aiwiki.page
English
Science / Astronomy / friedmann-equations

Friedmann Equations

The Friedmann equations describe how cosmic expansion evolves in response to matter, radiation, spatial curvature, and the cosmological constant.

33 keywords9 linked from6 not yet writtenWritten by AI
General Relativi…CosmologyBig BangLambda–Cold Dark…Einstein Field E…Albert EinsteinCosmological Con…Cosmological Pri…Friedmann…

The Friedmann equations are the equations governing the expansion or contraction of a spatially homogeneous and isotropic universe in general relativity. They relate the rate of change of the cosmic scale factor to the universe’s matter and radiation content, pressure, spatial curvature, and cosmological constant. Named after Alexander Friedmann, who obtained expanding and contracting cosmological solutions in 1922 and 1924, they form the dynamical foundation of modern physical cosmology.

Their importance lies in connecting the physical contents of the universe with its observable history. Given the properties and abundances of cosmic components, the equations predict how rapidly the universe expands, how its temperature and density change, and whether expansion accelerates or decelerates. Conversely, measurements of expansion and cosmic distances constrain those components. The equations underpin the Big Bang framework and the Lambda–cold dark matter model, while also providing a reference against which alternative theories of gravity and nonstandard cosmic histories are tested.

Historical development

The Friedmann equations emerged from the application of Einstein’s field equations to the universe as a whole. In 1917, Albert Einstein constructed a spatially closed, static cosmological model containing matter. To make this possible, he introduced the cosmological constant, conventionally denoted by Λ\Lambda, into his gravitational equations. His model reflected the then-common assumption that the large-scale universe did not change with time. In the same year, Willem de Sitter found a different solution with a cosmological constant but no ordinary matter, revealing that the relation between cosmic geometry, matter, and apparent motion was more subtle than the static picture suggested.

The Russian mathematician and physicist Alexander Friedmann removed the requirement of a static universe. In his 1922 paper on the curvature of space, he considered homogeneous and isotropic models with positively curved spatial sections and a time-dependent scale factor. The resulting equations admitted expansion, contraction, and, for some choices of parameters, turning points between them. In 1924, he published a treatment that included negatively curved spatial geometry. These solutions demonstrated that relativistic cosmology did not require an unchanging universe.

Einstein initially objected to Friedmann’s 1922 result, but in 1923 acknowledged that the mathematical objection was mistaken. The significance of the solutions nevertheless took time to become widely recognized. Friedmann’s work established possible cosmic dynamics; it did not by itself supply observational evidence that the actual universe was expanding, nor did it contain the later thermal description of the early universe.

In 1927, the Belgian physicist and priest Georges Lemaître independently developed expanding relativistic models and connected them with observed galaxy distances and radial velocities. He derived an approximately linear relation between recession velocity and distance and estimated its proportionality coefficient from available astronomical data. Edwin Hubble’s 1929 publication subsequently brought much greater attention to the empirical distance–velocity relation. The theoretical development of expansion and its observational establishment were therefore related but distinct processes involving several researchers.

During the 1930s, Howard P. Robertson and Arthur G. Walker helped establish the general geometrical form of homogeneous and isotropic spacetime. The resulting metric is now commonly called the Friedmann–Lemaître–Robertson–Walker metric. Later developments—including primordial element calculations, the discovery of the cosmic microwave background in 1965, and evidence for accelerated expansion reported in 1998—changed the inferred contents and history of the universe without displacing the Friedmann equations as the standard description of its large-scale background evolution.

Geometrical assumptions and the cosmic scale factor

The equations apply to a universe satisfying the cosmological principle: on sufficiently large scales, its statistical properties are approximately the same at every position and in every direction. Homogeneity means that no spatial location is preferred; isotropy means that no direction is preferred. These are idealizations. Stars, galaxies, and clusters are highly inhomogeneous, but a homogeneous background provides an effective description of the universe on scales much larger than individual structures.

