A New Robust Constraint on the Self-interaction Cross Section of Dark Matter with Double Radio Relic Clusters

Interpretation of mass-galaxy offsets is complicated by unknowns of the merger phase, geometry, and initial conditions.

“Their interpretation is further complicated by unknowns of the merger phase, geometry, and initial conditions.”

• cites D.Wittman et al. 2018 sourcePartly supported

  Section 5 (page unavailable) says the displacement ratio can change over time and cautions against interpreting offsets without regard to merger phase. Section 3.1 describes competing merger scenarios that make some offsets difficult to interpret. The cited passages do not specifically establish unknown initial conditions as a complication. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Many dark matter candidates arise from extensions of the Standard Model in which they reside in a hidden sector with its own gauge symmetries; in such constructions, self-interactions mediated by hidden-sector gauge bosons are generic rather than exceptional.

“Many dark matter candidates arise from extensions of the Standard Model in which they reside in a hidden sector with its own gauge symmetries; in such constructions, self-interactions mediated by hidden-sector gauge bosons are generic rather than exceptional (S.Tulin et al. 2013).”

• cites S.Tulin et al. 2013 sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

Persistent discrepancies on small scales, such as the core-cusp and too-big-to-fail problems, suggest that dark matter may possess a nonzero self-interaction cross section.

“However, persistent discrepancies on small scales, such as the core-cusp and too-big-to-fail problems, suggest that dark matter (DM) may possess a nonzero self-interaction cross section (SICS) (D.N. Spergel & P. J. Steinhardt 2000; S. Tulin & H. Yu 2018).”

• cites D.N. Spergel & P. J. Steinhardt 2000 sourcePartly supported

  On p. 1, the paper identifies shallow observed galaxy cores and an excess of predicted Local Group halos as small-scale discrepancies; on p. 2, it proposes dark-matter self-interactions as a resolution. This supports the core–cusp example and the general inference, but the paper does not identify the specific too-big-to-fail problem. Its discussion of excess halos should not be treated as that distinct problem. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites S. Tulin & H. Yu 2018 sourceSupported

Even in the presence of a nonzero cross section, the expected mass-galaxy offset is frequently too small to be detected with current observational precision.

“Furthermore, numerical simulations have shown that even in the presence of a nonzero cross section, the expected massgalaxy offset is frequently too small to be detected with current observational precision (F.Kahlhoefer et al. 2014).”

• cites F.Kahlhoefer et al. 2014 sourceSupported

A significant self-interaction cross section would induce a momentum-transfer drag, resulting in an offset between the dark matter mass centroid and the galaxies.

“Conversely, a significant SICS would induce a momentumtransfer drag, resulting in an offset between the DM mass centroid and the galaxies (M.Markevitch et al. 2004;D. Harvey et al. 2015).”

• cites M.Markevitch et al. 2004 sourceSupported
• cites D. Harvey et al. 2015 sourcePartly supported

  The main text says that an additional drag force would produce a lag of dark matter behind the stars. But the paper limits this drag-and-offset mechanism to frequent, low-momentum, long-range self-interactions; it says rare, high-momentum, short-range interactions tend instead to scatter dark matter away and cause mass loss. The cited claim omits that distinction. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

The symmetric geometry of systems hosting two relics on opposite sides indicates a nearly head-on binary merger viewed close to the plane of the sky.

“Systems hosting two relics on opposite sides of the cluster are especially useful: the symmetric geometry indicates a nearly head-on binary merger viewed close to the plane of the sky (R. J. van Weeren et al. 2012;N. Golovich et al. 2019)”

• cites R. J. van Weeren et al. 2012 sourcePartly supported

  The Discussion (section 7; page unavailable) says symmetric double radio relics are thought to arise from simple binary mergers and that binary mergers can produce two shock waves traveling in opposite directions. It does not say that this geometry indicates a nearly head-on merger or a view close to the plane of the sky; it also describes the cluster studied here as likely more complex than a simple binary merger. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites N. Golovich et al. 2019 sourcePartly supported

  Section 2 (Radio-relic Sample; page unavailable) says radio-relic selection is biased toward mergers in the plane of the sky. Section 5.2 (Cluster Redshift Histograms; page unavailable) also treats that viewing angle as plausible, but says a merger near apocenter is another possible explanation. Neither passage establishes that two relics on opposite sides indicate a nearly head-on binary merger. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

The interpretation of mass-galaxy offsets is further plagued by the inherent ambiguity of the merger phase and the projection effects of the viewing angle.

