Implication of neutrino backgrounds on the reach of next generation dark matter direct detection experiments

As direct dark matter experiments continue to increase in size, they will become sensitive to neutrinos from astrophysical sources. For experiments that do not have directional sensitivity, coherent neutrino scattering from several sources represents an important background to understand, as it can almost perfectly mimic an authentic weakly interacting massive particle (WIMP) signal. Here we explore in detail the effect of neutrino backgrounds on the discovery potential of WIMPs over the entire mass range of 500 MeV to 10 TeV. We show that, given the theoretical and measured uncertainties on the neutrino backgrounds, direct detection experiments lose sensitivity to light ($\sim 10\mathrm{GeV}$) and heavy ($\sim 100\mathrm{GeV}$) WIMPs with a spin-independent cross section below ${10}^{-45}$ and ${\text{10}}^{-49}{\mathrm{cm}}^2$, respectively.

Figure 1

Relevant neutrino fluxes which are backgrounds to direct dark matter detection experiments: Solar, atmospheric, and diffuse supernovae [8].

Figure 2

Neutrino-induced nuclear recoil spectra for the different neutrino sources, for a Ge target (left) and a Xe target (right).

Figure 3

Number of neutrino-induced nuclear recoils per ton-year for a Ge target (left) and Xe target (right) as a function of the energy threshold. Note that we have considered an upper limit on the nuclear recoil energy range of 100 keV.

Figure 4

Left (right) panel shows the energy spectra of the most relevant neutrino backgrounds for a Ge- (Xe-) type detector. Shown are a WIMP signal (black solid line), the total CNS (coherent neutrino scattering) background contribution (blue dashed line), standard electroweak neutrino-electron interaction (red line), and the contribution from the neutrino magnetic moment (cyan lines). Dashed red and cyan lines (dark and light) correspond to the consideration of an electron rejection factor of 99.5% and ${10}^5$ for a XENON-like and Ge-based CDMS-like experiment respectively. Dark and light cyan curves correspond to the experimental and theoretical upper limits on the neutrino magnetic moment respectively.

Figure 5

Left: Set of derived background-free sensitivity curves for exposures that attain one neutrino event, for different thresholds from 0.001 (purple) to 100 keV (red) in logarithmic steps. The black line is constructed by joining the best sensitivity for each mass, and represents a one neutrino event contour line in the WIMP-nucleon cross section vs WIMP mass plane. Right: Background-free exclusion limits (solid lines) for four different Xe-based experiments with threshold of 10 eV, 500 eV, 5 keV, and 10 keV and exposures of 10 kg-years, 2 ton-years, 100 ton-years, and 5,000 ton-years respectively. Also shown in dashed lines are the neutrino isoevent contour lines for 18.5 (blue), 657 (green), 4.5 (red), and 154 (magenta) events.

Figure 6

Distributions of the maximum likelihood of the CNS background under the WIMP only hypothesis for a Ge target (left) and a Xe target (right). The different intensities of colors correspond to the energy threshold considered, from light to dark: 1 eV, 10 eV, 100 eV, 1 keV, 2.5 keV, 5 keV, 7.5 keV, and 10 keV. These distributions have been computed by adjusting the experiment exposure such that we have a total of about 500 expected neutrino events for each different energy threshold and target.

Figure 7

Comparison between the nuclear recoil event rate as a function of energy from the CNS background (blue dashed line) and the best fit WIMP models for the different thresholds, deduced from Fig. 6, in the case of a Ge target (left) and Xe target (right). The different color intensities and thicknesses of the lines are from light thin to dark thick: 1 eV, 10 eV, 100 eV, 1 keV, 2.5 keV, 5 keV, 7.5 keV, and 10 keV.

Figure 8

Left: Distributions of the maximum likelihood of the CNS background under the WIMP only hypothesis for each neutrino component, considering a Xe target nucleus and no energy threshold. Right: Distributions of the maximum likelihood of the CNS background under the WIMP only hypothesis for six different target nuclei: Xe, Ge, Ar, Si, Ne, F with a common energy threshold of 1 keV. Distributions shown on the left and right panels have been computed by adjusting the experiment exposure such that we have a total of about 500 expected neutrino events for each configuration.

Figure 9

Discovery limits for Ge target (left panels) and Xe target (right panels), and for 0.1 keV threshold (top panels) and 2 keV thresholds (bottom panels). The different line styles, solid, dashed, and dotted-dashed respectively correspond to the three exposures 1, 0.1, and 0.01 ton-years. These discovery limits have been computed considering only the two dominant neutrino contributions: ${}^8\mathrm{B}$ and $hep$ that both have a 16% uncertainty on their integrated flux. Also shown are the exclusion sensitivity limits for each of the different exposures with a color intensity scaling with the exposure from light to dark gray. The bumpiness of the discovery limits is due to the finite size of the Monte Carlo samples (500 iterations) inducing a 5% to 10% statistical fluctuation over the WIMP mass range.

Figure 10

Discovery limits (solid lines) for arbitrarily large exposure experiments and for three different target nuclei: Xe (green curves), Ar (cyan curves), and Ne (purple curves). The short dashed lines correspond to the background-free exclusion sensitivity of such experiments. The green long dashed line corresponds to the case where we have considered a finite electron recoil rejection factor of 99.5% for a XENON-like detector without neutrino magnetic moment enhancement.

Figure 11

Discovery limits for Ar, Ge, and Xe vs exposure. Left: For $6\mathrm{GeV}/{\mathrm{c}}^2$ WIMPs, the Ge, Xe, and Ar exposures required to obtain 100 neutrino events are 240, 130, and 430 kg-years. Right: For $100\mathrm{GeV}/{\mathrm{c}}^2$ WIMPS, the Ge, Xe, and Ar exposures required to obtain one neutrino event are 32.5, 21.5, and 98 ton-years.

Figure 12

Left: Neutrino isoevent contour lines (long dash orange) compared with current limits and regions of interest. The contours delineate regions in the WIMP-nulceon cross section vs WIMP mass plane which for which dark matter experiments will see neutrino events (see Sec. 3d). Right: WIMP discovery limit (thick dashed orange) compared with current limits and regions of interest. The dominant neutrino components for different WIMP mass regions are labeled. Progress beyond this line would require a combination of better knowledge of the neutrino background, annual modulation, and/or directional detection. We show 90% confidence exclusion limits from DAMIC [46] (light blue), SIMPLE [47] (purple), COUPP [48] (teal), ZEPLIN-III [49] (blue), EDELWEISS standard [50] and low threshold [51] (orange), CDMS II Ge standard [52], low threshold [53] and CDMSlite [54] (red), XENON10 S2 only [55] and XENON100 [2] (dark green), and LUX [56] (light green). The filled regions identify possible signal regions associated with data from CDMS-II Si [1] (light blue, 90% C.L.), CoGeNT [57] (yellow, 90% C.L.), DAMA/LIBRA [58] (tan, 99.7% C.L.), and CRESST [59] (pink, 95.45% C.L.) experiments. The light green shaded region is the parameter space excluded by the LUX Collaboration.