{
  "id": "flrw-metric",
  "title": "Friedmann–Lemaître–Robertson–Walker Metric",
  "lang": "en",
  "source_lang": "en",
  "summary": "The Friedmann–Lemaître–Robertson–Walker metric describes a spatially homogeneous and isotropic universe and provides the geometric framework for modern cosmology.",
  "category": "science",
  "subcategory": "astronomy",
  "type": "concept",
  "aliases": [
    "FLRW metric",
    "Robertson–Walker metric",
    "Friedmann–Robertson–Walker metric",
    "FRW metric"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/flrw-metric/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/flrw-metric/",
      "json": "https://aiwiki.page/zh/flrw-metric.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "The Friedmann–Lemaître–Robertson–Walker metric, usually abbreviated FLRW metric, is the mathematical description of a spacetime whose spatial geometry is homogeneous and isotropic: at each cosmic time, all spatial locations are locally equivalent, and no spatial direction is preferred. It expresses the changing separation of freely moving cosmic observers through a single function, the scale factor. Within general relativity, it supplies the background geometry for models of an expanding or contracting universe, including the standard Lambda–cold dark matter model.\n\nThe metric is a statement about geometry and symmetry, not by itself a law of cosmic evolution. Its scale factor is determined only after gravitational field equations and the universe’s matter content have been specified. This distinction makes the FLRW framework broadly useful: it connects expansion, spatial curvature, light propagation, and observable distances without requiring a particular explanation of dark matter or accelerated expansion. The real universe contains galaxies, clusters, and other irregularities, so the metric is used as a large-scale background rather than as an exact description of every region."
    },
    {
      "heading": "Symmetry assumptions and the line element",
      "text": "The FLRW construction begins with the cosmological principle, understood here as spatial homogeneity and isotropy on hypersurfaces of constant cosmic time. Homogeneity means that the intrinsic geometry of such a hypersurface is the same at every point. Isotropy means that its geometry is invariant under rotations about each point. Together, these conditions require the spatial slices to have constant sectional curvature. They do not require the full four-dimensional spacetime to be homogeneous in time: the universe can have a distinguished age and can evolve from one cosmic epoch to another.\n\nWith metric signature \\((-+++)\\), a standard expression for the line element is\n\n\\[ds^2=-c^2dt^2+a^2(t) \\left[ \\frac{dr^2}{1-kr^2} +r^2\\left(d\\theta^2+\\sin^2\\theta\\,d\\phi^2\\right) \\right].\\]\n\nHere \\(c\\) is the speed of light, \\(t\\) is cosmic time, \\(a(t)\\) is the scale factor, and \\(r,\\theta,\\phi\\) are comoving spatial coordinates. In this convention, \\(r\\) is dimensionless, \\(a\\) has dimensions of length, and the normalized curvature index takes the values \\(k=+1,0,-1\\). An alternative convention makes the scale factor dimensionless and puts the length scale into the spatial coordinates or curvature parameter. Both conventions describe the same geometry, but their quantities must not be mixed within an equation.\n\nCosmic time is the proper time measured by observers whose comoving spatial coordinates remain fixed. These observers define the fundamental cosmic reference frame. In an ideal FLRW universe, they see the cosmic matter distribution as isotropic. The time coordinate is therefore physically associated with a preferred family of observers, rather than being an arbitrary clock shared by all possible observers. An observer moving relative to this family generally sees directional effects, such as a dipole in the cosmic radiation field.\n\nThe scale factor multiplies all spatial distances in the same way. For two comoving observers with fixed spatial separation in the reference geometry, their proper separation on a constant-\\(t\\) slice is proportional to \\(a(t)\\). No center of expansion is selected: every comoving observer describes nearby comoving observers as receding during an expanding phase. This does not imply that all physical objects expand. Atoms, planetary systems, and gravitationally bound galaxies are not represented by mutually comoving test particles.\n\nThe symmetry assumptions sharply restrict the geometry but leave \\(a(t)\\) undetermined. A static universe, a decelerating expansion, an accelerating expansion, and a contraction can all possess FLRW geometry. The metric can also be employed in gravitational theories other than general relativity; in that case the same symmetry reduction may lead to different equations for the scale factor."