“The interpretation of these offsets is further plagued by the inherent ambiguity of the merger phase and the projection effects of the viewing angle (D.Wittman et al. 2018).”

• cites D.Wittman et al. 2018 sourcePartly supported

  Section 5 (Summary and Discussion; page unavailable) says that averaging offsets without regard to merger phase can bias the inferred dark-matter cross section, supporting the merger-phase concern. The paper discusses projecting a galaxy–dark-matter offset onto a galaxy–gas axis in Section 3.1, but the passages found do not establish the claimed projection effects of the viewing angle. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Recent studies have demonstrated that mass-galaxy offsets are susceptible to large systematic uncertainties that undermine their reliability as a dark matter probe.

“However, recent studies have demonstrated that these massgalaxy offsets are susceptible to large systematic uncertainties that undermine their reliability as a dark matter probe (D.Wittman et al. 2018).”

• cites D.Wittman et al. 2018 sourcePartly supported

  The Introduction says the paper finds substantial errors in offsets used for an earlier merging-cluster analysis, and Section 4 reports that correcting them substantially loosens its dark-matter constraints. This supports concern about that analysis, but the claim extends it to the reliability of mass–galaxy offsets as a dark-matter probe generally, which the cited passages do not establish. Our copy provides section labels but no page numbers for these passages. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

These results place the constraint in mild tension with existing SIDM models that require σ/m ≳ 1 cm^2 g^-1 to resolve small-scale discrepancies in dwarf and spiral galaxies.

“These results place our constraint in mild tension with existing SIDM models that require σ/m ≳ 1 cm 2 g -1 to resolve small-scale discrepancies in dwarf and spiral galaxies (J.Zavala et al. 2013;O. D. Elbert et al. 2015;M. Kaplinghat et al. 2016).”

• cites J. Zavala et al. 2013 sourcePartly supported

  The Simulations and Results section (p. 2) says that a constant SIDM cross section of σT/m ≳ 1 cm² g⁻¹ can alleviate the too-big-to-fail problem for Milky Way dwarf spheroidals. The Introduction (p. 1) identifies cores in dwarf and low-surface-brightness galaxies as a broader discrepancy, but the paper does not establish that the stated threshold is required for spiral galaxies. The citing authors’ own ‘mild tension’ assessment is not a finding attributed to this source. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites O. D. Elbert et al. 2015 sourcePartly supported

  The Conclusions report that SIDM cross-sections of 0.5–50 cm² g⁻¹ alleviate the too-big-to-fail problem in dwarf galaxies, supporting the claim that values around or above 1 can address a small-scale discrepancy. The paper does not establish that at least 1 cm² g⁻¹ is required—it finds 0.5 can work—and the cited results concern dwarfs, not spiral galaxies. The citing authors’ assessment of tension with their own constraint is not a finding attributed to this source. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites M. Kaplinghat et al. 2016 sourceSupported

Galaxies often undergo significant sloshing motions within the dark matter halo on timescales much shorter than the overall cluster orbital period.

“Instead, galaxies often undergo significant sloshing motions within the dark matter halo on timescales much shorter than the overall cluster orbital period (S.Y. Kim et al. 2017).”

• cites S.Y. Kim et al. 2017 sourcePartly supported

  Section 4.1 (p. 12) reports that, after dark-matter halos coalesce in simulated self-interacting dark matter mergers, galaxy peaks can continue oscillating around the halo's barycenter by hundreds of kiloparsecs. The paper does not establish that this often happens in clusters generally or that the sloshing timescale is much shorter than the overall cluster orbital period. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Subsequent reanalysis by D. Wittman et al. (2018) identified substantial methodological flaws in the Harvey et al. (2015) study, noting that averaging such offsets is physically unsound because their magnitude and direction depend critically on whether the system is in an approaching or receding phase.

