decoding the nature of dark matter

PoS(CORFU2018)061

Decoding the nature of Dark Matter

Alexander Belyaev∗

School of Physics and Astronomy, University of Southampton, Southampton SO17 1BJ, UKE-mail: [email protected]

Determination of the nature of Dark Matter (DM) is one of the most fundamental problems ofparticle physics and cosmology. If DM is light enough and interacts with Standard Model parti-cles directly or via some mediators with a strength beyond the gravitational one, it can be directlyproduced at particle accelerators. The study of the complete set of dimension 5 and 6 effectiveDM operators which we present here demonstrates the LHC potential to distinguish operatorswith different spin of DM: they have a different energy dependence and respectively differentdistributions of the invariant mass of the DM pair which consequently leads to different missingtransverse energy distributions. This study can be directly applied beyond the EFT paradigm,as we demonstrate for three cases – Supersymmetry (DM with spin 1/2), inert two Higgs doubletmodel (i2HDM) (spin 0 of DM) and minimal isotriplet of vector dark matter model (MVDM). Wealso stress an importance of complementarity of the collider searches, DM direct detection exper-iments as well as relic density and CMB data. For several appealing DM models as an example,we show that the interplay between high and low energy data has unique power for an identifi-cation of DM nature. We also highlight prospects of new signature from DM theories such asdisappearing charge tracks which are characteristic for wide class of DM theories. Finally, we ad-vocate the importance of the framework which would combine the experience of HEP communityand would allow to effectively identify the underlying theory of DM.

Corfu Summer Institute 2018 "School and Workshops on Elementary Particle Physics and Gravity"(CORFU2018)31 August - 28 September, 2018Corfu, Greece

∗Speaker.

c© Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/

mailto:[email protected]

PoS(CORFU2018)061

Decoding the nature of Dark Matter Alexander Belyaev

1. IntroductionUnderstanding the nature of Dark Mat-

ter (DM) is one of the greatest puzzles ofmodern particle physics and cosmology. Al-though overwhelming observational evidencesfrom galactic to cosmological scales point tothe existence of DM [1, 2, 3], after decadesof experimental effort only its gravitational in-teraction has been experimentally confirmed.Currently, no information is available on theDM properties, such as its spin, mass, inter-actions other than gravitational, symmetry re-sponsible for its stability, number of states as-sociated to it, and possible particles that wouldmediate the interactions between DM and thestandard model (SM) particles.

If DM is light enough and interacts withSM particles directly or via some mediatorswith a strength beyond the gravitational one, itselusive nature can be detected or constrainedin different ways: a) from direct productionat colliders, resulting in a signature exhibit-ing an observed SM object, such as jet, Higgs,Z, or photon, that recoils against the missingenergy from the DM pair [4, 5, 6, 7]; b) viathe relic density constraint obtained throughthe observations of cosmic microwave back-ground (CMB) anisotropies, such as those ofWMAP and PLANCK collaborations [8, 1];c)from DM direct detection (DD) experiments,which are sensitive to elastic spin indepen-dent (SI) or spin dependent (SD) DM scatter-ing off nuclei [9, 10, 11, 12]; d) from DMindirect detection searches, that look for SMparticles produced in the decay or annihilationof DM present in the cosmos, both with highenergies observables (gamma-rays, neutrinos,charge cosmic rays) produced in the local Uni-verse [13, 14, 15, 16, 17, 18], and by studyingthe effects of energy produced by DM annihi-lation in the early universe on the properties ofthe CMB spectrum [19, 20, 1].

It is clear that decoding of the nature of

DM requires the respective signal at least inone of the search experiment. We do not haveone. However even without having this sig-nal at the moment we can already conclude onwhat kind of DM models are excluded already.Moreover, by exploring different signatures ofone particular model, their correlation and in-terplay we can prepare ourselves to discoveryof DM and their identification.

