Einasto Dark-Matter Halos

• Einasto-type dark-matter halos are spherical density models with continuously varying logarithmic slopes, providing a smooth alternative to fixed power-law profiles.
• They are defined by parameters such as r₋₂, ρ₋₂, and the curvature index n (or α), which together determine the halo's concentration and dynamical structure.
• Their versatility is applied in rotation curve analyses, weak lensing, and relativistic frameworks, offering better fits than traditional singular models.

Einasto-type dark-matter halos are a family of spherical density models in which the logarithmic density slope varies continuously with radius rather than approaching fixed inner and outer power laws. In the halo convention most widely used in cosmology, the profile is written as

$$\ln\!\left(\frac{\rho(r)}{\rho_{-2}}\right)= -\frac{2}{\alpha}\left[\left(\frac{r}{r_{-2}}\right)^\alpha-1\right],$$

with

$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$

where $r_{-2}$ is the radius at which the slope equals $-2$, $\rho_{-2}=\rho(r_{-2})$, and $\alpha$ is the curvature parameter. Equivalent notation uses the Einasto index $n=1/\alpha$. Einasto laws have become central to dark-matter halo studies because they fit simulated relaxed halos accurately, provide a mathematically regular alternative to singular double-power-law forms, and expose a physically significant departure from strict structural self-similarity rather than a mere fitting convenience (Ludlow et al., 2016).

1. Analytic definition and parameter conventions

The standard halo parameterization is

$$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$

which is equivalent to the $\alpha$-form above through $n=1/\alpha$ (Tissera et al., 2014). A central-density form often used in analytical work is

$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$0

with

$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$1

where $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$2 may be chosen as the 3D half-mass radius and $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$3 is fixed by the half-mass condition $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$4 (Retana-Montenegro et al., 2012). A fully normalized half-mass representation is also used,

$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$5

with $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$6 defined implicitly by $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$7 (Baes, 2022).

Across these conventions, the defining property is unchanged: the logarithmic slope is a power law in radius. This is the formal distinction from NFW and related double-power-law profiles, whose asymptotic inner and outer slopes are fixed. In the Einasto family, $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$8 remains the natural scale radius, while $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$9 or $r_{-2}$0 sets the normalization and $r_{-2}$1 or $r_{-2}$2 sets the curvature. Because the same symbol $r_{-2}$3 is also used in some formation papers for the spectral index of the initial power spectrum $r_{-2}$4, the two usages must be kept separate contextually (Ludlow et al., 2011).

2. Curvature, central structure, and asymptotic behavior

The shape parameter is the physically important third degree of freedom. In the $r_{-2}$5-notation, small $r_{-2}$6 means that the slope changes more gradually with radius; large $r_{-2}$7 means stronger curvature, corresponding to a shallower inner profile and a more rapidly steepening outer profile (Ludlow et al., 2016). Idealized collapse experiments state the same anti-correlation directly: low $r_{-2}$8 gives a steep center and shallow outskirts, whereas high $r_{-2}$9 gives a shallow center and steep outskirts (Nipoti, 2016). Since many observational studies use $-2$0, they describe the same behavior as increasing $-2$1 making the inner profile steeper and more cusp-like at fixed $-2$2 (Chemin et al., 2011).

Every finite-$-2$3 Einasto model has a finite central density. In the half-mass normalization,

$-2$4

so $-2$5 as $-2$6 and the density approaches

$-2$7

The family has two formal limits: as $-2$8, $-2$9; as $\rho_{-2}=\rho(r_{-2})$0, the model tends to a uniform-density sphere of radius $\rho_{-2}=\rho(r_{-2})$1 (Baes, 2022). The total mass is finite for finite $\rho_{-2}=\rho(r_{-2})$2,

$\rho_{-2}=\rho(r_{-2})$3

and the enclosed mass is

$\rho_{-2}=\rho(r_{-2})$4

so the gravitational potential is finite at the center and Keplerian at large radii (Retana-Montenegro et al., 2012).

This regular central behavior is one reason Einasto models outperform singular two-parameter forms in many simulation-based fitting exercises. It is not, however, equivalent to a literal constant-density core in the observational sense; the local slope still varies continuously, and different normalizations emphasize different aspects of “concentration” across the family (Baes, 2022).

3. Formation history, memory of initial conditions, and non-universality

The modern interpretation of Einasto-type halos is explicitly non-universal. In scale-free Einstein–de Sitter $\rho_{-2}=\rho(r_{-2})$5-body simulations with $\rho_{-2}=\rho(r_{-2})$6, relaxed halos are well fit by Einasto profiles, but the fitted $\rho_{-2}=\rho(r_{-2})$7 is not a universal constant. At fixed power-spectrum slope, $\rho_{-2}=\rho(r_{-2})$8 increases slightly but systematically with peak height $\rho_{-2}=\rho(r_{-2})$9, independent of whether one changes mass or redshift. More importantly, the $\alpha$0-$\alpha$1 relation itself depends on the spectral index: for $\alpha$2, $\alpha$3 rises from about $\alpha$4 for $\alpha$5 to about $\alpha$6 for $\alpha$7. Halo profiles are therefore only approximately self-similar within a given cosmology; across different initial spectra they retain a memory of the linear density field through the merger hierarchy and the mass-accretion history (Ludlow et al., 2016).

