Ruling out dark matter interpretation of the galactic GeV excess by gamma-ray data of galaxy clusters
Recently, some very tight constraints of annihilating dark matter have been obtained from gamma-ray data of the Milky Way and Milky Way dwarf spheroidal satellite galaxies. In this article, we report that there are two excellent galaxy clusters (A2877 and Fornax) which can provide interesting constraints for annihilating dark matter. The lower limits of the dark matter mass for the thermal relic annihilation cross section are 25 GeV, 6 GeV, 130 GeV and 100 GeV respectively for the , , and channels. For some configuration of our working assumptions, our results improve the Fermi-LAT upper limits of annihilation cross sections by a factor of for wide ranges of dark matter mass for , and channels, and a factor of for channel with dark matter mass GeV. These limits basically rule out most of the existing popular dark matter interpretation of the GeV excess in the Milky Way.
In the past few years, gamma-ray observations of the Milky Way initiated a hot research area about dark matter annihilation. Many studies Abazajian (); Calore (); Calore2 (); Daylan (); Abazajian2 () point out that the GeV excess in the Milky Way can be best explained by dark matter annihilation with dark matter mass GeV via or channels. However, later gamma-ray observations of the Milky Way dwarf spheroidal satellite (MW dSphs) galaxies by the Fermi-Large Area Telescope (Fermi-LAT) put strong constraints on the annihilation cross sections for different channels Ackermann (); Albert (). These limits give a certain tension for the dark matter interpretation of the GeV excess in the Milky Way. Besides, some recent studies suggest that the GeV excess in the Milky Way can be explained by pulsars’ emission Yuan (); Petrovic (); Gaggero (); Brandt (); Bartels2 (); Ploeg () or depletion of gamma rays in molecular clouds Boer (). Furthermore, recent results of Fermi-LAT gamma-ray detection strongly disfavour the dark matter interpretation of the GeV excess Ajello (). Therefore, the GeV excess in the Milky Way becomes a very controversial issue in dark matter astrophysics now.
Although most of the gamma-ray observations are now focusing on the Milky Way and the MW dSphs galaxies, some studies start to realize that galaxy clusters are also good candidates to constrain annihilating dark matter. For example, Ando & Nagai (2012) Ando () find that the gamma-ray data from the Fornax cluster is quite good to constrain the parameters of dark matter annihilation because its cosmic ray emission is not very significant. The cross section constrained by using the Fornax cluster data is cm s Ando (). Besides, Huang et al. (2012) Huang () use three years of Fermi-LAT data of several nearby galaxy clusters to constrain annihilating dark matter. The upper limits of annihilation cross section are also of the order cm s. Later studies of gamma-ray emission from galaxy clusters and galaxy groups give more stringent constraints on dark matter parameters. For example, Ackermann et al. (2015) Ackermann3 () and Lisanti et al. (2017) Lisanti () exclude the thermal relic annihilating dark matter via channel for GeV and GeV respectively. Also, using new Fermi-LAT data of nearby galaxy clusters can provide stringent constraints for gamma-ray line feature of annihilating dark matter Liang (). Besides gamma-ray data, by using radio data of galaxy clusters, the annihilation cross sections for some channels can be constrained down to cm s Storm (); Storm2 (); Beck (), though the uncertainties due to magnetic field strength are somewhat significant. Overall speaking, these limits are quite close to or even tighter than some of the parameters for the dark matter interpretation of the GeV excess in the Milky Way.
In this article, by using the latest gamma-ray data of galaxy clusters from Fermi-LAT, we show that the A2877 and Fornax clusters can give very stringent constraints for annihilating dark matter. Our results improve the current limits of annihilation cross sections for wide ranges of dark matter mass. These limits are tighter than the parameters involved for the dark matter interpretation of the GeV excess in the Milky Way.
