Velocity-Dependent P-Wave Annihilation

• Velocity-dependent p-wave annihilation is a dark matter process where the cross section scales with the square of the relative velocity, reducing signals in low-velocity environments.
• Astrophysical modeling shows that high-velocity regions, such as galaxy clusters and black hole spikes, yield enhanced indirect detection signals compared to dwarf spheroidals.
• Sommerfeld enhancement and resonant effects can boost p-wave annihilation, providing promising avenues for gamma-ray searches and tighter dark matter constraints.

Velocity-dependent p-wave annihilation describes a class of dark matter (DM) models in which the annihilation cross section is suppressed by the square of the pairwise relative velocity, leading to indirect signals with strong environmental and kinematic dependence. This phenomenon has significant consequences for indirect detection prospects, target selection, astrophysical modeling, and the particle physics constraints that can be derived from observational data.

1. Theoretical Foundation: Cross Section and Velocity Scaling

In nonrelativistic partial-wave expansion, the annihilation cross section of DM particles can be parametrized as

$\sigma v = a + b v^2 + \mathcal{O}(v^4)$

Here, $a$ corresponds to $s$-wave (velocity-independent) processes and $b v^2$ to $p$-wave (velocity-suppressed) processes. In many models, notably for Majorana DM or fermionic DM annihilating to light fermions, the $s$-wave is either forbidden or helicity-suppressed, making $p$-wave dominant (Kadota et al., 2021, Shelton et al., 2015).

For pure $p$-wave annihilation, this yields

$$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$

with $b$ a model-dependent constant reflecting the microscopic coupling structure. At freeze-out, $a$0 is large ($a$1), but for DM in galaxies today ($a$2 in the Milky Way (MW)), $a$3 is suppressed by $a$4 relative to freeze-out (Shelton et al., 2015, Kadota et al., 2021).

P-wave annihilation also admits quantum mechanical effects such as Sommerfeld enhancement, where the cross section further depends on velocity through long-range mediator-induced resonances, leading to highly nontrivial velocity scaling and possible resonance-induced boosts in indirect detection signals (Ding et al., 2021, Beneke et al., 2024).

2. Astrophysical J-Factor Generalization and Environmental Sensitivity

Astrophysical predictions in indirect searches depend on the J-factor, the integral along the line of sight (LOS) of the square of the DM density: $a$5 For $a$6-wave models, the relevant factor receives extra weight from the squared local DM velocity dispersion: $a$7 Typically, the local velocity distribution is approximated as Maxwell–Boltzmann, yielding $a$8 (Boddy et al., 2018, McKeown et al., 2021, Blanchette et al., 2022). Thus, the $a$9-wave J-factor is suppressed in environments with low velocity dispersion, as in dwarf spheroidals (dSphs), but can be greatly enhanced in hot environments such as galaxy clusters or near supermassive black holes (SMBHs), where velocity dispersions are much higher (Shelton et al., 2015, Kostić et al., 2023, McKeown et al., 2021).

Hydrodynamical simulations (FIRE-2, Auriga, APOSTLE) demonstrate that baryonic physics can systematically amplify the central $s$0 values and thus the $s$1-wave J-factor, especially in the central MW ($s$2–$s$3 at $s$4 from the GC compared to dark-matter-only (DMO) runs) (McKeown et al., 2021). The inclusion of gas cooling, star formation, and feedback increases the velocity dispersion and modifies the signal's morphology (flatter and rounder emission profile).

3. Consequences for Target Selection: Dwarfs, Clusters, and Black Hole Spikes

Because $s$5 varies strongly among environments, the prospects for indirect detection are highly target-dependent:

• Dwarf Spheroidals: $s$6–10 km/s, leading to $s$7 suppressed by $s$8. Limits on $s$9-wave cross sections from dwarfs are typically 2–3 orders of magnitude weaker than from the MW or clusters, failing to reach the canonical thermal benchmark ($b v^2$0–$b v^2$1 cm$b v^2$2/s) (Boddy et al., 2019, Zhao et al., 2017, Zhao et al., 2016). Even with stacking, bounds reach only $b v^2$3 cm$b v^2$4/s for $b v^2$5 GeV (Boddy et al., 2019).
• Milky Way and Large Extragalactic Halos: For velocity dispersions $b v^2$6–$b v^2$7 km/s, $b v^2$8 increases by $b v^2$9 over dSphs. Full-sky and stacking analyses of local volumes yield limits of $p$0 cm$p$1/s at 95% CL for $p$2 GeV, the strongest yet for $p$3-wave models, though still above the thermal value (Kostić et al., 2023, Baxter et al., 2022, Lacroix et al., 2022).
• Galaxy Clusters: Very high velocity dispersions ($p$4 km/s) make them exceptionally bright p-wave targets. Subhalo boosts are smaller (up to $p$5) than for s-wave, but $p$6 can reach $p$7 GeV$p$8 cm$p$9, shifting focus for next-generation searches (Lacroix et al., 2022).
• Black Hole Spikes: The highest $s$0 occurs near SMBHs (e.g., Sgr A*), where steep density spikes and increased velocity dispersions ($s$1 at $s$2–$s$3 pc) make the p-wave signal potentially observable as a bright gamma-ray point source. Fermi-LAT limits already constrain the $s$4-wave cross section down to the thermal relic region for favorable spike parameters (Shelton et al., 2015, Johnson et al., 2019).