This interpretation is reinforced by idealized dissipationless collapses from Gaussian random fields. There, smoother initial conditions dominated by long-wavelength modes produce rapid coherent collapse, few mergers, and high-$\alpha$8 halos; clumpier initial conditions dominated by short-wavelength modes produce several mergers, dense early substructure, extended outer envelopes, and low-$\alpha$9 halos. Across realizations with $n=1/\alpha$0 and $n=1/\alpha$1 from $n=1/\alpha$2 to $n=1/\alpha$3, the fitted $n=1/\alpha$4 decreases systematically as the fluctuation field becomes clumpier, while the deprojected Sérsic index $n=1/\alpha$5 increases (Nipoti, 2016).

A more explicitly dynamical reading is given by DARKexp, which treats halo non-universality as variation in a central dimensionless potential $n=1/\alpha$6. In that framework, the empirical Einasto parameter is approximately related to the potential depth by

$n=1/\alpha$7

so small $n=1/\alpha$8 corresponds to deeper central potential and larger binding energy per unit mass (Hjorth et al., 2015). A different dynamical route to low-index Einasto-like cores is proposed for galaxies by the moderate-relaxation scenario, in which the inner halo is assembled mainly by particles with large apocenters passing through the center. The resulting central density is close to an Einasto profile with $n=1/\alpha$9 and develops an extended $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$0 region at larger radii (Baushev, 2013).

Taken together, these results displaced the older notion that a single universal halo profile exists once radii and densities are scaled by $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$1 and $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$2. The third parameter is not incidental; it encodes assembly-dependent curvature.

4. Dynamical structure, phase space, and consistency constraints

Einasto density profiles are closely tied to the near-power-law pseudo-phase-space-density structure of simulated halos. Defining

$$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$3

cosmological halos are found to have $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$4 and $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$5 close to $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$6. Jeans-equilibrium calculations show that an Einasto density law naturally produces a $$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$7 profile that is almost a power law over the entire resolved range, and conversely that critical power-law-$$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$8 models can produce density profiles nearly indistinguishable from Einasto profiles over the same radii. The mapping between the two shape parameters is approximately

$$\rho(r)=\rho_{-2}\exp\!\left\{-2n\left[\left(\frac{r}{r_{-2}}\right)^{1/n}-1\right]\right\},$$9

The difference between the two descriptions appears only at radii $\alpha$0, below the convergence limits of the simulations analyzed in that study (Ludlow et al., 2011).

Dynamical admissibility imposes additional restrictions on the full Einasto family. For spherical isotropic systems, the distribution function obtained by Eddington inversion is physically acceptable only if it is non-negative everywhere. A systematic numerical survey found that all Einasto models with $\alpha$1 have a formal isotropic or Osipkov–Merritt distribution function that becomes negative in part of phase space; such models therefore cannot be supported by those orbital structures. All models with $\alpha$2 admit isotropic equilibria, and Osipkov–Merritt anisotropic realizations remain admissible provided the anisotropy radius exceeds a critical value $\alpha$3, which decreases with $\alpha$4 (Baes, 2022).

A separate no-go result concerns stellar tracers rather than the dark matter itself. For an ideal stellar system that is spherically symmetric, isotropic in velocity space, and cored in the sense that $\alpha$5, the Eddington inversion identity implies that an Einasto gravitational potential still forces the stellar distribution function to become negative somewhere under those assumptions. The key point is that, although $\alpha$6 at the exact center, the potential gradient is nonzero at finite radius inside a stellar core while $\alpha$7 remains nearly zero, producing the same inconsistency that had earlier been identified for NFW potentials (Almeida, 2024).

These results delimit the legitimate use of Einasto laws: density fitting alone is not sufficient, and distribution-function positivity can exclude parts of the formal parameter space.

5. Projection, lensing, and observable tracers

The projection of an Einasto halo is analytically nontrivial but tractable. Using Mellin-transform methods, the surface mass density, cumulative projected mass, deflection angle, convergence, average convergence, shear, magnification, and critical curves can all be written in terms of Fox $\alpha$8 functions for general $\alpha$9, with Meijer $n=1/\alpha$0 reductions for integer or half-integer $n=1/\alpha$1 (Retana-Montenegro et al., 2012). The central projected density is finite,

$n=1/\alpha$2

and the central convergence

$n=1/\alpha$3

must exceed unity for multiple imaging; unlike singular NFW lenses, Einasto halos do not automatically satisfy this condition (Retana-Montenegro et al., 2012).

Exact weak-lensing expressions for the shear and the first and second flexions were later derived in the same Mellin-transform framework, again in Fox $n=1/\alpha$4 and Meijer $n=1/\alpha$5 form. In that analysis, the shear and second flexion are especially useful for constraining halo concentration, while the shear and both flexions retain sensitivity to the Einasto index (Retana-Montenegro et al., 2012). An analytically motivated approximation to the projected surface density was also calibrated directly in terms of the 3D parameters $n=1/\alpha$6, with errors below about $n=1/\alpha$7 for $n=1/\alpha$8 and below $n=1/\alpha$9 even for $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$00 as large as $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$01, over $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$02 to roughly $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$03 (Dhar et al., 2011).

