The Gamma Ray Milky Way - astro.queensu.caastro.queensu.ca/~irwin/phy813/GRMW.pdf · 2 Gamma-ray...

15
The Gamma Ray Milky Way Colin Lewis Phys 813 December 9, 2014

Transcript of The Gamma Ray Milky Way - astro.queensu.caastro.queensu.ca/~irwin/phy813/GRMW.pdf · 2 Gamma-ray...

  • The Gamma Ray Milky Way

    Colin LewisPhys 813

    December 9, 2014

  • 1 Introduction

    Gamma rays (γ-rays) are photons of the most energetic electromagneticradiation, with frequencies below 10pm and energies exceeding 120 keV. Thelarge amount of energy associated with γ-rays is highly ionizing, and is easilyable to break down many atoms and organic material. The presence ofγ-rays on a planet would be disastrous to life like ours, so it is fortunate fororganisms on Earth that most radiation with energy higher than visible light(including γ radiation) is unable to easily penetrate Earth’s atmosphere - thedanger of sterilizing γ-rays reaching the surface is avoided. The opacity ofEarth’s atmosphere to gamma radiation, however, makes it challenging tomake observations of γ-ray emitting objects. Reaching the large amounts ofenergy required to produce a photon of gamma radiation is challenging too.Instead of the blackbody emission that most lower energy radiation comesfrom, γ-rays often involve the conversion of mass to energy or other highdensity, high energy nuclear reactions.

    2 Gamma-ray Telescopes and Instruments

    Figure 1: One of the MAGIC telescopes(MAGIC I), located at La Palma in theCanary Islands [ASPERA/R.Wagner,MPI Munich].

    There are two techniques used toobserve gamma radiation. On thesurface of Earth, we can do indi-rect observations of γ-rays by ob-serving the effects of gamma radi-ation hitting the atmosphere. Thishowever, requires a reconstructionof the γ spectrum, and is sub-ject to more uncertainty due tolow intensities over very short times(∼ 10 ns) - photomultipliers withwide angles are used to observethese events. Gamma rays causeparticle/antiparticle pair production(§3.3) upon atmospheric entry, withnon-thermal Bremsstrahlung radia-tion reducing the energy of the pair by creating lower energy gamma rays,which result in showers of particles. Detecting γ-rays from the surfaceis done by using instruments called imaging Air Cherenkov Telescopes(IACTs), which observe the cascades of secondary particles and attemptto reconstruct the γ image [Bringmann and Weniger, 2012]. Some impor-

    1

  • tant systems that use this method for astronomical observations are theHigh Energy Stereoscopic System (H.E.S.S), the Very Energetic Radia-tion Imaging Telescope Array System (VERITAS), and the Major Atmo-spheric Gamma-ray Imaging Cherenkov Telescopes (MAGIC) (Fig.2). VER-ITAS, H.E.S.S., and MAGIC all consist of arrays of telescopes, using five,four, and two reflectors respectively to reconstruct images after detect-ing Cherenkov light in the atmosphere [Baixeras, 2003, Konopelko, 1999].We can also make direct observations of γ-rays using space telescopes.Currently there are a few space telescopes dedicated to observing γ-rays:NASA’s Fermi Gamma-ray telescope launced, the Italian Space Agency’sAstro-Rivelatore Gamma a Immagini Leggero (AGILE),ESA’s INTErna-tional Gamma-Ray Astrophysics Laboratory (INTEGRAL) and the HighEnergy Transient Explorer 2 (HETE-2) are some of the more importantones [Tavani et al., 2009, Rando et al., 2013, Teegarden and Sturner, 1999].Some past telescopes that have also been important to Gamma-ray astronomyhave been NASA’s Swift, and the Italian-Dutch BeppoSAX, although theyinvolved more in looking at cosmological γ-rays than Galactic γ-rays.