“However, subsequent reanalysis by D. Wittman et al. (2018) identified substantial methodological flaws, noting that averaging such offsets is physically unsound because their magnitude and direction depend critically on whether the system is in an approaching or receding phase.”

• cites D. Wittman et al. (2018) sourceSupported

Galaxies are not perfect tracers of the dark matter potential.

“The primary challenge arises from the complex dynamical behavior of galaxies that are not perfect tracers of the dark matter potential.”

• cites S.Y. Kim et al. 2017 sourceSupported

If dark matter is truly collisionless, the dark matter halos should coincide with the galaxies.

“If DM is truly collisionless, the DM halos should coincide with the galaxies.”

• cites M. Markevitch et al. 2004 sourceSupported

Using the same prior ranges and methodology as for the observational gold sample, the TNG-Cluster validation yields a 68% (95%) upper limit on σ/m of ≲0.32 cm^2 g^-1 (≲0.75 cm^2 g^-1).

“The 68% (95%) upper limit is �0.32 cm 2 g -1 (�0.75 cm 2 g -1 ).”

• AssumptionRests on an assumption: The TNG-Cluster analogs selected are representative of the observational gold sample for the purpose of validating the method.

The joint constraint is consistent with the N^1/2 scaling seen in the mock recovery tests (Figure 4).

“consistent with the N 1/ scaling seen in our mock recovery tests (Figure 4).”

Applying the method to a gold sample of 11 cluster mergers hosting symmetric double radio relics yields an upper limit on the self-interaction cross section of σ/m < 0.22 (0.63) cm2 g-1 at the 68% (95%) confidence level.

“Applying this method to a gold sample of 11 cluster mergers hosting symmetric double radio relics, we determine an upper limit on the SICS of σ/m < 0.22 (0.63) cm 2 g -1 at the 68% (95%) confidence level.”

• Also stated in: § 3.2. Constraints on σ/m · plain

  “As shown in Figure 5, the combined analysis yields a 68% (95%) upper limit of σ/m < 0.22 cm 2 g -1 (<0.63 cm 2 g -1 ) on the dark matter self-interaction cross section.”

• AssumptionRests on an assumption: The 11 gold-sample double radio relic clusters are representative of cluster-scale collisions for constraining the self-interaction cross section.

The joint-likelihood analysis, marginalized over mass, viewing angle, collision velocity, merger phase, and impact parameter, yields an upper limit of σ/m < 0.22 (0.63) cm^2 g^-1 at the 68% (95%) confidence level.

“Our joint-likelihood analysis, marginalized over mass, viewing angle, collision velocity, merger phase, and impact parameter, yields an upper limit of σ/m < 0.22 (0.63) cm 2 g -1 at the 68% (95%) confidence level.”

The halo-to-halo distance is depressed by SIDM-induced drag.

“Because the propagation speed of merger shocks is nearly independent of the SICS, while the halo-to-halo distance is depressed by SIDM-induced drag, the ratio of the two distances translates directly into a constraint on σ/m.”

• Also stated in: § 2.1. Cluster Collision Simulation Setup · plain

  “However, the SIDM halos lose orbital energy via momentum-transfer drag, reaching apocenter earlier and at smaller separations.”

The total uncertainty budget in the current study is dominated by statistical variance from the limited size of the gold sample.

“While the total uncertainty budget in the current study is dominated by statistical variance from the limited size of the gold sample, the sensitivity of our results to prior assumptions must be assessed.”

The propagation speed of merger shocks is nearly independent of the self-interaction cross section.

“Because the propagation speed of merger shocks is nearly independent of the SICS, while the halo-to-halo distance is depressed by SIDM-induced drag, the ratio of the two distances translates directly into a constraint on σ/m.”

• Also stated in: § 2.5. Sensitivity of Halo and Shock Distances to SICS · plain

  “While the shock propagation speed remains nearly independent of the cross section, the postcollision orbital velocity of dark matter halos is sensitive to the SICS.”