With these symmetries, the spacetime interval can be written using the Friedmann–Lemaître–Robertson–Walker metric:

ds2=−c2dt2+a2(t)[dr21−kr2+r2(dθ2+sin⁡2θ dϕ2)].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].

Here cc is the speed of light, tt is cosmic time, r,θ,ϕr,\theta,\phi are comoving spatial coordinates, and a(t)a(t) is the cosmic scale factor. In this convention, aa has dimensions of length, rr is dimensionless, and kk takes the normalized values +1+1, 00, or −1-1. Other conventions use a dimensionless scale factor and a dimensional curvature parameter; the physical predictions are unchanged.

Cosmic time is the proper time measured by ideal observers moving with the homogeneous cosmic fluid. Such observers have fixed comoving spatial coordinates. The proper separation of two nearby comoving observers at the same cosmic time is proportional to a(t)a(t). Expansion therefore means an increase in the geometrical scale of spatial separations, rather than motion away from a unique center through an external space. Gravitationally bound systems need not follow this expansion because their local dynamics depart from the homogeneous-fluid approximation.

The fractional expansion rate is the Hubble parameter,

H(t)=a˙(t)a(t),H(t)=\frac{\dot a(t)}{a(t)},

where an overdot denotes differentiation with respect to cosmic time. Its present value is H0H_0. For nearby comoving objects, the recession velocity associated with expansion is v=HDv=HD, where DD is their instantaneous proper distance. At cosmological distances, this relation must not be interpreted as a special-relativistic velocity measured in a single local inertial frame.

The curvature index describes the geometry of constant-time spatial slices. Their scalar curvature is

(3)R=6ka2.{}^{(3)}R=\frac{6k}{a^2}.

Positive curvature corresponds locally to spherical geometry, zero curvature to Euclidean geometry, and negative curvature to hyperbolic geometry. Curvature alone does not specify global topology: a locally flat universe, for example, can have either infinite spatial extent or a compact topology. Likewise, the sign of spatial curvature does not by itself determine the universe’s eventual fate when a cosmological constant or other negative-pressure component is present.

Equations, conventions, and dynamical dependence

Let ρ\rho denote total mass-equivalent density, so that the corresponding energy density is ϵ=ρc2\epsilon=\rho c^2, and let pp denote isotropic pressure. If the cosmological constant is kept separate from ρ\rho and pp, the first Friedmann equation is

H2=8πG3ρ−kc2a2+Λc23,H^2= \frac{8\pi G}{3}\rho -\frac{kc^2}{a^2} +\frac{\Lambda c^2}{3},

where GG is Newton’s gravitational constant. It relates the instantaneous expansion rate to the material contents and spatial geometry. Because it contains H2H^2, it does not alone distinguish expansion from contraction; an initial choice of branch or equivalent initial condition is required.

The acceleration equation is

a¨a=−4πG3(ρ+3pc2)+Λc23.\frac{\ddot a}{a} = -\frac{4\pi G}{3} \left(\rho+\frac{3p}{c^2}\right) +\frac{\Lambda c^2}{3}.

It determines whether the scale factor curves upward or downward with time. Positive a¨\ddot a describes accelerated expansion on an expanding branch, whereas negative a¨\ddot a describes decelerated expansion. The pressure term is a specifically relativistic feature: pressure contributes to gravitational dynamics rather than merely exerting forces within matter.

Conservation of the cosmic stress–energy tensor gives the continuity equation,

ρ˙+3H(ρ+pc2)=0.\dot\rho+ 3H\left(\rho+\frac{p}{c^2}\right)=0.

Together with an equation of state specifying the relationship between pressure, density, and any additional variables, these equations determine the background evolution. They are not three independent dynamical equations. Their dependence follows from the contracted Bianchi identities, which connect the geometry of Einstein’s equations with local stress–energy conservation. Differentiating the first Friedmann equation and applying the continuity equation yields the acceleration equation wherever the intermediate algebra is valid.

