter projected along the line of sight.[5]
Across larger areas, cosmic shear probes foreground structure statistically. Grouping background galaxies by redshift adds depth information: farther sources sample longer paths through the universe. This tomography uses broad, overlapping sensitivities to foreground matter, rather than assigning an exact three-dimensional position to every mass element.[12]
What can imitate the signal?
Telescope optics and atmospheric blurring alter shapes; overlapping galaxy images complicate measurement; uncertain redshifts change the inferred geometry. Galaxies can also align naturally with their surroundings, producing intrinsic alignments. Precision analyses model these effects, along with changes to small-scale matter structure caused by gas and stellar feedback. More galaxies reduce random noise but do not automatically remove a shared calibration error.[13]
The 2025 KiDS-Legacy analysis illustrates the importance of this work. Improved redshift calibration, image processing, and modeling helped produce baseline cosmological constraints compatible with Planck. Lensing and the cosmic microwave background provide a valuable comparison; they do not exhibit one universal disagreement across all analyses.[14]
Finding planets through microlensing
When a compact foreground object passes close to our line of sight to a background star, its changing alignment can produce a microlensing light curve. The images usually remain unresolved, but their summed brightness changes. An isolated point lens and simple relative motion give an approximately symmetric rise and fall; companions, finite source size, and other effects complicate that ideal shape.[15]
A planet orbiting the lens star can introduce a brief additional feature. Astronomers compare models to test whether a planetary companion explains it. The inferred planet-to-star mass ratio and projected separation are useful clues, but they are not automatically an absolute planet mass or a complete orbit. Dense observations are essential because competing configurations can produce similar features.[16]
Duration is not a mass measurement
A long event can reflect a massive lens, slow relative motion, or a particular distance arrangement. Event duration alone cannot distinguish them. The same reasoning prevents a short event from automatically establishing an isolated planet.[15]
The apparent position moves too
The combined light of unresolved images can shift its apparent center. This astrometric microlensing, combined with brightness measurements and parallax information, helps constrain lens mass and distance. Nearby blended stars must be accounted for because they can also shift the measured light centroid.[17]
What lensing tells us about unseen matter
Lensing responds to the gravity of total matter: stars, gas, and any dark component. A colored mass map is a reconstruction from those gravitational effects. Comparing it with independent measurements of visible matter is what allows astronomers to infer a dark contribution.[5]
The Bullet Cluster provides a striking example. After two clusters collided, their hot, X-ray-emitting gas was displaced from the main galaxy concentrations. Lensing reconstructed mass peaks approximately following the galaxies rather than the gas, which contains most of the observed ordinary matter. Within general relativity, this separation provides strong evidence for additional unseen mass.[18]
It does not identify a dark matter particle or establish that dark matter can never interact with itself. Limits on such interactions depend on merger simulations and on comparing mass and galaxy positions consistently with the observations.[19]
Smaller concentrations of mass can also perturb strongly lensed images. Their abundance and properties can test dark matter models, including the distribution of small halos. Interpreting those perturbations requires careful treatment of the main lens, the background source, ordinary matter, and additional structures along the light path.[20]
Using a lens to measure cosmic distances
If a multiply imaged quasar changes brightness, the same variation can appear in its images at different times. The paths differ both in geometry and in gravitational delay. Monitoring the repeated pattern measures these time delays; a lens model then connects them to an absolute distance scale.[21]
Δt = (DΔt/c)Δφ
Here Δφ is the dimensionless difference in the modeled Fermat potential, combining path geometry and gravitational delay. The time-delay distance is DΔt = (1 + zl)DlDs/Dls, with lens redshift zl. For fixed remaining cosmological parameters, this distance varies inversely with H₀.[21]
The method supplies an expansion measurement with different assumptions from the local distance ladder. Its interpretation still depends on the lens mass profile, stellar motions, and foreground structure. The TDCOSMO collaboration’s 2025 analysis combined eight time-delay quasars with updated observations and flexible modeling of these ingredients. Its results are conditional inferences, rather than a single definitive value that all gravitational lenses independently deliver.[22]
Precise timing is one part of the measurement. A good account of the lens mass is also needed before a measured delay becomes a precise estimate of cosmic expansion.[21]
More images—and better ways to interpret them
Rubin: a repeatedly observed sky
The Vera C. Rubin Observatory began its Legacy Survey of Space and Time in June 2026. Its repeated multicolor imaging brings changing sources and large galaxy populations into the same observing program. LSST is the survey’s name; Rubin is the observatory conducting it.[23]
Euclid: mapping an operating survey
Euclid has conducted routine science since February 2024. Its visible-light imaging measures galaxy shapes, while near-infrared observations add photometry and spectroscopy. These complementary measurements support studies of matter structure and cosmic expansion.[24]
Roman: preparing complementary lensing surveys
NASA’s September 2026 update places the Nancy Grace Roman Space Telescope in commissioning during its journey toward Sun–Earth L2.[25] Its planned observations include wide-field galaxy imaging for weak lensing and repeated monitoring of stars toward the Galactic bulge for microlensing. Those programs use the same physical effect to investigate very different subjects: cosmic structure and planetary systems.[26]
As samples grow, the opportunity is to connect observations more tightly: shapes with redshifts, mass maps with gas and stars, and brightness changes with detailed lens models. The quality of those connections determines how much physical information each new lens provides.