2. Contact interactionsIn Table 1 we have summarised a minimal

set of independent dimension-5 and dimension-6 operators for complex scalar, Dirac fermionand complex vector DM coupling to quarks andgluons, adopting the widely used notations of[21, 22, 23]. These operators provide monojet-signature, the shapes of Emiss

T distributions forwhich is presented in in Fig. 1 from Ref. [23]for DM mass of 10 GeV. One can observe abig difference in Emiss

T shapes of the groupsof the operators, primarily split into groups ofoperators with scalar, femion and vector DM.The origin of the different Emiss

T shapes fromdifferent operators can be related to a combi-nation of effects. First, for a fixed Lorentzstructure of the SM part of the EFT opera-tors, the same invariant mass distribution of theDM pair, Minv(DM,DM), uniquely defines theshape of the Emiss

T distribution. Moreover, withthe increase of Minv(DM,DM), the Emiss

T shapefalls less and less steeply (again, for a given SMcomponent of the EFT operator).

The reason why the bigger invariant massof DM is correlated with flatter Emiss

T behaviourcan be explained by the phase space and par-ton density effects [23]: when Minv(DM,DM)is small, the radiation of a high PT jet will“cost” a large relative shift in x, the transferredmomentum of the parton, leading to a rapidlyfalling Emiss

T distribution; on the contrary, whenMinv(DM,DM) is large, the radiation of a highPT jet will “cost" a small relative shift in x,which will lead to a more slowly falling Emiss

T

1

PoS(CORFU2018)061


distribution in comparison to the first case.

Complex Scalar DM

g2∗Λ

φ †φ qq [C1]g2∗Λ

φ †φ qiγ5q [C2]g2∗Λ2 φ †i

←→∂µ φ qγµ q [C3]

g2∗Λ2 φ †i

←→∂µ φ qγµ γ5q [C4]

g2∗Λ2 φ †φGµν Gµν [C5]g2∗Λ2 φ †φ Gµν Gµν [C6]

Dirac Fermion DM

g2∗Λ2 χχ qq [D1]g2∗Λ2 χiγ5χ qq [D2]g2∗Λ2 χχ qiγ5q [D3]g2∗Λ2 χγ5χ qγ5q [D4]g2∗Λ2 χγµ χ qγµ q [D5]g2∗Λ2 χγµ γ5χ qγµ q [D6]g2∗Λ2 χγµ χ qγµ γ5q [D7]g2∗Λ2 χγµ γ5χ qγµ γ5q [D8]g2∗Λ2 χσ µν χ qσµν q [D9]g2∗Λ2 χσ µν iγ5χ qσµν q [D10]

Complex Vector DM

g2∗m2DM

Λ3 V †µV µ qq [V1]

g2∗m2DM

Λ3 V †µV µ qiγ5q [V2]

g2∗m2DM

2Λ4 i(V †ν ∂µV ν −V ν ∂µV †

ν )qγµ q [V3]g2∗m2

DM2Λ4 (V †

ν ∂µV ν −V ν ∂µV †ν )qiγµ γ5q [V4]

g2∗m2DM

Λ3 V †µVν qiσ µν q [V5]

g2∗m2DM

Λ3 V †µVν qσ µν γ5q [V6]

g2∗mDM2Λ3 (V †

ν ∂ νVµ +Vν ∂ νV †µ )qγµ q [V7P]

g2∗m2DM

2Λ4 (V †ν ∂ νVµ −Vν ∂ νV †

µ )qiγµ q [V7M]g2∗mDM

2Λ3 (V †ν ∂ νVµ +Vν ∂ νV †

µ )qγµ γ5q [V8P]g2∗m2

DM2Λ4 (V †

ν ∂ νVµ −Vν ∂ νV †µ )qiγµ γ5q [V8M]

g2∗mDM2Λ3 εµνρσ (V †

ν ∂ρVσ +Vν ∂ρV †σ )qγµ q [V9P]

0.5 g2∗mDM2Λ3 εµνρσ (V †

ν ∂ρVσ −Vν ∂ρV †σ )qiγµ q [V9M]


ν ∂ρVσ +Vν ∂ρV †σ )qγµ γ5q [V10P]


ν ∂ρVσ −Vν ∂ρV †σ )qiγµ γ5q [V10M]

g2∗m2DM

Λ4 V †µV µ Gρσ Gρσ [V 11]

g2∗m2DM

Λ4 V †µV µ Gρσ Gρσ [V 12]

Table 1: Minimal basis of operators of dimensionsix or less involving only complex scalar DM (φ ),Dirac fermion DM (χ) or complex vector DM (V µ )interacting with SM quarks (q) or gluons. Here wedenote the field strength tensor of the gluons as Gµν

and its dual as Gµν .