    },
    {
      "heading": "Spatial curvature, coordinate systems, and topology",
      "text": "The three allowed signs of \\(k\\) distinguish positively curved, flat, and negatively curved spatial geometry. For the convention above, the scalar curvature of a constant-time spatial slice is\n\n\\[{}^{(3)}R=\\frac{6k}{a^2}.\\]\n\nFor \\(k=\\pm1\\), the physical curvature radius is \\(a(t)\\). For \\(k=0\\), the intrinsic spatial curvature vanishes at every time, even though the four-dimensional spacetime can still be curved. A spatially flat expanding universe is therefore not generally equivalent to the flat spacetime of special relativity.\n\nA useful radial coordinate \\(\\chi\\) puts the line element in the form\n\n\\[ds^2=-c^2dt^2+a^2(t) \\left[d\\chi^2+S_k^2(\\chi)d\\Omega^2\\right],\\]\n\nwhere \\(d\\Omega^2=d\\theta^2+\\sin^2\\theta\\,d\\phi^2\\), and\n\n\\[S_k(\\chi)= \\begin{cases} \\sin\\chi,&k=+1,\\\\ \\chi,&k=0,\\\\ \\sinh\\chi,&k=-1. \\end{cases}\\]\n\nThe proper radial distance from the coordinate origin, measured along a constant-time slice, is \\(a\\chi\\), whereas the area of a sphere at that radial coordinate is \\(4\\pi a^2S_k^2(\\chi)\\). Their difference encodes spatial curvature. In positive curvature, sphere areas grow more slowly with radial distance than in Euclidean space; in negative curvature, they grow more rapidly.\n\nCoordinate coverage requires care. In the positively curved case, setting \\(r=\\sin\\chi\\) makes the \\(r\\) coordinate unsuitable for covering the entire spatial three-sphere in a single one-to-one chart. The apparent singularity at \\(r=1\\) in the first form of the metric is a coordinate problem, not a physical singularity. Curvature invariants and alternative coordinates distinguish this kind of chart boundary from a genuine breakdown of spacetime geometry.\n\nLocal curvature does not uniquely determine cosmic topology. The simply connected reference spaces are the three-sphere, Euclidean three-space, and hyperbolic three-space. Identifications of points can produce other spaces with the same local curvature. A flat three-torus, for example, is spatially finite despite having zero local curvature. Such identifications may reduce global rotational or translational symmetries, so cosmological discussions often distinguish an exactly globally homogeneous and isotropic space from a locally FLRW geometry with nontrivial topology.\n\nThe traditional labels “closed” for positive curvature and “open” for negative curvature can consequently be misleading. They are sometimes used to imply spatial finiteness or a particular cosmic fate, neither of which follows from curvature alone in a general cosmological model. A positively curved universe with a sufficiently large positive cosmological constant can expand indefinitely, while spatial topology can make a locally flat universe compact."
