How do galaxy cluster dynamics and gravitational lensing constrain dark matter over modified gravity differently?

# Galaxy Clusters, Lensing, and the Dark Matter vs. Modified Gravity Puzzle

This is a genuinely subtle question—and the archive shows why it matters: lensing and dynamics can look deceptively similar when testing gravity, but they're not identical probes.

The Core Degeneracy

The central problem is that different physical theories can produce the same lensing signal. Weak gravitational lensing depends on the cumulative mass along the line of sight, encoded in the Weyl potential (the sum of space and time geometry distortions). Modified gravity theories *and* dark matter with extra forces can both alter this potential in observationally similar ways [AG-2024.04-266]. For example, a fifth force coupled to dark matter produces mass distributions and growth rates that mimic those from f(R) gravity or other modified gravity variants. This is not just a minor nuisance—it's described as an "inevitable" degeneracy [AG-2024.04-266].

Why this matters: if lensing and growth-of-structure measurements (from redshift-space distortions) are your only tools, you cannot uniquely identify whether you're seeing modified gravity or an invisible fifth force.

How Galaxy Cluster Data Helps Break the Tie

Cluster-scale observations offer a concrete handle. Detailed lensing maps of massive clusters—combining both strong lensing (which bends light dramatically near the cluster core) and weak lensing (subtle bending at larger radii)—constrain the *internal density structure* of the cluster's total mass profile.

Disformal coupling models (a class of non-minimal dark matter–gravity couplings) predict a modified Poisson equation with a characteristic coupling length *L* [AG-2024.01-045]. When tested against CLASH survey cluster lensing data, this length is constrained to be extremely small—around 100 kpc or less at 1σ confidence [AG-2024.12-570]. This is a direct, local test: if dark matter truly couples to gravity differently, clusters *should* show it in their mass profiles. The fact that they don't rules out many modified coupling scenarios.

By contrast, MOND (Modified Newtonian Dynamics) fails catastrophically in clusters [AG-2024.10-060]. Even after accounting for a residual "missing mass," the density profile required to rescue MOND in clusters is bizarre—an inner constant-density core with a sharp outer cutoff—and differs markedly from dark matter predictions. This *falsifiability* is the point: clusters reveal that MOND and GR+dark matter are not equivalent.

The Key Difference: Geometry vs. Matter

Here's where cluster dynamics and lensing pull apart most clearly. A new test proposes to measure whether the growth of matter density fluctuations matches the growth of spacetime geometry [AG-2025.06-066]. In general relativity, these two growths are locked together by the same equations. In modified gravity, they can diverge. The test combines galaxy clustering (a dynamical probe sensitive to matter acceleration) with lensing (a geometric probe). When applied to current data, the two are found to agree to within 33% from early times to z ∼ 0.5 [AG-2025.06-066]—a null result that disfavors theories where geometry and matter decouple badly.

Meanwhile, f(R) and nDGP modified gravity models do enhance lensing signals relative to GR, producing larger Einstein radii and higher magnification factors [AG-2024.11-168]. So lensing is *sensitive* to these theories—it's just that the sensitivity alone doesn't tell you why the signal is enhanced (modified gravity vs. a new fifth force).

An Emerging Crisis: Substructure in Clusters

There is one regime where lensing data on clusters *does* point toward something beyond standard CDM: the internal density profiles and radial distribution of cluster subhalos. Strong lensing from subhalo cores is observed at rates an order of magnitude higher than cold collisionless dark matter predicts [AG-2026.01-1153]. This hints at self-interacting dark matter (SIDM), where particles scatter off one another and produce denser, more concentrated cores. Curiously, the outer parts of these subhalos behave like collisionless CDM, suggesting a composite picture [AG-2026.01-1153].

Summary

Cluster lensing constrains dark matter and modified gravity differently because:

1. Lensing measures cumulative mass, which both modified gravity and exotic dark matter can mimic.

2. Cluster internal structure breaks degeneracies by testing whether mass distributions match predictions of each theory locally.

3. Dynamics (growth of structure) and lensing (geometry) are separate probes that must agree in GR but can diverge in modified theories—and this mismatch is a direct signature.

4. Substructure offers an independent window: current data favor SIDM over collisionless CDM but don't yet decisively rule out modified gravity on cluster scales.