    3 Gamma-ray Sources

    3.1 Dark Matter Annihilation

    The dark, non-baryonic matter that causes galaxies to have flat rotation curvesout to radii far past the stellar distribution has a few theories surrounding itstrue nature. The existence of Weakly Interacting Massive Particles (WIMPS)is one of the more promising theories, with a leading candidate for dark matterparticles being one of the one predicted by supersymmetry, the neutralinoχ [Bergström, 1999]. This particle may be able to be observed throughscattering reactions with detectors on Earth, or in a particle collider, orindirectly via looking for results of annihilation. There are two processesinvolving χ that result in the production of gamma radiation:

    χχ→ Zγ (a)

    andχχ→ γγ. (b)

    The γ-rays that result from the reactions in [Eq. a,b] have a certain energy,causing emission lines of monochromatic γ-rays. The theorized energy for thesestates are Eγ ' Mχ and Eγ ' Mχ

    (1−M2Z/4M2χ

    ), with 5 GeV ≤ Mχ ≤ 1

    TeV,MZ ' 91GeV . [Ackermann et al., 2011] [Hooper and Goodenough, 2011].

    2

  • In order to indirectly detect WIMPs, we look for signs of these γ-rays due toχ annihilation - gamma-ray signatures are important because of the geometryof their travel, they are not perturbed easily and as such give good spatialsignatures [Bringmann et al., 2012]. The gamma ray flux in the Galactic halodue to χ neutralino annihilations into γγ is believed to be describable by:

    dJγdEdΩ

    (ξ) =〈σv〉χχ→γγ

    8πM2χ2δ(E − Eγ)

    l.o.s

    dsρ2dm(r), (1)

    where ξ is the angle to the galactic centre, 〈σv〉χχ→γγ is the partial annihilationcross section of the χχ→ γγ interaction, and ρ2dm is the dark matter densityas a function of the Galactocentric distance r [Weniger, 2012]. ρdm maybe one of several profiles, including the generalized Navarro-Frenk-White(NFW) profile [Navarro et al., 1996]. In addition to the line emission in γγand Zγ, gamma rays may also be given off in other reactions that involvethe χ neutralino. Internal (non-thermal) Bremsstrahlung (IB) radiation mayalso give off gamma radiation when χ neutralinos annihilate into chargedparticles [Bringmann et al., 2008]. IB causing events and intermediate events(like χχ → φφ, φ → γγ) are less likely than the ZZ and γZ products, andtherefore are only believed to make up a small amount by total percentageof the products resulting from χχ annihilation - the bulk of γ-ray radiationfrom χ neutralinos should be the monochromatic γ-ray emission lines fromthe χχ→ Zγ and χχ→ γγ [Birkedal et al., 2005].

    3.2 Gamma Ray Pulsars

    The high gravity and magnetic fields around a pulsar provide the extremeconditions needed for the production and emission of gamma radiation.The high rotation rate, combined with these conditions, make the envi-ronment of a neutron star very much like that of a particle accelerator[Caraveo, 2014]. There are two main classes of pulsars that give off γ radia-tion, ordinary pulsars and millisecond pulsars (MSPs). One of the supportingtheories as to how ordinary pulsars emit gamma radiation is called thePolar Cap (PC) model [Harding and Muslimov, 1998]. In the PC model,electron-positron pairs are formed in a plasma in the polar field lines atthe poles of a pulsar, where the particles move at ultrarelatvistic speeds[Arons, 1983]. The site of this behaviour is called a pair formation front(PFF). As the particles move at ultrarelativistic velocities, they may loseenergy via curvature radiation (CR) mostly, but also to inverse Comptonscattering (ICS), causing the release of γ rays [Harding and Muslimov, 1998].The released γ radiation may then cause the production of a secondary

    3

  • electron/positron pair (§3.3), as photons more energetic than 1 GeV areabsorbed into the magnetic field. These particles will give off more radiationin the form of more γ-rays as they revolve in the magnetic field. The newγ-rays go on to create more pairs, which release more synchrotron radia-tion again and cascade more generations of pairs until the synchrotron γradiation is not energetic enough to create a e−/e+ pair, and the escapingradiation allows us to observe the pulsar as a γ-ray pulsar[Caraveo, 2014].