Generally speaking, there are two different kinds of gamma-ray emissions for annihilating dark matter in galaxy clusters. One is ‘direct emission’ and the other is ‘indirect emission’. ‘Direct emission’ means the gamma-ray photons produced when dark matter particles annihilate each other. Since dark matter annihilation also produces electron and positron pairs, these electron and positron pairs can interact with the thermal electrons in galaxy clusters via Bremsstrahlung cooling. This process would emit gamma-ray photons if the energy of the electrons or positrons is high enough. This kind of emission is regarded as ‘indirect emission’.
ii.1 Direct emission
The total number of photons for direct emission with energy greater than produced by annihilating dark matter within a galaxy cluster can be calculated by
where is the dark matter density profile, is the radius of a galaxy cluster, and is the energy spectrum of gamma-ray produced per one annihilation, which can be obtained in Cirelli (). Based on the sample of galaxy clusters in Chen () and the observed limits of gamma-ray flux obtained in Ackermann2 (), there are two galaxy clusters (Fornax and A2877) which have the largest ratio , where . Larger the ratio of can give tighter constraints for annihilating dark matter. Therefore, we choose the Fornax and A2877 clusters to perform analysis. By assuming the dark matter density profiles of the A2877 and Fornax clusters follow the Navarro-Frenk-White (NFW) density profile Navarro () and using the corresponding parameters in Anderson (), we can get the total photons produced with energy greater than for direct emission.
On the other hand, recent numerical simulations show that substructures’ contribution for dark matter annihilation is very significant. High dark matter density of the substructures can boost the annihilation signal to more than 10-100 times for galaxy clusters Gao (); Anderhalden (); Sanchez (); Marchegiani (). However, the actual value of the boost factor in a galaxy cluster is controversial. Some high-resolution simulations can get the boost factor (e.g. Phoenix simulation project) Gao () while other studies point to a lower value of the order Anderhalden (); Sanchez (). Moreover, the effect of tidal interaction in galaxy clusters is also controversial. Some studies point out that tidal interaction would give a lower boost Sanchez () while some studies suggest that tidal stripping can increase the boost factor by times Bartels (). Here, for a conservative estimation, we neglect the effect of tidal interaction and take the lower bound of the boost factor, which can be described by a simple parametrization Sanchez ():
where and is the total mass of a galaxy cluster. The accuracy of this parametrization is better than 5% in the mass range Sanchez (). For the A2877 and Fornax clusters, the total mass are and respectively Anderson (). Therefore, the parametrization can be applied and the corresponding boost factors are 35 and 30 respectively. As discussed in Marchegiani (), the boost factor of the order are standard values for galaxy clusters. By taking these boost factors, we can calculate the direct emission flux for each galaxy cluster by , where is the distance of the galaxy cluster. Here, we are using the point-source approximation, which can be justified because we will compare the calculated fluxes with the ‘point-source upper limits’ in Ackermann2 ().
ii.2 Indirect emission
Dark matter annihilation also produces electron and positron pairs. These electrons and positrons would diffuse and cool down to a lower energy by synchrotron cooling, inverse Compton scattering, Bremsstrahlung cooling and Coulomb cooling. Since the diffusion coefficient for a typical galaxy cluster is about cm s Colafrancesco (), the diffusion time scale for a 1 GeV electron is about s. However, the total cooling time scale for the above cooling processes is just s. Therefore, the effect of diffusion is nearly negligible in galaxy clusters. The equilibrium electron energy spectrum can be simply obtained by Storm ()
where is the integrated energy spectrum of electrons produced per one annihilation and is the cooling function which is given by Colafrancesco ()
where is the thermal electron number density in cm, GeV, is the magnetic field strength in G, GeV s, GeV s, GeV s, GeV s and is the Lorentz factor of a high-energy electron or positron. We follow the -profile to model the thermal electron number density profile in a galaxy cluster:
where we have assumed G and for both galaxy clusters (the effect of the uncertainties in will be discussed below) Storm ().
For Bremsstrahlung emission (indirect emission), the total gamma-ray photons emitted per second for indirect emission with energy greater than can be given by Sarazin ()
By taking the same boost factors, we can calculate the indirect emission flux for each galaxy cluster by , with the point-source approximation.