Different oprators have differentt en-ergy behaviour and respective different in-variant mass distributions. Majority scalarDM operators have the smallest mean valueof Minv(DM,DM), while vector DM hasthe largest mean value of Minv(DM,DM) asdemonstrated in In Figure 2. Because ofrelation of the shape of Minv(DM,DM) andEmiss

T slope distributions one can distinguishmany operators and related underlying the-ories between ear other by the shape ofthe Emiss

T signal: C1-C2,C5-C6,D9-D10,V1-V2,V3-V4,V5-V6 and V11-12 pairs amongeach other [23](given thatt that the signal is ob-served).

missTE

0 200 400 600 800 1000 1200 1400

# E

vent

s (n

orm

aliz

ed to

one

)

4−10

3−10

2−10

1−10

1

= 13 TeVs = 10, DMMC1,C2C1Q,C2QC3,C4C5,C6D1-D4D1Q-D4QD1T-D4TD5-D8D9-D10V1-V2V1Q-V2QV3-V4, V7-V10V5-V6V5Q-V6QV11-V12

Figure 1: EmissT parton level distributions for a rep-

resentative subset of the EFT operators from Table 1for 13 TeV LHC energy and MDM = 10 GeV.

2

PoS(CORFU2018)061


(DM,DM)invM0 1000 2000 3000 4000 5000 6000 7000 8000

# E

vent

s (n

orm

aliz

ed to

one

)

5−10

4−10

3−10

2−10

1−10

1 = 13 TeVs = 10, DMM

C1,C2C1Q,C2QC3,C4C5,C6D1-D4D1Q-D4QD1T-D4TD5-D8D9-D10V1-V2V1Q-V2QV3-V4, V7-V10V5-V6V5Q-V6QV11-V12

Figure 2: Invariant mass of DM pair distribu-tions normalised to unity for EFT operators in Ta-ble 1 for 13 TeV LHC energy, MDM = 10 GeV andpT, jet ≥ 100 GeV cut applied.

Non-collider DM searches play an im-portant complementary role in probing DMparameter space. As an example in Fig. 3(top) we present the non-collider constraintsfor the operators D2, which exhibit pseudo-scalar interactions of fermion Dirac DMwith quarks. One can see that even formomentum-suppressed operator D2 (becauseof its pseudo-scalar nature) DM DD constraintsfrom Xenon[24] play an important role whichis comparable to collider constraints, presentedin Fig. 3 (bottom). It is important to stressthat both LHC and DM DD searches set an up-per limit on value of Λ. The LHC limit is ofthe order of 1 TeV for present LHC data whileDM DD searches the limit strongly depend onthe operator. For example for non-suppressedoperators conserving parity the limit on Λ isabout 3 orders of magnitude above the LHCone. On the other hand LHC limit is beyondDM DD searches for operators with suppressedelastic scattering cross sections on the nuclei(C2,C4,C6,D2,D3,D4,D6-10,V2,V4-V10).

D2: g*2

Λ2(χiγ5χ) (qq)

Λ < 2 mDM

Ωχ ∼ ΩDM

CMB

Xenon100

5 10 50 100 500 1000100

1000

104

105

106

mDM [GeV]

Λ[GeV

]

g* = 1

D2: g*2

Λ2(χγ5χ)(qq)

g* = 4 π

g* = 6

g* = 1

non-coll.