    Figure 2: The locations of γ-ray emis-sion is shown, with the OG, PC, andSG regions indicated in yellow, red, andblue. [Aliu et al., 2008].

    Other additional models describ-ing the γ emission of these pul-sars are the slot gap (SG) andouter gap (OG) pulsar models. TheSG model has a slot gap in themagnetic field lines of the pul-sar, which allows the electrons re-leased from the surface to acceler-ate to reach high altitudes, whilethe OG model has an outer gap,allowing them to reach even fur-ther out. All models have thesame basic processes involved, butthe energy measured in the emit-ted gamma-rays is different for dif-ferent altitudes - higher energygamma rays come from higherup in the pulsar’s magnetosphere[Aliu et al., 2008]. MSPs are ro-tating neutron stars that have pe-riods on the order of 1-10 mil-liseconds. The magnetic fieldsof millisecond pulsars have beenmeasured to be quite lower thanthat of the magnetic fields in or-dinary pulsars, being around 3orders of magnitude less thanthat of the average radio pulsar[Wijnands and van der Klis, 1998]. MSPs are believed to originally havebeen normal neutron stars in binary systems - many MSPs have been foundin binary systems. As the neutron star became older and lost some magneticfield strength, it also spun up by taking angular momentum from binary com-panions via accretion of matter [Kiziltan and Thorsett, 2009]. The differentmagnetic fields and rotation of MSPs suggests a different mechanism might

    4

  • Figure 3: Top: An all sky map of gamma radiation, showing the locations of γ-ray pulsars from Fermi-LAT. Bottom: Positions of radio-loud and radio-quietpulsars are indicated in green and blue, with MSPs indicated in red and orangetriangles. Here, black dots are other pulsars (no γ emission) and grey dots areglobular clusters. [NASA/DOE/Fermi LAT Collaboration][Caraveo, 2014].

    5

  • be needed to explain the γ radiation. In MSPs the process is similar, withprimary and secondary particles being accelerated through the magnetic fieldsof the pulsar. CR and ICS radiation also play a important parts in MSPs, butthere is also cyclotron resonant radio emission absorption by the ultrarelativis-tic pairs that are created. Synchrotron radiation emission then follows, andthe process continues like in ordinary pulsars [Harding et al., 2005]. Figure 3shows an all-sky γ − ray map that shows the location of several γ-ray pulsarsfound by Fermi.

    3.3 Particle-Antiparticle Annihilation

    When a particle and its antiparticle collide, they are completely annihilatedand give off high energy photons based on the energy of the particles. Theinteraction of a positron and electron is called the Dirac process, a reactionof the form:

    e+ + e− → 2γcreating 2 γ rays for conservation of linear momentum [Ruffini et al., 2010]. Inthe case of slowly moving particles, the energy is close to the rest mass energyE = mec

    2, and the two γ rays released are 511 keV photons [Hubbell, 2006],but for relativistic particles more emission may be present. Particle-antiparticlepairs may also be created by gamma rays through interaction with nuclei,electrons, or other gamma rays, and create gamma rays upon their annihi-lation. If a gamma ray has enough energy, it may spontaneously create anelementary particle and its anti-particle [Hubbell, 2006].