    },
    {
      "heading": "Comoving motion, expansion, and peculiar velocities",
      "text": "The expansion rate is described by the Hubble parameter,\n\n\\[H(t)=\\frac{\\dot a(t)}{a(t)}.\\]\n\nFor comoving observers separated radially by a fixed coordinate interval \\(\\Delta\\chi\\), the proper distance on a constant-time slice is \\(D(t)=a(t)\\Delta\\chi\\). Differentiation gives\n\n\\[\\dot D=HD.\\]\n\nThis is the geometric form of the Hubble expansion law. It is exact for this definition of simultaneous proper distance in an FLRW spacetime, although observational relations involving redshift and distance require additional treatment of light travel time.\n\nThe comoving observer congruence has expansion scalar \\(\\Theta=3H\\). Its shear and vorticity vanish: a small comoving volume changes size without acquiring a preferred elongation or bulk rotation. Comoving worldlines are geodesics in the ideal FLRW geometry. Although the cosmic fluid can possess nonzero pressure, homogeneity eliminates spatial pressure gradients that would otherwise accelerate fluid elements relative to this congruence.\n\nReal objects can have peculiar velocities, meaning motion relative to the local comoving frame. The observed motion of a galaxy combines background expansion with such local motion. In an unperturbed FLRW universe, the physical momentum magnitude of a freely moving particle scales as\n\n\\[p\\propto a^{-1}.\\]\n\nFor a nonrelativistic particle, this implies that its peculiar speed decreases approximately as \\(a^{-1}\\). For a freely propagating photon, the same momentum scaling gives the cosmological decrease in photon energy. Forces, interactions, and gravitational inhomogeneities can alter these simple relations.\n\nAn important distinction separates \\(\\dot D\\) from a locally measured relative velocity. At sufficiently large \\(D\\), the FLRW recession rate \\(HD\\) can exceed \\(c\\). This does not violate the local relativistic speed limit because it compares distances between separated worldlines using a chosen cosmological simultaneity convention. A local observer still measures any passing light ray to travel at \\(c\\). General relativity does not assign a unique special-relativistic relative velocity to arbitrary widely separated objects without specifying a comparison procedure.\n\nExpansion also should not be understood as a force that uniformly stretches every material structure. The metric describes the ideal large-scale flow. Bound systems require a more detailed local gravitational or electromagnetic description, and their characteristic sizes need not track \\(a(t)\\). Nor does the phrase “expanding space” mean that the universe must expand into an external arena. The scale factor describes changes in intrinsic spatial geometry; no surrounding higher-dimensional space is required by the theory."
    },
    {
      "heading": "Gravitational dynamics and the Friedmann equations",
      "text": "When the FLRW metric is substituted into the Einstein field equations, spatial symmetry reduces the gravitational dynamics to ordinary differential equations. The compatible matter source has the form of a perfect fluid, characterized by an energy density \\(\\varepsilon(t)\\), an isotropic pressure \\(p(t)\\), and the comoving four-velocity. Multiple homogeneous components can be combined by adding their energy densities and pressures. A directed energy flux or anisotropic stress would conflict with exact isotropy.\n\nWriting the cosmological constant separately, the Friedmann equations are\n\n\\[H^2= \\frac{8\\pi G}{3c^2}\\varepsilon +\\frac{\\Lambda c^2}{3} -\\frac{kc^2}{a^2},\\]\n\nand\n\n\\[\\frac{\\ddot a}{a} = -\\frac{4\\pi G}{3c^2}(\\varepsilon+3p) +\\frac{\\Lambda c^2}{3}.\\]\n\nHere \\(G\\) is Newton’s gravitational constant and \\(\\Lambda\\) is the cosmological constant. The first equation constrains the expansion rate, energy density, and spatial curvature. The second determines whether the expansion accelerates or decelerates. Positive pressure contributes to gravitational deceleration in this equation; pressure is not merely a mechanical resistance to expansion.\n\nLocal energy-momentum conservation gives the continuity equation,\n\n\\[\\dot\\varepsilon+3H(\\varepsilon+p)=0.\\]\n\nFor a comoving volume \\(V\\propto a^3\\), it can be expressed as\n\n\\[d(\\varepsilon V)=-p\\,dV.\\]\n\nThis resembles the adiabatic first law of thermodynamics. It describes local conservation in the expanding fluid, not necessarily a globally conserved total energy for the universe. A generic expanding FLRW spacetime lacks the time-translation symmetry needed for the usual conserved global energy associated with stationary systems.\n\nThe equations are not all independent. The geometric identities underlying the Einstein equations ensure that the continuity equation and suitable Friedmann equations are mutually consistent. To obtain a specific solution, an equation of state, initial conditions, and the value of \\(\\Lambda\\) must also be supplied. A common parametrization is \\(p=w\\varepsilon\\), with constant or time-dependent \\(w\\). If components exchange energy, their separate continuity equations contain transfer terms, while the total still satisfies the conservation equation above.\n\nThe cosmological constant can instead be treated as an effective fluid,\n\n\\[\\varepsilon_\\Lambda=\\frac{\\Lambda c^4}{8\\pi G}, \\qquad p_\\Lambda=-\\varepsilon_\\Lambda.\\]\n\nUnder that convention it is included in the total density and pressure rather than added separately to the Friedmann equations. Counting it in both ways would double its contribution. The acceleration equation shows why sufficiently negative pressure can produce accelerated expansion: with \\(\\Lambda\\) absorbed into the fluid description, acceleration requires the total \\(\\varepsilon+3p\\) to be negative."