5 10 50 100 500 1000100

1000

104

105

106

mDM [GeV]

Λ[GeV

]

Figure 3: Top: Non-collider constraints on D2 op-erator with fermion DM: (i) SI DM DD searches(shaded blue region below the lowest blue contour),(ii) constraints from relic density (above the yel-low dashed line), (iii) constraints from the CMB( shaded green area) and (iv) constraints from thevalidity of the EFT (Λ > 2mDM). Bottom: LHCmonojet constraints on D2 EFT operator. The areainside the red, orange and blue solid curves is ex-cluded by current LHC data at 95% CL for g? = 1,6 and 4π , respectively. The projected LHC lim-its for 300 fb−1 are indicated by dashed thin lines.The combined exclusion regions from CMB andDM DD searches for g? = 1 are given by the light-purple area. See details and complete set of plots inRef [25].

Moreover for operators with pseudo-vector currents which have suppressed DM DDrates, one should take into account effect oftheir running from TeV energy scale at theLHC down to low energy scale at DM DD ex-periments, due to which an operator acquirenon-negligible vector component [26, 27, 28].

3

PoS(CORFU2018)061


3. Beyond EFTThe analysis of Emiss

T shape presented herecan be applied to different scenarios, beyondthe EFT approach in general, where the DMmediator is not produced on-the-mass-shell,such as the case of t-channel mediator or me-diators with mass below 2MDM, where theMinv(DM,DM) is not fixed. This case cov-ers a wide range of theories. As an ex-ample in Fig. ?? the normalised shape forEmiss

T distribution from pp→ χ+1 χ−1 /χ

±1 χ0

1 →χ0

1 χ01 + softletons/jets Minimal Supersymmet-

ric Model(MSSM) signal and its dominant ir-reducible background Z + jet→ νν + jet (Z j)is presented for LHC@13TeV [29]. One cansee that Emiss

T distribution from the SM Z + jetbackground falls steeply than that from MSSMsignal. this is related to the fact that invariantmass of χ0

1 χ01 pair is larger than the mass of

Z-boson.Another example is given in Fig. 5 where

we present EmissT from h1h2 j inert two Higgs

doublet model (i2HDM) signal alongside theestimated (by CMS) experimental backgroundfor√

s = 13 TeV. Again, EmissT distribution

from SM BG falls more rapidly because themean value of invariant mass of DM is biggerthen the Z-boson mass.

Finally in Fig. ?? we present shapes ofpT distributions for pair production of thecharged component of the vector isotripletfrom MVDM model [30] and of a wino fromMSSM model, for different values of themasses. One can see that in case of MVDMmodel the pT distribution is flatter than in theMSSM case. This difference is again related tothe mean value of invariant mass of odd par-ticles distributions (V+ and χ+) respectivelywhich has large mean value in case of MVDMmodel. This is an important feature since onecan use MSSM results for MVDM reinterpre-tation using conservative estimation since theefficiency of the pT cuts will be higher for the

MVDM model.An important feature of the Emiss

T or pT

signal shapes for these different DM theoriesis that has the same explanation as for EFTstudy case above – it is related to the invari-ant mass of the invisible (or charged in caseof MVDM model) system. This feature pro-vides a very important way to increase signal-to-background ratio (S/B) (which is typicallybelow 1% for low Emiss

T cuts) by increasingthe value of Emiss

T or by performing the signal-background shape analysis [31].

(GeV)jTp

0 200 400 600 800 1000 1200 1400 1600 1800 2000-910

-810

-710

-610

-510

-410

-310

-210

-110

1

jχχj vs. pp->ννpp->Background

=93 GeVµ=500 GeVµ

Backgroundµ = 93 GeV

µ = 500 GeV

pjT [GeV]

Figure 4: Signal (dotted blue and dashed red) andZ j background (solid black) parton-level p j

T distri-butions for the 13 TeV LHC for the NSUSY sce-nario: normalised signal and Z j background distri-butions. See details in Ref [29].