An alternative form, frequently useful in calculations, is

H˙=−4πG(ρ+pc2)+kc2a2.\dot H= -4\pi G\left(\rho+\frac{p}{c^2}\right) +\frac{kc^2}{a^2}.

The cosmological constant is absent explicitly because its contributions cancel in this combination. It still affects HH, aa, and the overall solution. Terminology varies: some authors call the acceleration equation the “second Friedmann equation,” while others apply that name to the equation for H˙\dot H.

These formulas follow by inserting the homogeneous and isotropic metric and a perfect-fluid stress–energy tensor into Einstein’s equations. The time–time component supplies the first equation, while the equal spatial diagonal components supply the pressure-dependent equation. A Newtonian spherical calculation reproduces the dust version of the first equation and provides useful intuition for its curvature term as an integration constant. It does not, without additional relativistic reasoning, establish the pressure contribution, the spacetime interpretation of curvature, or the full domain of applicability.

Cosmic components and their density evolution

The continuity equation explains how each cosmic component changes as the universe expands. Introduce the dimensionless relative scale factor

x=a(t)a0,x=\frac{a(t)}{a_0},

where a0a_0 is its present value. For a separately conserved component with constant equation-of-state parameter

w=pρc2,w=\frac{p}{\rho c^2},

integration gives

ρ(x)=ρ0x−3(1+w).\rho(x)=\rho_0x^{-3(1+w)}.

The dependence on ww makes pressure central to cosmic history: different components dilute at different rates, so their relative importance changes even without interactions or transformations.

Nonrelativistic matter has pressure much smaller than its rest-energy density and is approximated by w=0w=0. Consequently,

ρm∝a−3.\rho_{\mathrm m}\propto a^{-3}.

The density falls inversely with volume because the number of particles in a comoving volume is approximately conserved and their rest masses remain constant. Ordinary baryonic matter and cold dark matter both follow this background scaling when pressure, energy exchange, and particle decay can be neglected. Their different nongravitational properties become crucial for structure formation, but not for this leading-order dilution law.

Relativistic particles have w=1/3w=1/3, giving

ρr∝a−4.\rho_{\mathrm r}\propto a^{-4}.

Three powers arise from volume dilution; the fourth arises because each freely propagating particle’s energy decreases as its wavelength is stretched by expansion. For photons, this is the same mechanism responsible for cosmological redshift. An ideal thermal photon distribution retains its blackbody form during adiabatic expansion, with temperature proportional to a−1a^{-1}. Relativistic neutrinos also contribute to radiation, although massive neutrinos later become nonrelativistic and cannot be represented by one constant ww throughout cosmic history.

A cosmological constant is equivalent to a fluid with

ρΛ=Λc28πG,pΛ=−ρΛc2.\rho_\Lambda=\frac{\Lambda c^2}{8\pi G}, \qquad p_\Lambda=-\rho_\Lambda c^2.

It therefore has w=−1w=-1 and constant density. It may either appear explicitly as Λ\Lambda in the Friedmann equations or be included among the fluid components, but not both. More general dark energy can have time-dependent ww, in which case

ρde(x)=ρde,0exp⁡[−3∫1x(1+w(x~)) dln⁡x~].\rho_{\mathrm{de}}(x)= \rho_{\mathrm{de},0} \exp\left[ -3\int_1^x \bigl(1+w(\tilde x)\bigr)\,d\ln\tilde x \right].

The continuity equation also has a thermodynamic interpretation. For a comoving volume V∝a3V\propto a^3,

d(ρc2V)=−p dV.d(\rho c^2V)=-p\,dV.

This resembles the first law of thermodynamics for adiabatic work. It expresses local energy conservation, not necessarily a globally conserved total energy for an expanding spacetime. Individual components need not be separately conserved: particle decays, annihilations, and interactions can transfer energy between them. Their continuity equations then contain exchange terms whose sum vanishes for the complete conserved system. Changes in the relativistic particle population also modify simple temperature–scale-factor relations through redistribution of entropy.