    3.4 Stellar Atmospheres & Flares

    Gamma rays are also created in small amounts in stellar atmospheres - thehigh temperatures and high energies associated with magnetic field eventslike solar flares also provides an environment in which low energy γ rayscan be formed from the interaction of relativistic particles with ambientplasma [Ramaty and Mandzhavidze, 1998]. Bremsstrahlung radiation fromhighly relativistic electrons and positrons also emits a continuum of γ rays, aselectrons and positrons collide with heavier particles in stellar chromospheres.Gamma radiation from protons, neutrons, and α particles colliding withheavier nuclei is also observed in the form of de-excitation lines, as well as anyparticle-antiparticle annihilation. We also observe γ radiation from positronsreleased from radioactive nuclei in decay reactions, adding to the lines from de-excitation and the continuum from the Bremsstrahlung radiation. Looking atthe γ-ray emission from stellar flares, it gives information on the abundancesand some physical quantities in stellar atmospheres. [Aschwanden, 2004]

    6

  • Figure 4: A theoretical spectrum ofthe γ-ray flux received from the sun,set to observations, showing the rangeof of fluxes for γ radiation. Thiscan give a perspective of how stel-lar flares in magnetic stars through-out the Galaxy emit γ-rays. ABremsstrahlung continuum with piondecay radiation dominates in the moreenergetic end with lower photon flux,while the less energetic but brighterend is dominated by emission lines.[Ramaty and Mandzhavidze, 1998].

    3.5 Gamma rays from Supernovae

    The high energies of γ rays are particularly hard to reach via traditionalnuclear transitions, so the most common sources of γ radiation involve extremeconditions. When a massive star ceases core fusion and progresses into asupernova, high amounts of energy in a high density medium allow for manycollisions of nuclei with other particles, particularly free neutrons - chargedparticles are often unable to overcome the Coulomb potential barrier neededto get close to the nucleus. When a neutron collides with a nucleus, we see areaction resulting in a more massive nucleus[Carroll and Ostlie, 2007]:

    AZX + n→ A+1Z X + γ

    which, if unstable, will then undergo a β-decay:

    A+1Z X →A+1Z+1 X + e− + ν̄e + γ

    If the β-decay is on a shorter time scale than the neutron capture, then theprocess is called an s-process, and if the decay is on a longer time scale thanthe neutron capture then the process is called an r-process. Both of theseprocesses involve the production of high energy γ-ray emission lines, but areparticular to supernovae [Carroll and Ostlie, 2007]. As such, we don’t seemuch Galactic γ ray production from supernovae nucleosynthesis due to thelow frequency of supernovae in the Milky Way. There are, however, smallamounts of γ rays from supernova remnants that we observe in the Milky Way(Fig 3.5). The expanding material after a supernova is believed to be a sitefor the formation of, highly relativistic and energetic particles, or cosmic rays.

    7

  • Long after a supernova takes place, shockwaves will continue to propagateoutward, propelling particles to relativistic energies as they encounter theshock front. [Blandford and Ostriker, 1978, Bell, 1978]. These relativistic

    Figure 5: Locations of some supernovae remnants as observed by the Fermitelescope. Note their position along the plane of the Galaxy, compared to theγ-ray pulsars that exist off the plane of the Milky Way, as well as along theplane3. [NASA/DOE/Fermi LAT Collaboration]

    particles may then collide with ambient gas surrounding the supernova, whichmay produce similar results to that in §3.4, as they may experience ICS andnon-thermal Bremsstrahlung radiation, releasing γ-rays. Alternatively, thecollisions with the ambient gas may prompt the emission of π0 pions whichthen decay and release γ-rays. Note that π0 pion decay is most likely to beof one of the two forms

    π0 → 2γ, or π0 → γ + e− + e+.

    Cosmic rays that escape supernovae and their remnants (or AGN and quasars)may travel through the intergalactic and interstellar media and produce similarresults. The small amount of cosmic rays that make it to gas clouds mayinteract with nuclei in the gas, leading to the emission of gamma rays asπ decay, ICS, and non-thermal Bremsstrahlung radiation. This gives of asmall amount of γ radiation as a faint glow [Knödlseder, 2010]. In addition tosupernova remnants, γ-rays may also be found coming from black holes, likethe stellar mass black hole Cygnus X-1. While not a steady source of γ-ray

    8

  • emission, some amounts of γ-ray signals may be detected coming from binaryblack holes due to inverse Compton scattering, and possibly synchrotroninteractions in accretion disks as the black hole accretes matter from itsbinary companion [Laurent et al., 2011].