200 400 600 800 1000 1200 [GeV]miss

TE

5−10

4−10

3−10

2−10

1−10

1

Nor

mal

ized

eve

nts

= 20 GeV h1M

= 40 GeV h1M = 50 GeV h1M

= 60 GeV h1M = 80 GeV h1M

= 100 GeV h1M = 200 GeV h1M

, 13 TeV)-1CMS background (2.3 fb

Figure 5: EmissT from h1h2 j i2HDM signal vs back-

4

PoS(CORFU2018)061


ground for√

s = 13 TeV. See details in [31]

200 400 600 800 1000

pjT (GeV)

10−5

10−4

10−3

10−2

10−1

100

dσ

dpj T

/σ

pp→ χ+χ−j/V +V −j at LHC@13 TeVMχ+ = 100 GeV

MV = 100 GeVMχ+ = 200 GeV

MV = 200 GeVMχ+ = 500 GeV

MV = 500 GeVMχ+ = 1000 GeV

MV = 1000 GeV

Figure 6: The shapes of pT distributions for pairproduction of the charged component of the vectorisotriplet from MVDM model and of a wino fromMSSM model, for different values of the masses.See details in [30]

The role of non-collider DM searchesis also crucial in case of these two com-plete and consistent models. And an exam-ple in Fig. 7 we present the projected LHCreach for MSSM monojet signal in the ∆M =

mχ+1−mchi01

, MDM = mchi01parameter space to-

gether with LUX anx Xenon1T DM DD exclu-sion Ref [29]. One can see that LHC would beable cover neutralino DM mass only below 250GeV (with the assumption that S/B of the orderof 3% will be under control) even with 3 ab−1

total integrated luminocity. It is worth to stressthough that LHC will cover the region inacces-sible by Xenon1T in small ∆M region, whileXenon1T is able to cover mDM well beyond theLHC reach for ∆M > 3−5 GeV, demonstratinga very important complementarity of DM DDto collider searches of DM. In case of i2HDMmodel collider sensitivity with mono-jet signa-ture is even more limited because of the lowerproduction rates of the scalar DM, h1, or itsinert partners (h2 and h+) and expected LHCreach is below 100 GeV for Mh1.

50 100 150 200 250 3000

5

10

15

20

25

30

35

mχ10 [GeV]

ΔM

[GeV

]

LHC13 2σ contour (M1>0)

LHC13 100 fb-1(3%)

LHC13 3 ab -1(3%)

LHC13 3 ab-1(5%)

LUX

XENON1TLEP

Figure 7: Exclusion contour lines for the 13 TeVLHC at the end of the LHC Run2 (light red region)and of the HL-LHC (light blue region). The regionexcluded by LUX and Xenon1T are also shown, to-gether with the LEP limit. See details in Ref [29].

4. Beyond mono-X signatureWhile mono-X (with X being jet,γ,Z,H, t

etc.) DM signatures at colliders are the mostgeneral ones, their rates is typically very low(usually at the percent level or even lower).Besides several others interesting but model-specific DM signature studies one should stressone signature which can be also considered asquite generic one. In case when DM, D0, is em-bedded into electroweak multiplet and its masssplit from the charged odd particle(s), D+, isgenerated only radiatively (preserving gaugeinvariance), the one can find that the value ofthis mass split is of the order of 0.2 GeV. Inthis case D+ has a very small width and respec-tively large life-time driven by its dominant de-cay to DM and pion: D+→D0π0 which makesD+ long lived particles (LLP). Production ofD+ in pairs or in association with DM leadsthen to the typical signature from charged LLP:disappearing charged tack (DCT) as soon as the

5

PoS(CORFU2018)061


track from LLP is long enough (from few cmto a meeter). In case of such signature the S/Bratio is much higher than in case of mono-jetsignal and therefore, substantially bigger DMmasses can be probed with charged LLPs fromDM sector [32, 33, 30]. As an example, wewould like to present here results for the min-imal vector triplet DM (V 0) model [30] whichpredicts the right amount of DM for MDM in the3-4 TeV range depending on DM coupling tothe Higgs boson.