    },
    {
      "heading": "Cosmic components and characteristic expansion histories",
      "text": "For a separately conserved component with constant equation-of-state parameter \\(w\\), the continuity equation integrates to\n\n\\[\\varepsilon\\propto a^{-3(1+w)}.\\]\n\nPressureless matter has \\(w=0\\), so its energy density scales as \\(a^{-3}\\). This describes dilution of a fixed number of particles in an expanding volume when their rest-mass energy dominates. It is a good background approximation for cold dark matter and for nonrelativistic ordinary matter when pressure is dynamically negligible.\n\nRadiation has \\(w=1/3\\), giving \\(\\varepsilon_{\\rm r}\\propto a^{-4}\\). Three powers arise from volume dilution and one from the reduction in each particle’s energy. For a thermal photon distribution expanding adiabatically without changes in the relevant particle content, the temperature scales as \\(T\\propto a^{-1}\\). Relativistic neutrinos also contribute to radiation, but as massive neutrinos become nonrelativistic their background behavior changes toward that of matter. Particle interactions and changes in the number of thermally coupled species complicate the earliest thermal history.\n\nIn a spatially flat universe dominated by a single constant-\\(w\\) component, with no separate cosmological constant and \\(w\\ne-1\\), the familiar expanding power-law solution is\n\n\\[a(t)\\propto(t-t_{\\rm B})^{2/[3(1+w)]}\\]\n\nfor the standard case \\(w>-1\\). Here \\(t_{\\rm B}\\) locates the extrapolated initial singularity of that solution. Radiation domination gives \\(a\\propto(t-t_{\\rm B})^{1/2}\\), while matter domination gives \\(a\\propto(t-t_{\\rm B})^{2/3}\\). Both expand with decreasing \\(\\dot a\\). The flat, pressureless, zero-\\(\\Lambda\\) case is commonly called the Einstein–de Sitter model.\n\nA positive cosmological constant instead supports exponential expansion in a spatially flat slicing,\n\n\\[a(t)\\propto e^{Ht}, \\qquad H=\\sqrt{\\frac{\\Lambda c^2}{3}},\\]\n\nwhen other components are negligible. This is a representation of de Sitter spacetime. The standard cosmological model passes approximately through radiation-dominated, matter-dominated, and dark-energy-dominated stages rather than following any one of these limiting solutions for all time. Models of cosmic inflation also use nearly exponential expansion, usually driven by an approximately vacuum-like early-universe component rather than by the late-time cosmological constant.\n\nSpatial curvature supplies another instructive limit. With \\(k=-1\\), no matter, and \\(\\Lambda=0\\), the expanding solution has \\(a=ct\\), after a choice of time origin. This Milne model is a portion of flat Minkowski spacetime written in expanding coordinates. It demonstrates that a changing scale factor does not alone establish nonzero four-dimensional curvature. Conversely, many matter-filled solutions have genuine curvature singularities when extrapolated to \\(a=0\\). Whether an FLRW solution begins, recollapses, bounces, or expands forever depends on its contents and dynamical assumptions, not merely on the sign of \\(k\\)."
