Phenomenological minimal supersymmetric standard model dark matter searches on ice

R. C. Cotta, K. T. K. Howe, J. L. Hewett, and T. G. Rizzo

Phys. Rev. D 85, 035017 – Published 16 February 2012

Abstract

We explore the capability of the IceCube/DeepCore array to discover signal neutrinos resulting from the annihilations of supersymmetric weakly interacting massive particles that may be captured in the solar core. In this analysis, we use a previously generated set of $\sim 70 k$ model points in the 19-dimensional parameter space of the phenomenological minimal supersymmetric standard model (pMSSM) which satisfy existing experimental and theoretical constraints. Our calculations employ a realistic estimate of the IceCube/DeepCore effective area that has been modeled by the IceCube collaboration. We find that a large fraction of the pMSSM models are shown to have significant signal rates in the anticipated IceCube/DeepCore 1825 d data set, including some prospects for an early discovery. Many models where the lightest supersymmetric particle only constitutes a small fraction of the total dark matter relic density are found to have observable rates. We investigate in detail the dependence of the signal neutrino fluxes on the lightest-supersymmetric-particle mass, weak eigenstate composition, annihilation products and thermal relic density, as well as on the spin-independent and spin-dependent scattering cross sections. Lastly, we compare the model coverage of IceCube/DeepCore to that obtainable in near-future direct-detection experiments and to pMSSM searches at the 7 TeV LHC.

10 More

Received 25 May 2011

DOI:https://doi.org/10.1103/PhysRevD.85.035017

Authors & Affiliations

R. C. Cotta¹, K. T. K. Howe^1,2, J. L. Hewett¹, and T. G. Rizzo¹

¹SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, California 94025 USA
²Physics Dept, SITP, 382 Via Pueblo Mall, Varian Lab Stanford University, Stanford California, 94305-4060, USA

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 85, Iss. 3 — 1 February 2012