In this model the SM is supplemented bya new massive vector boson Vµ in the adjointrepresentation of SU(2)L, e.g. by two newmassive vector particles: V 0 and V±. If Vµ

transforms homogeneously (i.e. Vµ → g†LVµgL

where gL ∈ SU(2)L) and Z2 symmetry is im-posed (which links the quartic V coupling tothe gauge coupling constant and makes theoryunitary is unitary before EW symmetry break-ing and in the absence of the Higgs boson asfound in Ref. [34]) then Lagrangian can bewritten as:

L = LSM−Tr

DµVνDµV ν

+ Tr

DµVνDνV µ

− g2

2Tr

[Vµ ,Vν

][V µ ,V ν ]

− igTr

Wµν [V µ ,V ν ]

+ M2TrVνV ν

+ a(Φ

†Φ)

TrVνV ν

where Dµ = ∂µ− ig[Wµ ,

]is the usual SU(2)L

covariant derivative in the adjoint representa-tion and LSM represents the SM Lagrangian.The main difference with respect to the modelin Ref. [34] is that the SU(2)L symmetry is bro-ken by the Higgs mechanism and the associatedgauge bosons have mass. We thus allow for acoupling of V to the Higgs scalar field Φ.

In Fig. 8(top) we present results for spin-independent cross-section for V 0-nucleon elas-tic scattering as a function of MV and for repre-sentative values of a. It is very important to

note that Xenon1T experiment excludes DMmass above 4 TeV, while from Fig. 8(bottom)one can see that LHC excludes now 1.4 TeVvector DM, while 100 TeV collider will be ableto exclude DM mass below 4 TeV and there-fore to probe the entire parameter space of themodel.

102 103 104

MV (GeV)

10−11

10−10

10−9

10−8

10−7

σp SI

(pb

)

Direct detection constraints

a = 0.1

a = 0.5

a = 1

a = 5

a = 4π

XENON1T(2018)

Unitarity constraints

ΩDMh2 = ΩPLANCKh

2

500 1000 1500 2000 2500 3000 3500 4000MV (GeV)

10−3

10−2

10−1

100

101

102

103

104

σeff

(fb)

LHC@13, @27TeV and FCC@100 TeV constraints from LLP searches

13 TeV LHC27 TeV LHC100 TeV FCC1.15 fb LHC limit2.0 fb FCC limit

Figure 8: Top: Spin-independent cross-section forV 0-nucleon elastic scattering as a function of MV

and for representative values of a. The continu-ous black curve represents the elastic cross-sectioncomputed with the values of MV and a that saturatethe measured DM relic density. The grey dashinghighlights the parameter space where perturbativeunitarity loss. Bottom: the effective cross-sectionsσe f f =σ(pp→V±V 0)+2σ(pp→V+V−) at lead-ing order for the vector isotriplet model for 13 and27 TeV LHC energies and for a 100 TeV future col-lider. The dashed lines corresponds to collider sen-sitivity. See details in [30].

6

PoS(CORFU2018)061


One should also note that in case ofi2HDM, DCT signature also allows to substan-tially enhance LHC potential and probe DMmass upto about 500 GeV [32] which is muchhigher than 100 GeV – the maximum DM masswhich can be probed via mono-jet signature.

5. Towards DM decoding framework

There is no framework at the momentwhich can solve the reverse engineering task– the task of decoding the nature of DM. It isnot surprising why – we are all eagerly look-ing for the signal first of all and busy withthe interpreting and exploring of our own mod-els. There is a huge amount of work has beendone on the model building, phenomenologyand experimental searches as well as on build-ing different tools examples of which has beengiven above. And there is really huge poten-tial of combining different methods and signa-tures to probe different models. What is miss-ing is the framework which join all these piecesin one tool which would help us to decode un-derlying theory, in particular its part related toDM. The task of decoding of the whole un-derlying theory sounds probably too ambitiousto the author, while decoding of its DM partsound more realistic since it contain specificand possibly much smaller piece of the the-ory. This framework requires the database ofmodels, database of various signatures and setof tools which will be able to effectively ex-plore not only parameter space of each particu-lar model, but also the model parameter spaceand compare predicted signatures with the ob-served ones. Such a framework would allowobjectively judge about preferred model or setof models which would fit signal best of all.An example of the prototype of such a frame-work actually already exists in the form of HighEnergy Physics Model Data Base (HEPMDB)(https://hepmdb.soton.ac.uk) [35],

created at Southampton University in 2011. Atthe moment HEPMDB is created as a web-server accessible to everybody and is able to:

1. collect HEP models for all multipurposeMatrix Element (ME) generators in theform of Feynman rules and parameterswritten in the format specific for a givenpackage;