    },
    {
      "heading": "Light propagation, redshift, and cosmological distances",
      "text": "For a radial light ray, \\(ds^2=0\\) gives\n\n\\[d\\chi=\\pm\\frac{c\\,dt}{a(t)}.\\]\n\nPhotons therefore connect emission and observation events through the entire intervening expansion history. Comparing successive wave crests emitted and received by comoving observers yields the cosmological redshift relation,\n\n\\[1+z=\\frac{a(t_0)}{a(t_{\\rm e})},\\]\n\nwhere \\(t_{\\rm e}\\) is the emission time and \\(t_0\\) the observation time. The same factor stretches observed time intervals: an emitted duration \\(\\Delta t_{\\rm e}\\) is observed as \\((1+z)\\Delta t_{\\rm e}\\), when the expansion is effectively unchanged over the event’s duration.\n\nThe redshift relation concerns comoving emission and observation in the unperturbed geometry. Actual observations also include peculiar-motion effects and gravitational frequency shifts associated with inhomogeneities. Cosmological redshift can be described using a chain of local comparisons between neighboring observers; it does not require assigning a single special-relativistic Doppler velocity to a distant galaxy. At low redshift the distinction is small enough that the familiar approximate relation \\(cz\\simeq H_0D\\) is useful.\n\nSeveral inequivalent distance measures are needed because the geometry evolves while light travels. Define the present-normalized scale factor \\(A(t)=a(t)/a_0\\), where \\(a_0=a(t_0)\\), and let\n\n\\[D_{\\rm C}(z)=c\\int_0^z\\frac{dz'}{H(z')}\\]\n\nbe the radial comoving distance expressed as a present-day length. The transverse comoving distance is\n\n\\[D_{\\rm M}= \\begin{cases} a_0\\sin(D_{\\rm C}/a_0),&k=+1,\\\\ D_{\\rm C},&k=0,\\\\ a_0\\sinh(D_{\\rm C}/a_0),&k=-1. \\end{cases}\\]\n\nThis distinction matters because curvature changes the relation between a radial interval and the area of a corresponding wavefront.\n\nThe angular-diameter distance \\(D_{\\rm A}\\) converts a small observed angular size into a transverse proper size at emission. The luminosity distance \\(D_{\\rm L}\\) relates an object’s intrinsic luminosity \\(L\\) to observed bolometric flux \\(F=L/(4\\pi D_{\\rm L}^2)\\). In FLRW geometry,\n\n\\[D_{\\rm A}=\\frac{D_{\\rm M}}{1+z}, \\qquad D_{\\rm L}=(1+z)D_{\\rm M}.\\]\n\nConsequently,\n\n\\[D_{\\rm L}=(1+z)^2D_{\\rm A}.\\]\n\nThis distance-duality relation assumes photons follow null geodesics and their number is conserved along the beam; absorption requires separate treatment.\n\nThe lookback time is another distinct quantity,\n\n\\[t_0-t_{\\rm e} = \\int_0^z\\frac{dz'}{(1+z')H(z')}.\\]\n\nMultiplying it by \\(c\\) does not generally give either the present proper separation or the angular-diameter distance. These differences are central to interpreting distant supernovae, galaxy sizes, and standard rulers. In standard expansion histories, angular-diameter distance eventually decreases with increasing redshift, so an object of fixed proper transverse size can appear larger when placed at sufficiently high redshift."