We can now obtain the total gamma-ray flux emitted from each of the galaxy clusters. Based on the observational results from Fermi-LAT Ackermann2 (), we obtain the fluxes for three lower limits of gamma-ray energy GeV, GeV and GeV. Also, we use the upper limits of the fluxes for these values of to constrain (see Table 1 for the point-source approximation upper limits for each ).
|Flux upper limits for GeV||8||11|
|Flux upper limits for GeV||3.6||4.9|
|Flux upper limits for GeV||0.19||0.26|
In Fig. 1, we plot the upper limits for the annihilation cross section for 4 popular channels (, , and ). We can see that the gamma-ray flux upper limits can give very stringent constraints for the annihilation cross section. In standard cosmology, the thermal relic annihilation cross section is cm s Steigman (). The lower limits of dark matter mass for the thermal relic annihilation cross section are GeV, GeV, GeV and GeV for the , , and channels respectively. Although these limits are nearly the same as the current one obtained by Fermi-LAT data of the MW dSphs Ackermann (); Albert (); Geringer (), overall speaking, we get tighter constraints for the upper limits of annihilation cross sections, except for the channel with GeV. Our results improve the upper limits of annihilation cross sections by a factor of for wide ranges of for , and channels, and a factor of for channel ( GeV) (see Fig. 2).
Furthermore, our results are more stringent than that obtained by other recent studies with the data of galaxy clusters. For example, recent studies using the data of Virgo cluster Ackermann3 () and Coma cluster Marchegiani () obtain the lower limit of GeV for dark matter annihilating via channel. Our results improve this limit by a factor of 2.5. Also, our results improve the limits obtained in Huang (); Storm () (using the data of other clusters, such as Ophiuchus, A2199 and AWM7) by nearly an order of magnitude. Nevertheless, by using radio data of galaxy clusters, some recent studies can give more stringent constraints of annihilation cross sections. For example, Beck & Colafrancesco Beck () obtain GeV for the thermal relic annihilation cross section via or by using the radio data of the Coma cluster. However, these studies constrain dark matter annihilation by synchrotron radiation which may suffer from large uncertainties of magnetic field strength. Although our study also includes the consideration of magnetic field strength, we consider the effect of Bremsstrahlung radiation due to magnetic field, but not synchrotron radiation. As discussed below, the uncertainties of magnetic field do not affect our result significantly. Therefore, using gamma-ray data is more reliable than using radio data for constraining annihilating dark matter, especially in using data of galaxy clusters. Based on the above discussion, we suggest that the Fornax and A2877 clusters are better candidates for constraining annihilating dark matter.
Moreover, our constraints here can safely rule out all of the existing popular dark matter interpretation of the GeV excess in our Milky Way (via or ) Calore (); Daylan (); Abazajian2 () (see Fig. 3). Besides, some recent models of dark matter interpretation of the positron excess and antiproton excess in our Milky Way are also ruled out (e.g. and channels, see Fig. 3) Mauro (); Cuoco ().
We have also checked with the effects by varying the magnetic field strength, dark matter profile and boost factor for both galaxy clusters. First of all, the magnetic field strength mainly affects the indirect emission for low energy ( GeV). The contribution of indirect emission varies with dark matter mass mildly. The largest values of the ratio for the , , and channels are 26%, 19%, 3% and 1% respectively (mainly for GeV). In the above calculations, we used a typical central magnetic field strength G and a typical index . We have tried another extreme value G Storm () and change the index to , we find that the overall upper limits of annihilation cross sections are just larger by %. Therefore, the effect of the systematic uncertainties of magnetic field strength (or contribution from indirect emission) is not very significant. The reason is that the most stringent constraints mainly come from the data of gamma-ray flux GeV. Changing the magnetic field strength does not affect too much for the gamma-ray flux for GeV.
For the dark matter profile, we have used the NFW profile to model the density profile of galaxy clusters. We have tried another profile, Einasto profile Merritt (), to model the Fornax and A2877 clusters as some recent studies suggest that the Einasto profile is a better description of dark matter halos on large scale Sereno (). We finally get tighter constraints after using the Einasto profile (see Fig. 4). Here, we only show the effect for the Fornax cluster because it gives the most stringent constraints. Generally speaking, using the NFW profile would give a more conservative result.
For the boost factor, as mentioned above, we used the most conservative values to model the boost factors of Fornax and A2877 clusters. These values depend on the virial mass of the clusters. By using an older value of virial mass for Fornax cluster () Sanchez2 (), the boost factor of the Fornax cluster is 29, which gives only about 3% larger for the upper limits. We have also tried another extreme empirical relation to calculate the boost factor (based on the simulation results from the Phoenix simulation project Gao ()), the boost factor of the Fornax cluster is 525. The resultant constraints are much tighter (see Fig. 5).