    4 Gamma-ray Observations

    4.1 The Crab Nebula, and its pulsar B0531+21

    Several observations have been done on the Crab Nebula and the pulsar thatremains of the supernova that caused the crab nebula in 1054 AD. The Crabsystem emits radiation at many wavelengths, including extremely high energyγ-rays of up to 100 TeV from the nebula. Many observations of the Crab

    Figure 6: Left: Crab nebula emission with Chandra X-ray in light blue, HSToptical in green and dark blue, Spitzer infrared in red, and centre of gravity of500GeV, 250GeV, and 160GeV indicated with the filled plus sign, circle, andsquare, respectively. The dotted circle shows the 39% containment radius ofthe γ emission. [Albert et al., 2008]. Right: The spectral energy distributionof the Crab Nebula from several sources, including MAGIC, HESS, andFermi.[MAGIC Collaboration et al., 2014]

    system have been done using many telescopes, with the MAGIC, H.E.S.Sand Fermi telescopes doing the majority of γ-ray measurements over time.Observations of the Crab pulsar have improved our knowledge of the PC, OG,and SG models [MAGIC Collaboration et al., 2014], and we see many signsof inverse Compton scattering from the nebula as expected from (§3.5). Thepulsed emission coming from the Crab pulsar itself ranges up to and exceeds

    9

  • 25MeV, suggesting that an OG model may be the source of the observedγ-ray emission [Aliu et al., 2008].

    4.2 Milky Way Halo & Dark Matter Annihilation

    The central cusp of the Milky Way, as well as some sub-haloes have beenshown to exhibit signs of dark matter annihilation via γ-ray observations insome studies. The observed γ radiation is not correlated to the Fermi bubbles(§4.4), and observation of an emmision-line type spectra with a 130 GeV peak(to 4.5 σ, Fig. 4.2) as detected by Fermi observations leans toward the ideathat both the χχ→ γγ and χχ→ γZ are sources of γ-ray emission as darkmatter neutralinos annihilate [Tempel et al., 2012].

    Figure 7: Evidence of line-like γ-ray emission at (l, b) = (−1◦,−0.7◦) on the leftand (l, b) = (−10◦, 0◦) on the right, with the 130 GeV peak clearly visible asshown by the red line over the black background radiation line, as observationssuggest possible χχ annihilation into γ radiation [Tempel et al., 2012].

    4.3 Galactic Globular Clusters

    A relatively high amount of gamma rays are observed in globular clusters (seeFig.3). The presence of millisecond pulsars in globular clusters seems to bethe reason for this - globular clusters have large populations of millisecondpulsars, and have been shown to have more binary systems than the Galacticdisk by mass. Often, it is difficult to resolve the individual pulsars in globularclusters due to intrinsic dimness and interstellar scattering, but by looking atthe γ-ray signatures the presence of (sometimes several) MSPs can be inferred[Abdo and Ackermann, 2010].

    10

  • 4.4 The Galactic Centre and the Fermi Bubble

    Gamma observations along the disk and toward the centre of the Milky Wayshow diffuse γ-ray emission released from protons interacting with ionizedgas and dust in the ISM, causing π0 decay, along with ICS emission and CRemission like mentioned in §3.5 and §3.4 [Hooper and Slatyer, 2013]. This canbe seen along the plane of the galaxy in (Fig 4.4). Fermi-LAT observationsmade of the galactic centre (GC) have shown giant plasma bubbles in theγ spectrum, extending several kiloparsecs above and below the plane of theMilky Way, after masking of bright sources and removing of the diffuseemission from the plane of the galaxy. The environment from which these