    },
    {
      "heading": "Conformal time and causal horizons",
      "text": "The causal structure of FLRW spacetime becomes especially transparent through conformal time. With the dimensionful scale-factor convention used here, define the dimensionless coordinate \\(\\eta\\) by\n\n\\[d\\eta=\\frac{c\\,dt}{a(t)}.\\]\n\nThe metric then becomes\n\n\\[ds^2=a^2(\\eta) \\left[-d\\eta^2+d\\chi^2+S_k^2(\\chi)d\\Omega^2\\right].\\]\n\nThe overall positive conformal factor changes measured lengths and proper times but not local light cones. Radial light rays satisfy \\(d\\chi=\\pm d\\eta\\), so they appear as lines at \\(45^\\circ\\) in a diagram of radial conformal coordinates.\n\nA particle horizon concerns how far signals could have traveled since the earliest boundary of a model. Its comoving radial extent at time \\(t\\), when finite, is\n\n\\[\\chi_{\\rm p}(t) = \\int_{t_{\\rm i}}^t\\frac{c\\,dt'}{a(t')}.\\]\n\nThe corresponding radial proper distance on that time slice is \\(a(t)\\chi_{\\rm p}(t)\\). In the usual radiation- and matter-dominated big-bang solutions, the integral from the initial singularity is finite. Regions sufficiently far apart therefore have no overlapping causal past within a simple extrapolation of those expansion laws. This causal issue underlies the horizon problem addressed by inflationary cosmology.\n\nAn event horizon asks a different question: whether a signal emitted at time \\(t\\) can ever reach a sufficiently distant comoving observer. Its comoving radial extent is\n\n\\[\\chi_{\\rm e}(t) = \\int_t^{t_{\\rm f}}\\frac{c\\,dt'}{a(t')},\\]\n\nprovided the integral converges, where \\(t_{\\rm f}\\) is the future endpoint of the model. Eternal expansion does not guarantee an event horizon. Matter-dominated power-law expansion has none, whereas sustained exponential expansion has a finite event horizon. The distinction depends on the entire future expansion history, not only on the current value of the expansion rate.\n\nThe Hubble radius \\(c/H\\) is different from either horizon. It is the proper radial distance at which \\(HD=c\\), and it often characterizes the relative importance of expansion in perturbation calculations. It is not generally an absolute boundary for light propagation. Photons emitted beyond the Hubble radius can reach an observer in suitable cosmological histories because the Hubble radius and the photon’s proper distance both change with time. In flat de Sitter expansion, the Hubble radius and event-horizon radius coincide, but that equality is special.\n\nGlobal spatial geometry can further affect causal paths. In a compact universe, light may in principle traverse spatial identifications, and in a positively curved simply connected universe it can travel along great circles. Horizon integrals describe available conformal travel distances; their interpretation must also account for the spatial manifold. A finite particle horizon, an event horizon, and a finite total spatial volume are separate properties."
    },
    {
      "heading": "Observational use and cosmological parameters",
      "text": "The FLRW metric becomes an observational model when its expansion history is expressed through measurable parameters. Using mass-equivalent density \\(\\rho=\\varepsilon/c^2\\), the present critical density is\n\n\\[\\rho_{\\rm c0}=\\frac{3H_0^2}{8\\pi G}.\\]\n\nDensity parameters are defined by \\(\\Omega_{i0}=\\rho_{i0}/\\rho_{\\rm c0}\\), while the curvature parameter is\n\n\\[\\Omega_{k0}=-\\frac{kc^2}{a_0^2H_0^2}.\\]\n\nThus positive spatial curvature corresponds to negative \\(\\Omega_{k0}\\), a sign convention that can otherwise cause confusion. The present-day Friedmann equation gives a sum of unity when all density components and curvature are included consistently.\n\nFor matter, radiation, a cosmological constant, and curvature, the background expansion law is\n\n\\[\\frac{H^2(z)}{H_0^2} = \\Omega_{\\rm r0}(1+z)^4 +\\Omega_{\\rm m0}(1+z)^3 +\\Omega_{k0}(1+z)^2 +\\Omega_{\\Lambda0}.\\]\n\nThis expression assumes the relevant matter is nonrelativistic and separately conserved, while radiation remains relativistic. Precision calculations treat transitions such as massive-neutrino evolution more carefully. A general homogeneous dark-energy component with equation of state \\(w(z)\\) instead contributes a factor proportional to\n\n\\[\\exp\\!\\left[ 3\\int_0^z\\frac{1+w(z')}{1+z'}\\,dz' \\right].