Overall speaking, the assumed parameters and the profiles used are conservative. Therefore, the upper limits obtained are conservative limits.
Iv Discussion and Conclusion
Recently, it has been reported that the excess gamma-ray emission in the Milky Way can be best explained by dark matter annihilation. The proposed dark matter mass is about GeV. On the other hand, standard cosmology indicates that the thermal relic annihilation cross section should be cm s so that it gives the correct amount of dark matter in our universe Steigman (). In this article, we obtain the upper limits of the annihilation cross sections for four popular channels by using the Fermi-LAT gamma-ray upper limits of two galaxy clusters (A2877 and Fornax). We improve the upper limits of annihilation cross sections by a factor of for wide ranges of for , and channels, and a factor of for channel (for GeV). Tighter constraints are obtained because we have considered the ‘indirect emission’ (Bremsstrahlung emission) and the effect of substructures in galaxy clusters. In particular, the effect of substructures is significant in galaxy clusters as they are large and massive structures. Besides, we have used the most updated gamma-ray data for analysis, which provide more stringent limits for annihilating dark matter. Furthermore, among a large sample of nearby galaxy clusters, we found that the A2877 and Fornax clusters are the best candidates to constrain dark matter annihilation cross section because they have larger values of .
Besides, the lower limits of the dark matter mass for the thermal relic annihilation cross section are larger than the mass expected in dark matter interpretation of the gamma-ray excess. We show that most of the existing popular dark matter interpretation of the GeV excess in the Milky Way are ruled out. Our results basically support the latest Fermi-LAT analysis of the Milky Way GeV excess Ajello ().
In fact, searching dark matter particles is one of the most important tasks in particle physics. However, the null detection in direct detection experiments may indicate that dark matter mass is probably larger than 1 TeV Tan (); Akerib (). Therefore, our results are quite consistent with these experimental results that dark matter mass may be much larger than the previous expected values. We expect that future observations of TeV gamma rays Abdallah (); Archambault () and anti-proton detection Cavasonza (); Feng () might provide some new insight of dark matter annihilation.
This work is supported by a grant from The Education University of Hong Kong (Activity code: 04234).
Vi Competing financial interests
The authors declare no competing financial interests.
Vii Author contribution statement
Design of the study and writing of the manuscript was performed by Man Ho Chan. Analysis of the results was performed by Man Ho Chan and Chung Hei Leung. All authors approved the submitted version of the manuscript.
Viii Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
- (1) Abazajian, K. N., Canac, N., Horiuchi, S. & Kaplinghat, M. Astrophysical and dark matter interpretations of extended gamma-ray emission from the Galactic Center. Phys. Rev. D 90, 023526 (2014).
- (2) Calore, F., Cholis, I. & Weniger, C. Background model systematics for the Fermi GeV excess. J. Cosmol. Astropart. Phys. 03, 38 (2015).
- (3) Calore, F., Cholis, I., McCabe, C. & Weniger, C. A tale of tails: dark matter interpretations of the Fermi GeV excess in light of background model systematics. Phys. Rev. D 91, 063003 (2015).
- (4) Daylan, T. et al. The characterization of the gamma-ray signal from the central Milky Way: a compelling case for annihilating dark matter. Phys. of Dark Univ. 12, 1 (2016).
- (5) Abazajian, K. N. & Keeley, R. E. Bright gamma-ray Galactic Center excess and dark dwarfs: strong tension for dark matter annihilation despite Milky Way halo profile and diffuse emission uncertainties. Phys. Rev. D 93, 083514 (2016).
- (6) Ackermann, M. et al. [Fermi-LAT Collaboration] Searching for dark matter annihilation from Milky Way dwarf spheroidal galaxies with six years of Fermi Large Area Telescope data. Phys. Rev. Lett. 115, 231301 (2015).
- (7) Albert, A. et al. [Fermi-LAT, DES Collaborations] Searching for dark matter annihilation in recently discovered Milky Way satellites with Fermi-LAT. Astrophys. J. 834, 110 (2017).
- (8) Yuan, Q. & Zhang, B. Millisecond pulsar interpretation of the Galactic Center gamma-ray excess. J. High Energy Astrophys. 3, 1-8 (2014).