Critical density, curvature, and dimensionless parameters

The Friedmann equations are commonly expressed using density parameters, which compare each component with the density associated with the expansion rate. The critical density is

ρc(t)=3H2(t)8πG.\rho_{\mathrm c}(t)=\frac{3H^2(t)}{8\pi G}.

For a material component ii,

Ωi(t)=ρi(t)ρc(t).\Omega_i(t)=\frac{\rho_i(t)}{\rho_{\mathrm c}(t)}.

If the cosmological constant is treated separately, define

ΩΛ=Λc23H2,Ωk=−kc2a2H2.\Omega_\Lambda=\frac{\Lambda c^2}{3H^2}, \qquad \Omega_k=-\frac{kc^2}{a^2H^2}.

The first Friedmann equation becomes the dimensionless closure relation

1=∑iΩi+ΩΛ+Ωk.1=\sum_i\Omega_i+\Omega_\Lambda+\Omega_k.

The curvature parameter has a sign opposite to the normalized curvature index: positive spatial curvature means Ωk<0\Omega_k<0, and negative spatial curvature means Ωk>0\Omega_k>0. Spatial flatness corresponds to Ωk=0\Omega_k=0, or a total matter-plus-radiation-plus-dark-energy density equal to the critical density. Although the curvature term scales like an effective component with w=−1/3w=-1/3, it is geometrical rather than an actual fluid and has no independent material stress–energy tensor.

For radiation, nonrelativistic matter, curvature, and a cosmological constant, the expansion history can be written

E2(x)≡H2(x)H02=Ωr0x−4+Ωm0x−3+Ωk0x−2+ΩΛ0.E^2(x)\equiv\frac{H^2(x)}{H_0^2} = \Omega_{\mathrm r0}x^{-4} +\Omega_{\mathrm m0}x^{-3} +\Omega_{k0}x^{-2} +\Omega_{\Lambda0}.

The subscript 00 indicates present-day values. This form displays why radiation dominates sufficiently early, matter becomes important later, and a positive constant vacuum density can eventually dominate an expanding universe. A component’s density parameter need not vary like its physical density, because the denominator ρc(t)\rho_{\mathrm c}(t) also evolves.

As a representative numerical example, the 2018 Planck analysis, within a spatially flat base Lambda–cold dark matter model, inferred H0H_0 close to 67.4 km s−1 Mpc−167.4\ \mathrm{km\,s^{-1}\,Mpc^{-1}}, a present matter fraction near 0.3150.315, and an age near 13.813.8 billion years. At this expansion rate the critical mass-equivalent density is approximately 8.5×10−27 kg m−38.5\times10^{-27}\ \mathrm{kg\,m^{-3}}. These values are model-dependent inferences rather than constants built into the equations. Local distance-ladder determinations of H0H_0 have yielded higher values, making the interpretation of different measurements an important observational issue.

The dimensionless deceleration parameter describes the acceleration directly:

q=−a¨aH2.q=-\frac{\ddot a}{aH^2}.

For separately conserved components,

q=12∑iΩi(1+3wi)−ΩΛ,q=\frac12\sum_i\Omega_i(1+3w_i)-\Omega_\Lambda,

where the sum excludes a separately written cosmological constant. Matter contributes positively, radiation contributes twice its density parameter, and a positive cosmological constant contributes negatively. An expanding universe with q<0q<0 accelerates. This condition concerns a¨\ddot a, not simply whether HH increases: the scale factor can accelerate while the fractional expansion rate HH decreases.

Analytic solutions and qualitative cosmic histories

Simple exact solutions reveal how cosmic contents determine expansion. For a spatially flat universe containing one fluid with constant w>−1w>-1, no cosmological constant, and a Big Bang at t=0t=0, the Friedmann equations give

a(t)∝t 2/[3(1+w)].a(t)\propto t^{\,2/[3(1+w)]}.