\\]\n\nDifferent observations constrain different combinations of geometry and dynamics. Type Ia supernovae primarily trace luminosity distance as a function of redshift. Baryon acoustic oscillations supply a statistical standard ruler in the distribution of matter, constraining transverse distances and radial expansion rates. The cosmic microwave background provides an early-universe acoustic scale whose observed angular size strongly constrains the distance to last scattering relative to the sound horizon. These methods share assumptions about source physics or the early universe as well as the FLRW background.\n\nAs an example of a parameter determination within a specified model, the Planck 2018 analysis of cosmic microwave background temperature, polarization, and lensing data in spatially flat base Lambda–cold dark matter cosmology found \\(H_0\\) near \\(67.4\\) kilometres per second per megaparsec, a matter density fraction near \\(0.315\\), and an age near \\(13.8\\) billion years. These are fitted properties of that model and dataset, not values imposed by the FLRW metric. Measurements of \\(H_0\\) using other methods need not return the same central value.\n\nThe age of an expanding big-bang model follows from\n\n\\[t_0-t_{\\rm B} = \\int_0^1\\frac{dA}{A\\,H(A)}.\\]\n\nIt is not generally equal to \\(H_0^{-1}\\), because the expansion rate changes over time. Likewise, apparent constraints on spatial curvature depend on the adopted expansion and primordial-perturbation model. Combining probes helps break degeneracies, but a successful parameter fit tests a package of geometric and physical assumptions rather than homogeneity and isotropy in isolation."
    },
    {
      "heading": "Perturbations, averaging, and limits of the background description",
      "text": "An exactly FLRW universe has no galaxies, clusters, or cosmic web: its matter fields are spatially uniform. The usual theory of structure formation therefore writes the physical geometry and matter distribution as an FLRW background plus perturbations. Perturbation theory tracks small deviations in density, velocity, and metric variables. At linear order, these deviations can be separated into scalar, vector, and tensor sectors according to their transformation properties on the spatial background.\n\nScalar perturbations describe density fluctuations and gravitational potentials and provide the principal seeds for large-scale structure. Vector perturbations represent rotational patterns and generally decay in the simplest expanding perfect-fluid models unless sustained by sources. Tensor perturbations describe propagating gravitational waves. The FLRW scale factor enters all three sectors, setting how physical wavelengths and characteristic propagation scales evolve. Fourier methods are particularly simple for flat spatial slices; curved spaces require the appropriate spatial harmonics.\n\nRelativistic perturbation theory has a gauge issue: a small coordinate change alters which events are compared with the chosen background. A density perturbation evaluated at fixed coordinate time can therefore change under a redefinition of time slicing. Observable predictions require either explicitly gauge-invariant combinations or a carefully specified gauge. For example, the relation between a measured galaxy number count and an underlying matter perturbation includes velocity, redshift, and light-propagation effects, not only a local density contrast.\n\nOn sufficiently small scales, density contrasts become nonlinear. Galaxies and clusters then require nonlinear dynamics, often modeled through numerical simulations of matter evolving in an expanding background. Weak gravitational potentials can coexist with very large density contrasts, so the success of a weak-field treatment does not imply that the density perturbation itself is small. Light traveling through these structures also undergoes gravitational lensing, producing magnification and shear around the background distance-redshift relation.\n\nA deeper issue is whether averaging an inhomogeneous universe produces exactly the same evolution as an FLRW model with averaged matter. The Einstein equations are nonlinear, so averaging and solving the equations need not commute. This motivates research on cosmological backreaction. It does not by itself establish that inhomogeneities replace dark energy or cause the observed acceleration; quantitative results depend on the averaging prescription, model, and observable being considered.\n\nThe background approximation also has limits near very strong gravitational fields and in extrapolations toward an initial singularity. A classical FLRW solution does not supply a theory of quantum gravity, nor does it settle what, if anything, precedes the earliest phase to which the solution applies. Its empirical role is more restricted: it supplies a controlled large-scale geometry against which cosmic evolution, structure, and light propagation can be calculated and tested."