2. collect models’ sources which can beused to generate HEP models for variousME generators using FeynRules [36] orLanHEP [37];

3. allow users to perform simulations fortheir own models or models availableat HEPMDB using the full power ofthe High Performance Computing (HPC)IRIDIS cluster standing behind the HEP-MDB itself. Connection to HPC clusteris one of the key features of the HEP-MDB: it provides a web interface to vari-ous ME generators (CalcHEP, Madgraphand Whizard at the moment) which canthen also be run directly on the HPCcluster avoiding problems related to in-stalling the actual software, which cansometimes be quite cumbersome;

4. collect simulated events and plot distri-butions using web interface.

Though the signature database at HEP-MDB is at the development stage, users can in-dicate some essential features of the signatureswhich model can provide, such as presence ofresonance, Emiss

T etc. The next step of develop-ment of HEPMDB will include an addition ofvarious packages to event analysis. Probablythe most important feature of HEPMDB is thatit can be developed by the whole HEP com-munity – any registered user can add his/herown model and signature which would be usedfor identification of underlying theory when theexperimental signal will be observed.

7

PoS(CORFU2018)061


6. ConclusionsIn the absence of DM signal we can still do

a lot – we can prepare ourselves ot its discov-ery and identification. Emiss

T shape is quite in-strumental in understanding the underlying the-ory at colliders, while direct and indirect DMsearches are very powerful in complementingcollider searches especially in the parameterspace with large DM mass. We also advocatethe usage of new DM signatures such as disap-pearing charge tracks ones which allows to sub-stantially extend collider exploration of largeDM mass. Moreover we would like to stressthe crucial role of 100 TeV pp collider which islikely to explore the complete parameter spaceof thermal DM. We show that collider and non-collider DM searches have a unique power todecode the nature of Dark Matter on the exam-ples of several appealing DM theories. Suchcomplementarity and usage of different signa-tures would allow us to decode the nature ofDM, signals from which we are expecting inthe near future. Finally, we advocate the impor-tance of the framework which would combinethe experience of HEP community and wouldallow to identify effectively underlying theory.

AcknowledgementsAuthor would like to thank organisers

of CORFU2019 workshop for the invita-tion and very warm hospitality and acknowl-edges partial support from the STFC grantST/L000296/1, and Soton-FAPESP grant. ABalso thanks the NExT Institute and Royal Soci-ety International Exchange grant IE150682.

References

[1] Planck Collaboration, P. A. R. Ade et. al.,Astron. Astrophys. 594 (2016) A13[1502.01589].

[2] G. R. Blumenthal, S. M. Faber, J. R. Primackand M. J. Rees, Nature 311 (1984) 517–525.[,96(1984)].

[3] J. S. Bullock, T. S. Kolatt, Y. Sigad, R. S.Somerville, A. V. Kravtsov, A. A. Klypin,J. R. Primack and A. Dekel, Mon. Not. Roy.Astron. Soc. 321 (2001) 559–575[astro-ph/9908159].

[4] ATLAS Collaboration, M. Aaboud et. al.,JHEP 01 (2018) 126 [1711.03301].

[5] CMS Collaboration, C. Collaboration.

[6] ATLAS Collaboration, Tech. Rep.ATLAS-CONF-2018-005, CERN, Geneva,Apr, 2018.

[7] ATLAS Collaboration, M. Aaboud et. al.,Eur. Phys. J. C77 (2017), no. 6 393[1704.03848].

[8] WMAP Collaboration, G. Hinshaw et. al.,Astrophys. J. Suppl. 208 (2013) 19[1212.5226].

[9] M. W. Goodman and E. Witten, Phys. Rev.D31 (1985) 3059. [,325(1984)].

[10] XENON Collaboration, E. Aprile et. al.,Phys. Rev. Lett. 119 (2017), no. 18 181301[1705.06655].

[11] LUX Collaboration, D. S. Akerib et. al.,Phys. Rev. Lett. 118 (2017), no. 2 021303[1608.07648].