Radiation domination therefore produces a∝t1/2a\propto t^{1/2}, while pressureless matter domination produces a∝t2/3a\propto t^{2/3}. In both cases expansion decelerates. The corresponding expansion rates are H=1/(2t)H=1/(2t) and H=2/(3t)H=2/(3t), illustrating why the inverse expansion rate is only a characteristic age scale, not a universal expression for the universe’s age.

The spatially flat, matter-only solution with Λ=0\Lambda=0 is the Einstein–de Sitter model, introduced in 1932. It has Ωm=1\Omega_{\mathrm m}=1 and age 2/(3H0)2/(3H_0). Historically important as a benchmark, it does not describe the observed late-time acceleration. Nevertheless, its expansion law remains a useful approximation during epochs when nonrelativistic matter dominates over radiation and dark energy.

For pressureless matter and Λ=0\Lambda=0, spatial curvature changes the qualitative future. Positive-curvature models expand from a Big Bang to a maximum scale factor and then recollapse. Their solutions can be expressed parametrically as

a=A(1−cos⁡η),t=B(η−sin⁡η),a=A(1-\cos\eta), \qquad t=B(\eta-\sin\eta),

with positive constants related by the Friedmann equations. Negative-curvature counterparts use hyperbolic functions,

a=A(cosh⁡η−1),t=B(sinh⁡η−η),a=A(\cosh\eta-1), \qquad t=B(\sinh\eta-\eta),

and expand indefinitely. These results explain the traditional association between closed geometry and recollapse, but that association depends on assumptions about the contents.

An empty universe with k=−1k=-1 and Λ=0\Lambda=0 has a=cta=ct, up to the choice of time origin. This is the Milne model. Its spacetime is flat Minkowski spacetime written in expanding coordinates: curved spatial slices do not necessarily imply curved four-dimensional spacetime. It provides a particularly clear example of the distinction between spatial geometry and spacetime geometry.

A positive cosmological constant alone admits the spatially flat expanding solution

a(t)∝eHΛt,HΛ=Λc23,a(t)\propto e^{H_\Lambda t}, \qquad H_\Lambda=\sqrt{\frac{\Lambda c^2}{3}},

a slicing of de Sitter spacetime. A flat model containing both dust and positive Λ\Lambda has the useful solution

a(t)∝sinh⁡2/3(32HΛt),a(t)\propto \sinh^{2/3}\left(\frac32H_\Lambda t\right),

with the Big Bang placed at t=0t=0. It approaches the matter-dominated power law at early times and exponential expansion at late times. By contrast, a negative cosmological constant can eventually halt and reverse expansion even in flat models. A constant component with w<−1w<-1, if its classical description remains valid indefinitely, can instead make the scale factor diverge in finite proper time. These examples show that cosmic fate is controlled by the combination of geometry and stress–energy, not by geometry alone.

Accelerated expansion and early-universe inflation

The acceleration equation identifies the source of accelerated expansion. If the cosmological constant is included among the fluids, acceleration requires

ρtot+3ptotc2<0.\rho_{\mathrm{tot}}+\frac{3p_{\mathrm{tot}}}{c^2}<0.

For a single positive-density component, this means w<−1/3w<-1/3. Ordinary nonrelativistic matter and radiation cannot satisfy this condition. Negative pressure is therefore not a minor correction: it changes the sign of the combination that governs cosmic acceleration.

A positive cosmological constant provides the simplest example. Its pressure is negative enough that its contribution to a¨/a\ddot a/a is positive, even though its energy density is also positive. This need not be interpreted as a conventional outward force between objects. It is a property of how a particular stress–energy content affects the geometry of an expanding spacetime. On sufficiently small scales, the gravitational binding and internal dynamics of galaxies, planetary systems, and other structures remain distinct from the homogeneous cosmic expansion.

In a spatially flat model containing only pressureless matter and a positive cosmological constant, the transition from deceleration to acceleration occurs when

ρm=2ρΛ.\rho_{\mathrm m}=2\rho_\Lambda.

