The 20 GeV Galactic Halo Excess: Pixel-Level Confirmation and Consistency with Sub-TeV WIMP Annihilation

A recent analysis of 15 years of Fermi-LAT data reported a spherically symmetric, halo-like component of the Galactic diffuse emission that peaks near 20GeV. We independently reproduce this cell-aggregated analysis, then extend it to a pixel-level likelihood on the native 0.125 maps, adding energy-dependent point-spread-function forward folding and masking bright sources. Both methods replicate the 20GeV halo spectrum, with the pixel-level normalisation about 20% above the cellwise fit across NFW emissivity scalings rho p, p in 1,2,2.5. This 20GeV halo is a high-latitude feature, distinct from the inner-Galaxy excess, and consistent with sub-TeV dark matter (WIMP) annihilation. It is centrally concentrated, strongly disfavouring extragalactic emission. Fitting prompt s-wave annihilation spectra, best-fit masses are m chi 0.55 TeV (W+W-) and 0.72 TeV (b) with sigma v 1 times10-24 cm cubed,s-1, in about!4 - 5 times tension with dwarf spheroidal galaxy limits. However, accounting for foreground modelling and J-factor systematic uncertainties widens the tension window to R 1.6 - 9.3, leaving the s-wave interpretation viable. To close the tension, we consider alternative particle dark matter models. p-wave annihilation misses relic abundance constraints by about!7 orders of magnitude. A decay interpretation evades dwarf limits but is disfavoured by the isotropic gamma-ray background. The only viable velocity structure consistent with dwarf limits, present-day halo rates, and relic density is low-velocity-enhanced annihilation (resonant Sommerfeld or Breit-Wigner). This supplies the required about!45 times boost from a thermal relic. Fully resolving the dwarf tension requires a fine-tuned resonance peaking at the halo velocity and falling for colder systems.