Cosmic Ray Anomalies from the MSSM?

TL;DR

This paper investigates whether specific MSSM models can explain cosmic ray anomalies, like the positron excess, with small boost factors and consistent gamma-ray constraints, highlighting a bino-like LSP scenario.

Contribution

It identifies a subclass of MSSM models with bino-like LSP and tau pair annihilation that better fits cosmic ray data with moderate boost factors, compared to traditional models.

Findings

A bino-like LSP MSSM model fits CR data well with boost ~100.

Models with lightest stau close in mass to LSP are favored.

Significant improvement in fit compared to non-SUSY models.

Abstract

The recent positron excess in cosmic rays (CR) observed by the PAMELA satellite may be a signal for dark matter (DM) annihilation. When these measurements are combined with those from FERMI on the total ($e^++e^-$) flux and from PAMELA itself on the $\bar p/p$ ratio, these and other results are difficult to reconcile with traditional models of DM, including the conventional mSUGRA version of Supersymmetry even if boosts as large as $10^{3-4}$ are allowed. In this paper, we combine the results of a previously obtained scan over a more general 19-parameter subspace of the MSSM with a corresponding scan over astrophysical parameters that describe the propagation of CR. We then ascertain whether or not a good fit to this CR data can be obtained with relatively small boost factors while simultaneously satisfying the additional constraints arising from gamma ray data. We find that a specific subclass of MSSM models where the LSP is mostly pure bino and annihilates almost exclusively into $\tau$ pairs comes very close to satisfying these requirements. The lightest $\tilde \tau$ in this set of models is found to be relatively close in mass to the LSP and is in some cases the nLSP. These models lead to a significant improvement in the overall fit to the data by an amount $\Delta \chi^2 \sim 1/$dof in comparison to the best fit without Supersymmetry while employing boosts $\sim 100$. The implications of these models for future experiments are discussed.