- (9) Petrovic̀, J., Serpico, P. D. & Zaharijas, G. Millisecond pulsars and the Galactic Center gamma-ray excess: the importance of luminosity function and secondary emission. J. Cosmol. Astropart. Phys. 02, 023 (2015).
- (10) Gaggero, D., Taoso, M., Urbano, A., Valli, M. & Ullio, P. Towards a realistic astrophysical interpretation of the gamma-ray Galactic Center excess. J. Cosmol. Astropart. Phys. 12, 056 (2015).
- (11) Brandt, T. D. & Kocsis, B. Disrupted globular clusters can explain the Galactic Center gamma-ray excess. Astrophys. J. 812, 15 (2015).
- (12) Bartels, R., Krishnamurthy, S. & Weniger, C. Strong support for the millisecond pulsar origin of the Galactic Center GeV excess. Phys. Rev. Lett. 116, 051102 (2016).
- (13) Ploeg, H., Gordon, C., Crocker, R. & Macias, O. Consistency between the luminosity function of resolved millisecond pulsars and the Galactic Center excess. J. Cosmol. Astropart. Phys. 08, 015 (2017).
- (14) de Boer, W., Bosse, L., Gebauer, I., Neumann, A. & Biermann, P. Molecular clouds as the origin of the Fermi gamma-ray GeV-excess. Phys. Rev. D 96, 043012 (2017).
- (15) Ajello, M. et al. [Fermi-LAT Collobration] Characterizing the population of pulsars in the Galactic bulge with the Fermi Large Area Telescope. ArXiv:1705.00009.
- (16) Ando, S. & Nagai, D. Fermi-LAT constraints on dark matter annihilation cross section from observations of the Fornax cluster. J. Cosmol. Astropart. Phys. 7, 017 (2012).
- (17) Huang, X., Vertongen, G. & Weniger, C. Probing dark matter decay and annihilation with Fermi LAT observations of nearby galaxy clusters. J. Cosmol. Astrooart. Phys. 01, 042 (2012).
- (18) Ackermann, M. et al. [Fermi-LAT Collaboration] Search for extended gamma-ray emission from the Virgo galaxy cluster with Fermi-LAT. Astrophys. J. 812, 159 (2015).
- (19) Lisanti, M., Mishra-Sharma, S., Rodd, N. L. & Safdi, B. R. A search for dark matter annihilation in galaxy groups. ArXiv:1708.09385.
- (20) Liang, Y.-F. et al. Search for a gamma-ray line feature from a group of nearby galaxy clusters with Fermi LAT pass 8 data. Phys. Rev. D 93, 103525 (2016).
- (21) Storm, E., Jeltema, T. E., Profumo, S. & Rudnick, L. Constraints on dark matter annihilation in clusters of galaxies from diffuse radio emission. Astrophys. J. 768, 106 (2013).
- (22) Beck, G. & Colafrancesco, S. A multi-frequency analysis of dark matter annihilation interpretations of recent anti-particle and -ray excesses in cosmic structures. J. Cosmol. Astropart. Phys. 05, 013 (2016).
- (23) Storm, E., Jeltema, T. E., Splettstoesser, M. & Profumo, S. Synchrotron emission from dark matter annihilation: predictions for constraints from non-detections of galaxy clusters with new radio surveys. Astrophys. J. 839, 33 (2017).
- (24) Cirelli, M. et al. PPPC 4 DM ID: a poor particle physicist cookbook for dark matter indirect detection. J. Cosmol. Astropart. Phys. 10, E01 (2012).
- (25) Chen, Y., Reiprich, T. H., Böhringer, H., Ikebe, Y. & Zhang, Y.-Y. Statistics of X-ray observables for the cooling-core and non-cooling core galaxy clusters. Astron. Astrophys. 466, 805-812 (2007).
- (26) Ackermann, M. [Fermi-LAT Collaboration] Search for cosmic-ray-induced gamma-ray emission in galaxy clusters. Astrophys. J. 787, 18 (2014).
- (27) Navarro, J. F., Frenk, C. S. & White, S. D. M. Astrophys. J. 490, 493 (1997).
- (28) Anderson, B. et al. Search for gamma-ray lines towards galaxy clusters with the Fermi-LAT. J. Cosmol. Astropart. Phys. 02, 026 (2016).