• Also stated in: § 2.1. Cluster Collision Simulation Setup · plain

  “The shock-to-shock distance is nearly identical in the collisionless (σ/m = 0 cm 2 g -1 , red) and self-interacting (σ/m = 3 cm 2 g -1 , black) DM cases.”

The shifts produced by all four prior variations remain well within the bootstrap range.

“Notably, the shifts produced by all four prior variations remain well within this bootstrap range.”

In the TNG-Cluster simulation, 15 analogs were identified that experience major binary mergers with small impact parameters and clear symmetric double radio relics.

“we identify 15 analogs within the TNG-Cluster simulation that experience major binary mergers with small impact parameters and clear symmetric double radio relics, as shown in Figure 6.”

The most constraining single system is a factor of ∼3.2 weaker than the joint 68% limit of 0.22 cm^2 g^-1 (0.63 cm^2 g^-1 at 95%).

“Even the most constraining single system is thus a factor of ∼3.2 weaker than the joint 68% limit of 0.22 cm 2 g -1 (0.63 cm 2 g -1 at 95%)”

Merging galaxy clusters are a promising laboratory for measuring the self-interaction cross section of dark matter.

“Merging galaxy clusters are a promising laboratory for measuring the self-interaction cross section (SICS) of dark matter.”

The TNG-Cluster validation demonstrates that the method recovers consistent posteriors when applied to clusters drawn from a cosmological zoom-in simulation, in which the full diversity of halo profiles and environments is naturally present.

“the TNG-Cluster validation (Section 4.1) plays a critical role in this regard: it demonstrates that our method recovers consistent posteriors when applied to clusters drawn from a cosmological zoom-in simulation, in which the full diversity of halo profiles and environments is naturally present.”

• AssumptionRests on an assumption: The TNG-Cluster simulation naturally contains the full diversity of halo profiles and environments needed to validate the method.

Comparing interpolated predictions to directly simulated shock-to-shock and halo-to-halo distances, agreement is well within 2% in all cases.

“Comparing the interpolated predictions to the directly simulated shock-to-shock and halo-to-halo distances, we find agreement well within 2% in all cases.”

The hybrid quadratic-linear interpolation captures the nonlinear dependence on M2 with sufficient accuracy for Bayesian inference.

“This confirms that the hybrid quadratic-linear interpolation captures the nonlinear dependence on M 2 with sufficient accuracy for our Bayesian inference.”

This represents the most stringent constraint obtained to date from cluster-scale collisions.

“This represents the most stringent constraint obtained to date from cluster-scale collisions.”

From mock recovery tests, a sample of ∼100 double radio relic systems would tighten the limit to the ∼0.1 cm^2 g^-1 level, at which velocity-dependent models become meaningfully distinguishable.

“From our mock recovery tests (Figure 4), a sample of ∼100 double radio relic systems would tighten this limit to the ∼0.1 cm 2 g -1 level, at which velocity-dependent models become meaningfully distinguishable.”

The method provides the most stringent and marginalized constraint on the dark matter self-interaction cross section to date from cluster-scale collisions.

“providing the most stringent and marginalized constraint on the DM selfinteraction cross section to date from cluster-scale collisions”

• Also stated in: § 3.2. Constraints on σ/m · plain

  “This is the most stringent constraint on the SICS derived from cluster-scale collisions to date.”

A lack of observed offset does not necessarily imply a lack of self-interaction.

“This strong gravitational coupling means that a lack of observed offset does not necessarily imply a lack of selfinteraction, leading to significantly weaker and more ambiguous constraints than previously assumed.”

The shock-to-shock distance, traced by double radio relics, can serve as a merger chronometer that time-stamps the post-pericenter dynamical phase.

“In this paper, we overcome these obstacles by introducing the shock-to-shock distance, traced by double radio relics, as a merger chronometer that time-stamps the post-pericenter dynamical phase.”

By jointly fitting the observed d_h and d_s against simulation predictions, we can isolate the SICS-dependent suppression of the halo separation.