A recurrent misconception is that projected Einasto halos are simply Sérsic profiles in disguise. They are not. Although the two families are formally analogous, the projection of a 3D Einasto profile is not itself exactly Sérsic-like, and Sérsic fits depend strongly on whether one works in linear or logarithmic surface density. Structural parameters inferred from Sérsic fits to projected Einasto systems therefore require caution (Dhar et al., 2011).

These projection differences matter observationally. If a true Einasto halo is analyzed with an NFW lensing model, the reduced tangential shear can still be fit extremely well, yet inferred masses and concentrations become biased. For very massive halos, weak-lensing fits can overestimate mass and underestimate concentration by about $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$04, and the induced bias steepens the observed mass–concentration relation. Distinguishing Einasto from NFW by shear alone requires either exceptionally massive single clusters or the stacking of thousands of group-scale systems (Sereno et al., 2015).

Beyond direct lensing, low-metallicity stellar halos can act as tracers of the Einasto structure of the underlying dark halo. In simulated Milky-Way-mass galaxies, the Einasto parameters of the extremely metal-poor stellar halo correlate with those of the dark halo at the level $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$05, and the dark-matter profile shape can be reconstructed to within residuals of about $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$06 after renormalization when in situ contamination is small (Tissera et al., 2014). In simulations, fully model-based identification is also possible: DarkMix treats the particle field as a finite mixture of spherical Einasto components plus background, returning halo centers, half-mass radii, shape indices, expected particle counts, and soft membership probabilities in merging environments (Hurtado-Gil et al., 2022).

6. Astrophysical applications, empirical tensions, and relativistic extensions

In galaxy rotation-curve work, Einasto profiles became a practical standard because they outperform simpler two-parameter halo laws while remaining flexible enough to span cored and cusp-like behavior. In the THINGS sample, the Einasto halo fits rotation curves significantly better than either NFW or pseudo-isothermal models, yet the preferred indices are typically much smaller than those of dissipationless $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$07CDM halos: intermediate- and low-mass galaxies favor $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$08, whereas simulated galaxy halos are often associated with $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$09 or higher. The cusp–core controversy therefore persists inside the Einasto framework itself, as a mismatch between observed low-$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$10 halos and dark-matter-only expectations (Chemin et al., 2011).

The SPARC sample sharpened this empirical picture. Most fitted Einasto indices lie between $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$11 and $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$12, with $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$13 common, and the halo parameters obey strong internal correlations such as

$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$14

for the parameter set aligned with the tightest radial-acceleration relation. For cored systems with $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$15, the core size correlates positively with stellar disk scale length, and the average dark-matter density within $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$16 correlates tightly with the baryon-induced circular velocity at the same radius (Ghari et al., 2018). A related comparison with a semidegenerate fermion model found that low-$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$17, cored Einasto halos are precisely the regime in which microphysical cored alternatives can fit competitively, while higher-$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$18 systems remain more naturally described by the empirical Einasto family (Siutsou et al., 2014).

On cluster scales, Einasto halos have been embedded in hydrostatic polytropic models for the intracluster medium. In seven Chandra-observed clusters, an Einasto-based model reproduces surface-brightness and temperature profiles about as well as the Vikhlinin et al. and Bulbul et al. frameworks, with fitted shape parameters $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$19 spanning roughly $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$20 to $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$21 (Mirakhor, 2018).

Einasto profiles also admit fully relativistic reinterpretations. An Einstein-cluster construction combines an Einasto density law with collisionless particles on circular geodesics, producing a static anisotropic halo with $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$22 and $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$23, a well-defined relativistic tangential pressure profile, and SPARC rotation-curve fits that remain close to the Newtonian phenomenology while adding explicit stress-energy structure (Acharyya et al., 2023). A different general-relativistic construction treats the Einasto density as an isotropic perfect-fluid source and finds regular, horizonless compact configurations; for $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$24, the dominant-energy-condition thresholds are $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$25, $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$26, and $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$27, respectively (Yue et al., 29 Jan 2026). At the black-hole scale, an Einasto halo has also been used as a perturbing matter distribution around a Schwarzschild-like spacetime, with Event Horizon Telescope shadow measurements of Sgr A$$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$28 yielding $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$29 at $$\frac{d\ln\rho}{d\ln r}=-2\left(\frac{r}{r_{-2}}\right)^\alpha,$$30 under the adopted stellar-orbit mass prior, while the Einasto index remains only weakly constrained (Errehymy et al., 8 Jul 2026).

Einasto-type halos are therefore best understood not as a single universal law but as a broad, physically informative family. Their defining curvature parameter encodes assembly history, phase-space structure, and observational systematics across simulation, lensing, stellar-halo tracing, galactic rotation curves, cluster thermodynamics, and relativistic halo modeling.