- (29) Gao, L., Frenk, C. S., Jenkins, A., Springel, V. & White, S. D. M. Where will supersymmetric dark matter first be seen? Mon. Not. R. Astron. Soc. 419, 1721-1726 (2012).
- (30) Anderhalden, D. & Diemand, J. Density profiles of CDM microhalos and their implications for annihilation boost factors. J. Cosmol. Astropart. Phys. 04, 009 (2013).
- (31) Sànchez-Conde, M. A. & Prada, F. The flattening of the concentration-mass relation towards low halo masses and its implications for the annihilation signal boost. Mon. Not. R. Astron. Soc. 442, 2271-2277 (2014).
- (32) Marchegiani, P. & Colafrancesco, S. The role of dark matter sub-halos in the non-thermal emission of galaxy clusters. J. Cosmol. Astropart. Phys. 11, 033 (2016).
- (33) Bartels, R. & Ando, S. Boosting the annihilation boost: tidal effects on dark matter subhalos and consistent luminosity modeling. Phys. Rev. D 92, 123508 (2015).
- (34) Colafrancesco, S., Profumo, S. & Ullio, P. Multi-frequency analysis of neutralino dark matter annihilations in the Coma cluster. Astron. Astrophys. 455, 21-43 (2006).
- (35) Sarazin, C. L. & Kempner, J. C. Nonthermal Bremsstrahlung and hard X-ray emission from cluster of galaxies. Astrophys. J. 533, 73 (2000).
- (36) Steigman, G., Dasgupta, B. & Beacom, J. F. Precise relic WIMP abundance and its impact on searches for dark matter annihilation. Phys. Rev. D 86, 023506 (2012).
- (37) Geringer-Sameth, A., Koushiappas, S. M. & Walker, M. G. Comprehensive search for dark matter annihilation in dwarf galaxies. Phys. Rev. D 91, 083535 (2015).
- (38) Di Mauro, M., Donato, F., Fornengo, N. & Vittino, A. Dark matter vs. astrophysics in the interpretation of AMS-02 electron and positron data. J. Cosmol. Astropart. Phys. 05, 031 (2016).
- (39) Cuoco, A., Krämer, M. & Korsmeier, M. Novel dark matter constraints from antiprotons in the light of AMS-02. Phys. Rev. Lett. 118, 191102 (2017).
- (40) Merritt, D., Graham, A. W., Moore, B., Diemand, J. & Terzic, B. Empirical models for dark matter halos. I. Nonparametric construction of density profiles and comparison with parametric models. Astron. J. 132, 2685-2700 (2006).
- (41) Sereno, M., Fedeli, C. & Moscardini, L. Comparison of weak lensing by NFW and Einasto halos and systematic errors. J. Cosmol. Astropart. Phys. 01, 042 (2016).
- (42) Sànchez-Conde, M. A., Cannoni, M., Zandanel, F., Gòmez, M. E. & Prada, F. Dark matter searches with Cherenkov telescopes: nearby dwarf galaxies or local galaxy clusters? J. Cosmol. Astropart. Phys. 12, 011 (2011).
- (43) Tan, A. et al. Dark matter results from first 98.7 days of data from the PandaX-II experiment. Phys. Rev. Lett. 117, 121303 (2016).
- (44) Akerib, D. S. et al. Results from a search for dark matter in the complete LUX exposure. Phys. Rev. Lett. 118, 021303 (2017).
- (45) Abdallah, H. et al. [H.E.S.S. Collaboration] Search for dark matter annihilations towards the inner Galactic halo from 10 years of observations with H.E.S.S. Phys. Rev. Lett. 117, 111301 (2016).
- (46) Archambault, S. et al. [VERITAS Collaboration] Dark matter constraints from a joint analysis of dwarf spheroidal galaxy observations with VERITAS. Phys. Rev. D 95, 082001 (2017).
- (47) Cavasonza, L. A., Gast, H., Krämer, M., Pellen, M. & Schael, S. Constraints on leptophilic dark matter from the AMS-02 experiment. Astrophys. J. 839, 36 (2017).
- (48) Feng, J. & Zhang, H.-H. Dark matter search in space: combined analysis of cosmic ray antoproton-to-proton flux ratio and positron flux measured by AMS-02. ArXiv:1701.02263.