Reuse & Permissions

Access Options

Article Available via CHORUS

Download Accepted Manuscript

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
We display points representing models in our flat-prior pMSSM model set in the detected neutrino flux, $Φ_{ν}^{D}$ [see Eq. (7)], vs LSP mass plane. Grey points represent all of the models in this set while orange points denote pMSSM models which are out of capture/annihilation equilibrium according to Eq. (5), and blue points represent models whose LSPs form substantially all of dark matter, $Ω_{{\tilde{χ}}_{1}^{0}} h^{2} > 0.10$ ( $R \approx 1$ ).Reuse & Permissions
Figure 2
We display points representing models in our flat-prior pMSSM model set in the detected neutrino flux, $Φ_{ν}^{D}$ [see Eq. (7)], vs $R$ [see Eq. (4)] plane. Grey points represent all of the models in this set, orange points denote pMSSM models which are out of capture/annihilation equilibrium according to Eq. (5) and blue points represent models whose LSPs form substantially all of dark matter, $Ω_{{\tilde{χ}}_{1}^{0}} h^{2} > 0.10$ ( $R \approx 1$ ). $R \approx 1$ models are, of course, situated at the very right-hand edge this figure.Reuse & Permissions
Figure 3
We display points representing models in our flat-prior pMSSM model set in the detected neutrino flux, $Φ_{ν}^{D}$ [see Eq. (7)], vs $σ_{S I, p} \times R$ plane. Grey points represent all of the models in this set, orange points denote pMSSM models which are out of capture/annihilation equilibrium according to Eq. (5) and blue points represent models whose LSPs form substantially all of dark matter, $Ω_{{\tilde{χ}}_{1}^{0}} h^{2} > 0.10$ ( $R \approx 1$ ). We note that we include all models which satisfied the direct-detection limits that were in place when these models were generated [14] (wherein a factor of 4 error was allowed for nuclear form factor uncertainties). We will consider the effect of recent direct-detection limits (including the very recent XENON100 limit [52]) in Sec. 4c.Reuse & Permissions
Figure 4
We display points representing models in our flat-prior pMSSM model set in the detected neutrino flux, $Φ_{ν}^{D}$ [see Eq. (7)], vs $σ_{S D, p} \times R$ plane. Grey points represent all of the models in this set, orange points denote pMSSM models which are out of capture/annihilation equilibrium according to Eq. (5) and blue points represent models whose LSPs form substantially all of dark matter, $Ω_{{\tilde{χ}}_{1}^{0}} h^{2} > 0.10$ ( $R \approx 1$ ). We note that $σ_{S D, p} \times R$ is the quantity which is most strongly correlated with high detected-neutrino rates.Reuse & Permissions
Figure 5
Comparison of the basic results for our flat-prior pMSSM model set with those for our log-prior pMSSM model set. Flat-prior models are represented by grey points while log-prior models are represented by black points. There are $\sim 68.4 k$ models in the flat-prior set and $\sim 2.9 k$ models in the log-prior set. We note the general similarity between these two model samples.Reuse & Permissions
Figure 6
Using the approximate exclusion criterion described in Sec. 3b, we compare histograms of models that are expected to be excluded by IC/DC (red) against the full model set (grey) as functions of the LSP relic density and of the ratio $2 Γ_{a} / C_{c}$ [recall Eq. (5)]. As described in the text, exclusion performance is not strongly correlated with LSP relic density, and we see that IC/DC may be able to discover LSPs with $R ≪ 1$ . In contrast to this, we see that essentially none of the models that can be excluded are out-of-equilibrium.Reuse & Permissions
Figure 7
We display points representing models in our flat-prior pMSSM model set in the detected neutrino flux, $Φ_{ν}^{D}$ , vs solar capture rate, $C_{c}$ , plane. Purple points represent models that are in capture/annihilation equilibrium according to Eq. (5), and orange points denote pMSSM models which are out of capture/annihilation equilibrium.Reuse & Permissions
Figure 8
We display points representing models in our flat-prior pMSSM model set in the capture rate, $C_{c}$ , vs elastic scattering, $σ_{S I, p} \times R$ or $σ_{S D, p} \times R$ planes. Grey points represent all of the models in this set, orange points denote pMSSM models which are out of capture/annihilation equilibrium according to Eq. (5) and blue points represent models whose LSPs form substantially all of dark matter, $Ω_{{\tilde{χ}}_{1}^{0}} h^{2} > 0.10$ ( $R \approx 1$ ). The strong correlation between $C_{c}$ and $σ_{S D, p} \times R$ explains the strong correlation observed between $Φ_{ν}^{D}$ and $σ_{S D, p} \times R$ (Fig. 4). The strong correlation between $C_{c}$ and $σ_{S D, p} \times R$ (and relatively weak correlation between $C_{c}$ and $σ_{S I, p} \times R$ ) is explained by relative importance of spin-dependent scattering in solar WIMP capture and the fact that $σ_{S D, p} / σ_{S I, p} > 10^{3}$ for most of the models in our set (cf. Figs. 9, 10).Reuse & Permissions
Figure 9
SUSY-model dependence in the $σ_{S D, p} \times R$ vs $σ_{S I, p} \times R$ plane. Flat-prior models that are estimated to be excluded in the IC/DC search are displayed as red points, and those that are not expected to be excluded are displayed as blue points.Reuse & Permissions
Figure 10
We describe the distribution of spin-dependent and spin-independent WIMP-proton cross sections in our flat- and log-prior model sets. We display points in the detected neutrino flux, $Φ_{ν}^{D}$ , vs $σ_{S D, p} / σ_{S I, p}$ plane. We also histogram the ratio $σ_{S D, p} / σ_{S I, p}$ for both model sets. Flat-prior models are described by grey points/bars while log-prior models are described by black points/bars. We observe that most of the models in both sets have $σ_{S D, p} / σ_{S I, p} > 10^{3}$ (the small subset of models for which $σ_{S D, p} < σ_{S I, p}$ are seen to have contributions to $σ_{S D, p}$ from $Z^{0}$ and squark diagrams which largely cancel).Reuse & Permissions
Figure 11