[12] PandaX-II Collaboration, X. Cui et. al.,Phys. Rev. Lett. 119 (2017), no. 18 181302[1708.06917].

[13] T. R. Slatyer in Theoretical Advanced StudyInstitute in Elementary Particle Physics:Anticipating the Next Discoveries in ParticlePhysics (TASI 2016) Boulder, CO, USA, June6-July 1, 2016, 2017. 1710.05137.

[14] Fermi-LAT Collaboration, M. Ackermannet. al., Phys. Rev. Lett. 115 (2015), no. 23231301 [1503.02641].

[15] VERITAS Collaboration, B. Zitzer, PoSICRC2015 (2016) 1225 [1509.01105].

[16] Fermi-LAT, MAGIC Collaboration, M. L.Ahnen et. al., JCAP 1602 (2016), no. 02 039[1601.06590].

[17] H.E.S.S. Collaboration, H. Abdallah et. al.,Phys. Rev. Lett. 117 (2016), no. 11 111301[1607.08142].

8

http://arXiv.org/abs/1502.01589

http://arXiv.org/abs/astro-ph/9908159












PoS(CORFU2018)061


[18] H.E.S.S. Collaboration, A. Abramowskiet. al., Phys. Rev. Lett. 110 (2013) 041301[1301.1173].

[19] S. Galli, F. Iocco, G. Bertone andA. Melchiorri, Phys. Rev. D80 (2009) 023505[0905.0003].

[20] S. Galli, F. Iocco, G. Bertone andA. Melchiorri, Phys. Rev. D84 (2011) 027302[1106.1528].

[21] J. Goodman, M. Ibe, A. Rajaraman,W. Shepherd, T. M. Tait et. al., Phys.Rev.D82 (2010) 116010 [1008.1783].

[22] J. Kumar, D. Marfatia and D. Yaylali, Phys.Rev. D92 (2015), no. 9 095027[1508.04466].

[23] A. Belyaev, L. Panizzi, A. Pukhov andM. Thomas, JHEP 04 (2017) 110[1610.07545].

[24] XENON Collaboration, E. Aprile et. al.,Phys. Rev. Lett. 121 (2018), no. 11 111302[1805.12562].

[25] A. Belyaev, E. Bertuzzo, C. Caniu Barros,O. Eboli, G. Grilli Di Cortona, F. Iocco andA. Pukhov, 1807.03817.

[26] R. J. Hill and M. P. Solon, Phys. Lett. B707(2012) 539–545 [1111.0016].

[27] M. T. Frandsen, U. Haisch, F. Kahlhoefer,P. Mertsch and K. Schmidt-Hoberg, JCAP1210 (2012) 033 [1207.3971].

[28] L. Vecchi, 1312.5695.

[29] D. Barducci, A. Belyaev, A. K. M. Bharucha,W. Porod and V. Sanz, JHEP 07 (2015) 066[1504.02472].

[30] A. Belyaev, G. Cacciapaglia, J. Mckay,D. Marin and A. R. Zerwekh, 1808.10464.

[31] A. Belyaev, T. R. Fernandez Perez Tomei,P. G. Mercadante, C. S. Moon, S. Moretti,S. F. Novaes, L. Panizzi, F. Rojas andM. Thomas, 1809.00933.

[32] A. Belyaev, G. Cacciapaglia, I. P. Ivanov,F. Rojas-Abatte and M. Thomas, Phys. Rev.D97 (2018), no. 3 035011 [1612.00511].

[33] R. Mahbubani, P. Schwaller and J. Zurita,JHEP 06 (2017) 119 [1703.05327].[Erratum: JHEP10,061(2017)].

[34] A. R. Zerwekh, Int. J. Mod. Phys. A28(2013) 1350054 [1207.5233].

[35] M. Bondarenko, A. Belyaev, J. Blandford,L. Basso, E. Boos, V. Bunichev et. al.,1203.1488.

[36] A. Alloul, N. D. Christensen, C. Degrande,C. Duhr and B. Fuks, Comput. Phys.Commun. 185 (2014) 2250–2300[1310.1921].

[37] A. Semenov, 1005.1909.

9





















decoding the nature of dark matter

Documents