Research and mission status checked: 8 September 2026.
Sources and further reading
Original research, scientific reviews, and official survey information. Checked in September 2026. The original illustrations explain lensing geometry and observational signatures; they are not telescope images or measured data.
- NASA / ESA / STScI — Hubble Frontier Fields and Parallel FieldsDescribes the six cluster fields and neighboring deep fields observed by the Frontier Fields program.
- Perlick (2004; arXiv 2010), Gravitational Lensing from a Spacetime PerspectiveLightlike geodesics and the precise redshift dependence of surface brightness.
- Treu (2010), Strong Lensing by GalaxiesMagnification improves access to distant sources while source reconstruction and selection require a lens model.
- Wambsganss (1998), Gravitational Lensing in AstronomyStrong images, weak statistical distortions, and unresolved microlensing describe related observational regimes.
- Umetsu (2020), Cluster–Galaxy Weak LensingCluster lensing probes projected total matter without requiring dynamical or hydrostatic equilibrium.
- Adamo et al. (2024) — Bound Star Clusters Observed in a Lensed Galaxy 460 Myr After the Big BangWebb resolves compact star clusters in the Cosmic Gems arc, with physical properties inferred through lens modeling.
- Raney et al. (2020) — Systematic Versus Statistical Uncertainties in Frontier Fields Masses and MagnificationsShows that different mass models can agree on broad profiles while predicting different local magnifications.
- Narayan & Bartelmann (1996), Lectures on Gravitational LensingPoint-mass deflection, the thin-lens equation, and Einstein angles.
- Hogg (1999, revised 2000), Distance Measures in CosmologyAngular-diameter distances used in the lens equation are not simply additive.
- Schneider & Sluse (2013), Mass-Sheet Degeneracy, Power-Law Models and External ConvergenceDifferent lens mass profiles can reproduce the same image geometry while changing magnification and inferred distances.
- Bartelmann & Schneider (2001), Weak Gravitational LensingGalaxy shapes constrain reduced shear; convergence and shear jointly determine magnification.
- Hu (1999), Power Spectrum Tomography with Weak LensingSource-redshift bins add depth information through overlapping lensing kernels.
- Mandelbaum (2018), Weak Lensing for Precision CosmologyShape calibration, point-spread functions, redshift distributions, blends, intrinsic alignments, and baryonic effects.
- Wright et al. (2025) — KiDS-Legacy Cosmic-Shear ConstraintsA completed weak-lensing survey illustrates the importance of redshift calibration, intrinsic alignments, and gas-feedback modeling.
- Mao (2012) — Astrophysical Applications of Gravitational MicrolensingUnresolved multiple images produce changing total brightness; event duration depends on mass, distances, and relative motion.
- Tsapras (2018) — Microlensing Searches for ExoplanetsPlanets perturb microlensing images, leaving brief light-curve features that require careful modeling and dense observations.
- Sahu et al. (2022) — Astrometric Microlensing by a Dark Stellar RemnantThe changing light centroid provides an additional way to constrain a microlens’s angular scale and mass.
- Clowe et al. (2006) — A Direct Empirical Proof of the Existence of Dark MatterCompares Bullet Cluster lensing peaks with the galaxies and displaced X-ray-emitting gas.
- Robertson, Massey & Eke (2017) — What Does the Bullet Cluster Tell Us About Self-Interacting Dark Matter?Shows how merger modeling and measurement choices affect limits on dark matter self-interactions.
- Vegetti et al. (2024) — Strong Gravitational Lensing as a Probe of Dark MatterExplains sensitivity to small mass concentrations and the modeling needed to distinguish their effects.
- Birrer et al. (2024) — Time-Delay CosmographyRepeated brightness variations, arrival-time delays, and lens models together constrain absolute cosmological distances.
- TDCOSMO Collaboration (2025) — Cosmological Constraints from Strong-Lensing Time DelaysModern time-delay analyses combine lens images, stellar motions, and foreground structure while allowing flexible mass profiles.
- Rubin Observatory (2026) — The Legacy Survey of Space and Time Is UnderwayRubin began its ten-year imaging survey in June 2026, repeatedly recording the southern sky.
- ESA (2026) — Euclid’s Operating SurveyEuclid has conducted routine science since February 2024, mapping galaxies to study cosmic structure and expansion.
- NASA (updated 7 September 2026) — Roman commissioningRoman is undergoing instrument checkout and calibration while traveling toward Sun–Earth L2.
- NASA — Roman’s Weak-Lensing and Microlensing ScienceRoman is designed to combine wide-field weak-lensing measurements with repeated monitoring for Galactic microlensing events.
Cosmology and the Universe’s Large-Scale Structure
- Cosmic Inflation: Theory and Evidence
- The Cosmic Web: Filaments, Voids, and Superclusters
- The Cosmic Microwave Background’s Detailed Structure
- Baryon Acoustic Oscillations
- Redshift Surveys and Mapping the Universe
- Gravitational Lensing: A Natural Cosmic Telescope · You are here
- Measuring the Hubble Constant: The Tension
- Dark Energy Surveys
- Anisotropies and Inhomogeneities
- Current Debates and Outstanding Questions