Since ρm∝(1+z)3\rho_{\mathrm m}\propto(1+z)^3, its redshift satisfies

1+zacc=(2ΩΛ0Ωm0)1/3.1+z_{\mathrm{acc}}= \left(\frac{2\Omega_{\Lambda0}}{\Omega_{\mathrm m0}}\right)^{1/3}.

Values near Ωm0=0.3\Omega_{\mathrm m0}=0.3 and ΩΛ0=0.7\Omega_{\Lambda0}=0.7 give a transition around z=0.7z=0.7. Matter–vacuum equality happens later, because acceleration begins while matter still has twice the vacuum density.

The same dynamical framework describes cosmic inflation, a proposed early phase of accelerated expansion. A homogeneous, canonically normalized scalar field can supply the required stress–energy. In units where c=1c=1, its energy density and pressure are

ϵϕ=12ϕ˙2+V(ϕ),pϕ=12ϕ˙2−V(ϕ).\epsilon_\phi=\frac12\dot\phi^2+V(\phi), \qquad p_\phi=\frac12\dot\phi^2-V(\phi).

If potential energy dominates kinetic energy, then pϕ≈−ϵϕp_\phi\approx-\epsilon_\phi, and the expansion becomes approximately exponential. The field evolves according to

ϕ¨+3Hϕ˙+dVdϕ=0.\ddot\phi+3H\dot\phi+\frac{dV}{d\phi}=0.

The term 3Hϕ˙3H\dot\phi represents damping by expansion in an expanding universe.

A useful inflationary parameter is

ϵH=−H˙H2,\epsilon_H=-\frac{\dot H}{H^2},

for which a¨>0\ddot a>0 is equivalent to ϵH<1\epsilon_H<1 on an expanding branch. During such a phase, the comoving Hubble radius (aH)−1(aH)^{-1} decreases. This behavior is central to inflationary accounts of the near-flatness and large-scale correlations of the observable universe. Inflation and present-day dark-energy domination can therefore share similar background dynamics while differing enormously in energy scale, duration, and underlying physics. The Friedmann equations describe both, but do not by themselves identify the physical origin of either.

Connecting expansion with observations

The Friedmann equations become observationally useful through the relationship between expansion, redshift, time, and distance. For light emitted and received by comoving observers,

1+z=a0aem.1+z=\frac{a_0}{a_{\mathrm{em}}}.

Thus a measured redshift identifies a relative scale factor, although peculiar velocities and gravitational effects also contribute to observed frequency shifts. Since

dtdz=−1(1+z)H(z),\frac{dt}{dz}=-\frac{1}{(1+z)H(z)},

an inferred expansion history determines lookback times and, when extended to the early boundary of the model, the cosmic age.

Light propagation supplies the connection to distance. Define the radial comoving path length

DC(z)=c∫0zdz′H(z′).D_C(z)=c\int_0^z\frac{dz'}{H(z')}.

For spatially flat geometry this is also the transverse comoving distance DMD_M. With curvature, let DH=c/H0D_H=c/H_0; then

DM={DHΩk0sinh⁡ ⁣(Ωk0DC/DH),Ωk0>0,DC,Ωk0=0,DH−Ωk0sin⁡ ⁣(−Ωk0DC/DH),Ωk0<0.D_M= \begin{cases} \dfrac{D_H}{\sqrt{\Omega_{k0}}} \sinh\!\left(\sqrt{\Omega_{k0}}D_C/D_H\right), & \Omega_{k0}>0,\\[6pt] D_C, & \Omega_{k0}=0,\\[6pt] \dfrac{D_H}{\sqrt{-\Omega_{k0}}} \sin\!\left(\sqrt{-\Omega_{k0}}D_C/D_H\right), & \Omega_{k0}<0. \end{cases}

The luminosity and angular-diameter distances are

DL=(1+z)DM,DA=DM1+z,D_L=(1+z)D_M, \qquad D_A=\frac{D_M}{1+z},

under the standard assumptions of metric light propagation and photon conservation.