We display points representing our flat-prior models in the $(DetectedFlux) / (RawFlux)$ [Eq. (8)] vs LSP mass plane. The full flat-prior model set is displayed as grey points and models whose annihilations occur predominantly through a given final-state channel are overlaid in other colors. Models with $⟨ σ v ⟩_{t \bar{t}} / ⟨ σ v ⟩ > 0.85$ (cyan), with $⟨ σ v ⟩_{τ \bar{τ}} / ⟨ σ v ⟩ > 0.70$ (green), with $⟨ σ v ⟩_{b \bar{b}} / ⟨ σ v ⟩ > 0.93$ (red), with $⟨ σ v ⟩_{W^{+} W^{-}} / ⟨ σ v ⟩ > 0.99$ (blue) and with $⟨ σ v ⟩_{Z^{0} Z^{0}} / ⟨ σ v ⟩ > 0.42$ (magenta) are shown. Purities are chosen to obtain subsets of models of similar size (in our model set there is a maximum purity for annihilations into the $Z^{0} Z^{0}$ final state, $⟨ σ v ⟩_{Z^{0} Z^{0}} / ⟨ σ v ⟩ \leq 0.445$ , as the Higgsino-like LSPs which annihilate well to $Z^{0} Z^{0}$ via ${\tilde{χ}}_{2}^{0}$ exchange also annihilate well to $W^{+} W^{-}$ via ${\tilde{χ}}_{1}^{+}$ exchange).Reuse & Permissions
Figure 12
We display points representing our flat-prior models in the $(DetectedFlux) / (RawFlux)$ [Eq. (8)] vs LSP mass plane. The full flat-prior model set is displayed as grey points, and models categorized according to LSP eigenstate composition are overlaid in other colors. By convention, our LSPs are described in terms of their neutralino mass matrix entries as: ${\tilde{χ}}_{1}^{0} = Z_{11} \tilde{B} + Z_{12} {\tilde{W}}^{3} + Z_{13} {\tilde{H}}_{1}^{0} + Z_{14} {\tilde{H}}_{2}^{0}$ . Higgsino models are defined as having $(| Z_{13} |^{2} + | Z_{14} |^{2}) > 0.99$ and are displayed here in green. Wino models are defined as having $| Z_{12} |^{2} > 0.99$ and are displayed here in blue. Bino models are defined as having $| Z_{11} |^{2} > 0.99$ and are displayed here in red. Mixed models are defined as having $| Z_{11} |^{2}$ , $| Z_{12} |^{2}$ and $(| Z_{13} |^{2} + | Z_{14} |^{2})$ , all $< 0.8$ , and are displayed here in magenta.Reuse & Permissions
Figure 13
Detected neutrino flux histograms illustrating the dependence of $Φ_{ν}^{D}$ on final-state annihilation channel. The full flat-prior model set is described in the grey histogram while subsets of these models whose annihilations occur predominantly through a given final-state channel are displayed in other colors. Models with $⟨ σ v ⟩_{t \bar{t}} / ⟨ σ v ⟩ > 0.85$ (cyan), with $⟨ σ v ⟩_{τ \bar{τ}} / ⟨ σ v ⟩ > 0.70$ (green), with $⟨ σ v ⟩_{b \bar{b}} / ⟨ σ v ⟩ > 0.93$ (red), with $⟨ σ v ⟩_{W^{+} W^{-}} / ⟨ σ v ⟩ > 0.99$ (blue) and with $⟨ σ v ⟩_{Z^{0} Z^{0}} / ⟨ σ v ⟩ > 0.42$ (magenta) are shown. Purities are chosen to obtain subsets of models of similar size (in our model set, there is a maximum purity for annihilations into the $Z^{0} Z^{0}$ final state, $⟨ σ v ⟩_{Z^{0} Z^{0}} / ⟨ σ v ⟩ \leq 0.445$ , as the Higgsino-like LSPs which annihilate well to $Z^{0} Z^{0}$ via ${\tilde{χ}}_{2}^{0}$ exchange also annihilate well to $W^{+} W^{-}$ via ${\tilde{χ}}_{1}^{+}$ exchange).Reuse & Permissions
Figure 14
Detected neutrino flux histograms illustrating the dependence of $Φ_{ν}^{D}$ on LSP eigenstate composition. By convention, our LSPs are described in terms of their neutralino mass matrix entries as: ${\tilde{χ}}_{1}^{0} = Z_{11} \tilde{B} + Z_{12} {\tilde{W}}^{3} + Z_{13} {\tilde{H}}_{1}^{0} + Z_{14} {\tilde{H}}_{2}^{0}$ . Higgsino models are defined as having $(| Z_{13} |^{2} + | Z_{14} |^{2}) > 0.99$ and are displayed here in green. Wino models are defined as having $| Z_{12} |^{2} > 0.99$ and are displayed here in blue. Bino models are defined as having $| Z_{11} |^{2} > 0.99$ and are displayed here in red. Mixed models are defined as having $| Z_{11} |^{2}$ , $| Z_{12} |^{2}$ and $(| Z_{13} |^{2} + | Z_{14} |^{2})$ , all $< 0.8$ , and are displayed here in magenta.Reuse & Permissions
Figure 15
Comparison of IC/DC and spin-independent direct-detection searches. We display all points in the flat-prior model set in grey, models that are estimated to be excluded by the IC/DC solar WIMP search in red and those which are not excluded in blue. Current experimental limits from the CDMS [50] and XENON100 [51, 52] Collaborations are displayed as red lines (CDMS 2010), black solid lines (XENON100 2010) and black dashed lines (XENON100 2011), respectively. Near-future projected experimental limits from LUX [53], SuperCDMS [54], COUPP 60 kg and COUPP 500 kg [55] are displayed as brown, magenta, orange, green dashed and green dotted lines, respectively.Reuse & Permissions
Figure 16
Comparison of IC/DC and spin-dependent direct-detection searches. We display all points in the flat-prior model set in grey, models that are estimated to be excluded by the IC/DC solar WIMP search in red and those which are not estimated to be excluded in blue. Current experimental limits from the AMANDA [46] and IceCube-22 [43] Collaborations are displayed as orange and magenta lines, respectively (with the assumption of soft or hard channel annihilations represented by dashed or solid lines, respectively). Near-future projected experimental limits from the COUPP [55, 56] 4 kg, 60 kg and 500 kg searches in green solid, dashed and dotted lines, respectively. The IceCube/DeepCore limit estimated in Ref. 57 (assuming hard channel annihilations, DM which is in capture/annihilation equilibrium and DM which has $R = 1$ ) is displayed as a black line.Reuse & Permissions
Figure 17
Comparison of IC/DC and the ATLAS 4j0l ( $\sqrt{s} = 7 TeV$ , $1 {fb}^{- 1}$ and 50% systematic uncertainty assumed) searches. Here “PASS” and “FAIL” denote discovered or not discovered, respectively. We display all points in the flat-prior model set in grey, models that are estimated to be discoverable by both the IC/DC solar WIMP search and the ATLAS search are displayed in pink, those expected to be seen by IC/DC but missed in the ATLAS search in green, those expected to be missed by IC/DC but seen in the ATLAS search in purple and those which are estimated to be unobservable by either search are displayed in black.Reuse & Permissions

Physical Review D

covering particles, fields, gravitation, and cosmology