    },
    {
      "heading": "Historical development and terminology",
      "text": "The modern FLRW framework emerged from several distinct advances rather than from a single derivation. In 1917, Albert Einstein applied general relativity to cosmology using a static, spatially closed model supported by a cosmological constant. Willem de Sitter introduced another cosmological solution that year, with no ordinary matter and a positive cosmological constant. Their models established the importance of global relativistic geometry but did not yet provide the standard matter-filled, evolving cosmology.\n\nAlexander Friedmann’s 1922 paper demonstrated nonstatic solutions of the Einstein equations with positively curved homogeneous spatial sections. His 1924 work also considered negative spatial curvature. These solutions showed that general relativity did not require the universe to be static and that expansion or contraction could follow naturally from the field equations. Their immediate significance was mathematical: different cosmic histories could be derived by varying the matter content, curvature, and cosmological constant.\n\nGeorges Lemaître independently developed an expanding relativistic universe in 1927 and explicitly connected its dynamics to astronomical observations. He related the recession of extragalactic nebulae to their distances and estimated an expansion rate from available data. Edwin Hubble’s 1929 distance–velocity study subsequently became central to the observational establishment of the expansion relation. The geometrical model, the dynamical equations, and the observational evidence were thus developed through overlapping but distinguishable lines of work.\n\nHoward P. Robertson and Arthur Geoffrey Walker clarified the general geometric basis of homogeneous and isotropic cosmological line elements in work during the 1920s and 1930s, including Robertson’s mid-1930s analyses and Walker’s 1937 treatment. Their role was to characterize the spacetime form allowed by the symmetry assumptions, independently of choosing a particular matter model and solving its dynamics. This is the reason “Robertson–Walker metric” is often used when the emphasis is specifically geometrical.\n\nThe names “Friedmann–Robertson–Walker,” “Friedmann–Lemaître–Robertson–Walker,” and “Robertson–Walker” remain in use, often with overlapping meanings. “FLRW” makes explicit the combined contributions to dynamical cosmology and the symmetry-based metric. More narrowly, a Friedmann model usually means an FLRW geometry whose scale factor satisfies the Einstein equations for specified cosmic components. The distinction matters whenever the metric is used in a modified gravitational theory or as a purely kinematic description of observations.\n\nAfter the discovery of the cosmic microwave background in 1965, hot big-bang cosmology acquired a major observational foundation. Later measurements of microwave-background anisotropies, cosmic structure, and the distance–redshift relation made FLRW geometry the background for precision parameter inference. The framework’s historical continuity lies in the separation between a highly symmetric spacetime geometry, equations for its evolution, and observations used to determine which particular cosmic history describes the universe."
    }
  ],
  "links": [
    "spacetime",
    "general-relativity",
    "lambda-cdm-model",
    "cosmological-principle",
    "speed-of-light",
    "proper-time",
    "special-relativity",
    "hubble-parameter",
    "pressure",
    "photon",
    "einstein-field-equations",
    "friedmann-equations",
    "cosmological-constant",
    "equation-of-state",
    "dark-matter",
    "neutrino",
    "de-sitter-spacetime",
    "cosmic-inflation",
    "redshift",
    "doppler-effect",
    "supernova",
    "dark-energy",
    "baryon-acoustic-oscillations",
    "cosmic-microwave-background",
    "perturbation-theory",
    "gravitational-wave",
    "gravitational-lensing",
    "quantum-gravity",
    "albert-einstein",
    "big-bang"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}