4. Impact on Gamma-Ray Signals and Model Constraints

The $s$5-wave velocity suppression profoundly impacts both the amplitude and morphology of cosmic gamma-ray signatures:

• Galactic Center: Compared to s-wave, the innermost regions of the GC are more suppressed, with $s$6 at $s$7 from the center for NFW-like profiles. Baryonic effects in FIRE-2 further amplify $s$8 at intermediate angles by up to factors of $s$9–$p$0, providing a window for detection near the thermal target with improved modeling (McKeown et al., 2021, Boddy et al., 2018).
• Extragalactic Background and Angular Power: The addition of the $p$1-wave term modifies the overall normalization of the extragalactic gamma-ray background. Significant $p$2-wave-induced shape changes (hardening) require $p$3, a regime not typically realized in the MSSM but present in scenarios with a highly suppressed s-wave (1009.3530, Campbell et al., 2011). For thermal relics, the relic density constraint forces the s-wave component to nearly vanish when $p$4-wave dominates, leading to amplitude suppression by $p$5. Observable shape modifications would point to non-thermal DM production (1009.3530, Campbell et al., 2011).
• Subhalos and Boost Factors: In contrast to s-wave, $p$6-wave annihilation in subhalos is comparatively insignificant due to low $p$7, making the smooth halo the dominant emission source; cluster subhalos can still provide boosts up to $p$8 (Piccirillo et al., 2022, Lacroix et al., 2022).
• SMBH-Induced Spikes: Thermal $p$9-wave models can be probed with $p$0-ray data from the Galactic Center. For an adiabatic spike, Fermi-LAT observations exclude $p$1 cm$p$2/s for $p$3–$p$4 GeV (Shelton et al., 2015, Johnson et al., 2019). Spectral searches for box/line features from cascade annihilation can yield even stronger constraints.

5. Resonant and Non-Perturbative Effects: Sommerfeld Enhancement

When dark matter couples to a light mediator, Sommerfeld enhancement can amplify the $p$5-wave cross section nontrivially. In a Yukawa potential, p-wave Sommerfeld factors exhibit an off-resonant $p$6 scaling and, near quasi-bound states, Breit–Wigner resonance spikes at specific velocities: $p$7 where $p$8, $p$9 are the energy and width of the quasi-bound state. At resonance, $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$0 in galactic and dSph contexts can be enhanced by $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$1–$$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$2, dramatically tightening indirect detection constraints and potentially shifting or closing otherwise allowed parameter space (Beneke et al., 2024, Ding et al., 2021).

Implications include:

• Simultaneous accommodation of the observed cosmic-ray positron excess and the DM relic density, while evading gamma-ray and CMB constraints, is possible in narrow resonance parameter bands (Ding et al., 2021).
• Even after velocity averaging, Sommerfeld resonance provides substantial boost in $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$3 for wino DM or light-mediator models (Beneke et al., 2024).

6. Constraints from Primordial Black Holes and Cosmological Observables

In mixed DM scenarios with primordial black holes (PBHs) and p-wave annihilating WIMPs, PBH-induced DM spikes can exceed the velocity suppression in compact regions, enhancing annihilation rates. However, the annihilation core is extremely compact, and the net gamma-ray or energy injection signal is still typically smaller than for s-wave. This leads to weaker constraints on the allowed PBH fraction by up to two orders of magnitude relative to s-wave, with bounds scaling weakly with PBH mass ($$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$4) (Kadota et al., 2021, Si et al., 1 Jan 2026). Observables such as the CMB $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$5-distortion parameter and the global 21-cm signal at cosmic dawn provide powerful complementary probes (Si et al., 1 Jan 2026).

7. Uncertainties, Systematics, and Prospects for Future Searches

Key uncertainties include:

• Astrophysical J-Factor Modeling: The primary uncertainty for $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$6-wave annihilation is the DM density profile; the velocity dispersion is more robust and generally constrained within $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$7 in simulations (Board et al., 2021, Blanchette et al., 2022).
• Subhalo and Halo Mass Functions: These contribute to systematic uncertainty in stacked or extragalactic analyses; the dominant emission is from the smooth halo in $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$8-wave models, minimizing substructure systematics (Kostić et al., 2023).
• Velocity Distribution Modeling: Approximations such as global Maxwell–Boltzmann fits and power-law relations between circular velocity and peak speed have proven accurate to $$\langle \sigma v \rangle_p = b\,\langle v^2 \rangle$$9 over broad ranges of galactic environments (Blanchette et al., 2022).
• Spectral and Angular Discriminants: $b$0-wave and $b$1-wave annihilation can produce similar morphologies when varying inner halo slopes, but are distinguishable at sub-degree angular scales around the GC. Current and next-generation $b$2-ray telescopes with high angular resolution and exposure are required to separate them conclusively (Boddy et al., 2018, McKeown et al., 2021, Baxter et al., 2022).
• Experimental Limits: State-of-the-art limits from galaxy clusters and extragalactic stacking analyses are 2–3 orders of magnitude stronger than dSph-based limits, but still above the canonical thermal relic value, indicating that $b$3-wave dark matter of this type remains an open possibility (Kostić et al., 2023, Vienneau et al., 16 Sep 2025).

---

Velocity-dependent $b$4-wave annihilation models fundamentally alter the phenomenology of indirect dark matter searches. The quadratic velocity suppression of the cross section makes such models difficult to probe in cold environments but highlights environments of high velocity dispersion—including SMBH-induced density spikes, galaxy clusters, and massive extragalactic halos—as optimal targets for indirect detection efforts. Ongoing improvement in simulation-based astrophysical modeling, combined with expanded $b$5-ray dataset sensitivity, continues to tighten constraints on velocity-suppressed dark matter models, with Sommerfeld-enhanced scenarios offering especially promising—and constraining—resonant signatures (McKeown et al., 2021, Shelton et al., 2015, Beneke et al., 2024, Kostić et al., 2023).