“By jointly fitting the observed d h and d s against our simulation predictions, we can isolate the SICS-dependent suppression of the halo separation.”

Internal oscillation of galaxies can produce transient offsets driven by local gravitational dynamics rather than by dark matter self-interaction.

“This internal oscillation can produce transient offsets driven by local gravitational dynamics rather than by dark matter self-interaction.”

The validation confirms that the method can achieve high precision by combining multiple systems.

“This validation confirms that our method can achieve high precision by combining multiple systems, and motivates our strategy of applying the joint analysis to the 11 gold mergers in Section 3.”

The constraining power arises from the combination of 11 independent merger systems, each contributing an individual posterior that is marginalized over seven nuisance parameters: the two halo masses, gas profile slope, collision velocity, impact parameter, merger phase, and viewing angle.

“The constraining power arises from the combination of 11 independent merger systems, each contributing an individual posterior that is marginalized over seven nuisance parameters: the two halo masses, gas profile slope, collision velocity, impact parameter, merger phase, and viewing angle.”

The relic pair directly marks the positions of both shock fronts, enabling a clean measurement of the shock-to-shock separation.

“and the relic pair directly marks the positions of both shock fronts, enabling a clean measurement of the shock-to-shock separation.”

The joint constraint is a genuine ensemble property rather than the imprint of any individual cluster.

“demonstrating that the result is a genuine ensemble property rather than the imprint of any individual cluster”

The accumulation of independent data points effectively averages out merger-to-merger variations in nuisance parameters such as merger phase and viewing angle.

“While a single cluster collision provides only marginal information, the accumulation of independent data points effectively averages out the merger-to-merger variations in nuisance parameters such as merger phase and viewing angle.”

This is the first constraint from cluster collisions that fully marginalizes over mass uncertainty, viewing angle, collision speed, merger phase, impact parameter, and gas profile slope.

“This is the first constraint from cluster collisions that fully marginalizes over mass uncertainty, viewing angle, collision speed, merger phase, impact parameter, and gas profile slope.”

The observational constraint on σ/m relies on idealized simulations that do not capture physical complexities such as surrounding large-scale structures, halo substructures, and mass accretion.

“For our constraint on σ/m with observations, we rely on idealized simulations, which do not capture physical complexities such as surrounding large-scale structures, halo substructures, and mass accretion.”

• AssumptionRests on an assumption: The idealized simulations used for the observational constraint adequately capture the relevant merger dynamics despite neglecting large-scale structure, substructure, and mass accretion.

The projected halo-to-halo distance follows d_h2D = d_h3D cos α.

“The projected halo-to-halo distance is straightforward and can be modeled following a standard trigonometric projection:”

In a typical cluster merger, the hot intracluster medium experiences ram pressure stripping and lags behind the collisionless galaxies and dark matter halos.

“In a typical cluster merger, the hot intracluster medium (ICM) experiences ram pressure stripping and lags behind the collisionless galaxies and DM halos.”

The ratio of shock-to-shock distance to halo-to-halo distance translates directly into a constraint on the self-interaction cross section.

“Because the propagation speed of merger shocks is nearly independent of the SICS, while the halo-to-halo distance is depressed by SIDM-induced drag, the ratio of the two distances translates directly into a constraint on σ/m.”

• Also stated in: § 2.1. Cluster Collision Simulation Setup · boosted

  “The halo-to-halo distance relative to the shock-to-shock distance is therefore a robust probe of self-interaction strength.”

Because galaxies remain gravitationally bound to the dark matter halos, which provide the dominant contribution to the local potential, they tend to stay centered within the halo despite the momentum-transfer drag acting on the dark matter particles.

“Because galaxies remain gravitationally bound to the dark matter halos, which provide the dominant contribution to the local potential, they tend to stay centered within the halo despite the momentum-transfer drag acting on the dark matter particles.”

The method recovers a known input cross section for a single mock merger under idealized tight priors.

“Under these idealized conditions, the resulting posterior distribution for σ/m is sharply peaked at the true input value (blue curve in Figure 3).”