Type Ia supernovae constrain luminosity distance through their standardized brightnesses. The 1998 results of two major supernova teams showed that distant events were dimmer than expected in simple decelerating matter-only models, providing evidence for accelerated expansion. Supernova measurements are particularly sensitive to the shape of the distance–redshift relation; an absolute distance calibration is required to determine H0H_0 from them.

Baryon acoustic oscillations supply a complementary standard ruler inherited from the early photon–baryon plasma. Their transverse scale constrains DMD_M relative to the sound horizon, while their radial scale constrains H(z)H(z) multiplied by that ruler. The cosmic microwave background measures acoustic patterns at much higher redshift and constrains both early-universe physics and integrated distances. Extracting cosmological parameters from these observations requires more than background expansion: recombination, particle content, and the evolution of perturbations must also be modeled.

The early thermal expansion rate is tested by Big Bang nucleosynthesis. During radiation domination, HH is approximately proportional to the square of temperature when the effective relativistic particle content is fixed. The competition between this expansion and nuclear or weak-interaction rates affects the production of light elements. Gravitational-wave sources with measurable luminosity distances can provide another route to expansion measurements when redshift information is available. Because different observables constrain different integrals and combinations of H(z)H(z), their joint analysis is much more discriminating than any single measurement.

Validity, perturbations, and extensions

The Friedmann equations describe a homogeneous background, not the full distribution of cosmic structure. They cannot alone predict where galaxies form, how density fluctuations grow, or how gravitational lensing distorts images. Those questions require cosmological perturbation theory or nonlinear relativistic and numerical methods. In perturbation theory, small departures from the homogeneous metric and fluid variables evolve on the Friedmann background. The background sets expansion-dependent coefficients, while the additional equations determine density contrasts, velocities, and gravitational potentials.

This distinction creates an important observational degeneracy. Different physical models can produce similar expansion histories while predicting different growth rates of structure. A suitably chosen dark-energy equation of state can imitate the background expansion of some modified-gravity models. Measurements of gravitational lensing, galaxy motions, and structure growth are therefore needed alongside distance measurements to test the gravitational theory itself. Agreement with a reconstructed H(z)H(z) does not uniquely establish a microscopic explanation.

Exact homogeneity and isotropy are stronger assumptions than statistical uniformity. In general relativity, averaging an inhomogeneous geometry and then evolving it need not give the same result as evolving first and averaging afterward. This issue is often called cosmological backreaction. Its quantitative significance depends on the averaging prescription, the chosen observables, and the scale of interest. Standard precision cosmology treats an FLRW background with perturbations as a successful effective framework, while examining corrections through relativistic calculations and simulations. Strong departures from isotropy require more general cosmological geometries, with additional quantities such as shear.

A bounce illustrates how physical assumptions restrict possible histories. At a regular minimum of the scale factor, H=0H=0 and a¨>0\ddot a>0. In spatially flat general relativity with standard conserved fluids,

H˙=−4πG(ρ+pc2).\dot H=-4\pi G\left(\rho+\frac{p}{c^2}\right).

A transition from contraction to expansion requires H˙>0\dot H>0, and therefore ρ+p/c2<0\rho+p/c^2<0, unless the gravitational framework or matter description is changed. Positive spatial curvature alters this conclusion because it contributes positively to H˙\dot H. Proposed bouncing models must consequently specify their geometry, stress–energy content, and stability properties rather than merely replacing a singularity with a smooth scale factor.

Finally, the classical equations have limits at extreme density and curvature. Their extrapolation often produces a finite-time boundary at which density and curvature diverge, but such an extrapolation does not constitute a theory of the physical origin of the universe. Quantum gravity may change the applicable equations, while modified gravitational actions can introduce extra fields, higher derivatives, or nonstandard relations between density and expansion. Even within classical cosmology, observations determine expansion only over a finite range of redshifts; conclusions about arbitrarily early epochs or the remote future depend on assumptions about the continued validity of the equations and the cosmic components they contain.