    Figure 8: The same gamma-ray observations of the Milky Way as shownin Fig 3 and 5, with the contrast stretched and the bright sources maskedout, showing diffuse emission on the left due to π0 decay and emission fromcosmic rays interacting with ISM, and on the right are the Fermi bubblesafter the diffuse emission on the left is removed [NASA/DOE/Fermi LATCollaboration].

    bubbles arise seems to be around the supermassive black hole at the centre,Sagittarius A (Sgr A∗), surrounded by molecular clouds and star clusters.These Fermi bubbles bubbles show signs of strong shocks, and are coincidentwith huge magnetic outflows, which suggests that these events may be related.There are a few theories surrounding their formation: the bubbles can bedue to WIMP annihilation events, recent AGN activity, or due to recent starformation close to the GC. A few simulations show that recent star formingactivity around the GC, with feedback leading to accretion of matter aroundSgr A∗ can account for this γ emission, while the structure of the bubbleformation leads many to believe that dark matter annihilation can not be thesource of this [Zubovas et al., 2011, Su et al., 2010, Carretti et al., 2013].

    11

  • 5 Conclusion

    The radiation of low intensity γ-rays is fairly common in the Milky Way- but very high energy occurrences are needed to produce any γ radiation.Stars, pulsars, black holes, dark matter, and cosmic rays all have conditionsthat create energetic interactions needed to be a γ-ray emitter in our galaxy.Observations of the Galactic centre are ideal for looking at γ rays, and cangive insight as to previous activity in the Milky Way.

    6 References

    [Abdo and Ackermann, 2010] Abdo, A. A. and Ackermann, M. t. (2010). A population ofgamma-ray emitting globular clusters seen with the Fermi Large Area Telescope. A&A,524:A75.

    [Ackermann et al., 2011] Ackermann, M., Ajello, M., and Albert, A. t. (2011). ConstrainingDark Matter Models from a Combined Analysis of Milky Way Satellites with the FermiLarge Area Telescope. Physical Review Letters, 107(24):241302.

    [Albert et al., 2008] Albert, J., Aliu, E., Anderhub, H., Antoranz, P., and Armada, A. t.(2008). VHE γ-Ray Observation of the Crab Nebula and its Pulsar with the MAGICTelescope. ApJ, 674:1037–1055.

    [Aleksić et al., 2011] Aleksić, J., Alvarez, E. A., and Antonelli, L. A. t. (2011). Observationsof the Crab Pulsar between 25 and 100 GeV with the MAGIC I Telescope. ApJ, 742:43.

    [Aliu et al., 2008] Aliu, E., Anderhub, H., and Antonelli, L. A. t. (2008). Observation ofPulsed γ-Rays Above 25 GeV from the Crab Pulsar with MAGIC. Science, 322:1221–.

    [Arons, 1983] Arons, J. (1983). Pair creation above pulsar polar caps - Geometricalstructure and energetics of slot gaps. ApJ, 266:215–241.

    [Aschwanden, 2004] Aschwanden, M. J. (2004). Physics of the Solar Corona. An Introduc-tion. Praxis Publishing Ltd.

    [Baixeras, 2003] Baixeras, C. (2003). The MAGIC telescope. Nuclear Physics B ProceedingsSupplements, 114:247–252.

    [Bell, 1978] Bell, A. R. (1978). The acceleration of cosmic rays in shock fronts. I. MNRAS,182:147–156.

    [Bergström, 1999] Bergström, L. (1999). Non-baryonic dark matter. Nuclear Physics BProceedings Supplements, 70:31–42.

    [Bergström et al., 1998] Bergström, L., Ullio, P., and Buckley, J. H. (1998). Observabilityof γ rays from dark matter neutralino annihilations in the Milky Way halo. AstroparticlePhysics, 9:137–162.

    12

  • [Birkedal et al., 2005] Birkedal, A., Matchev, K. T., Perelstein, M., and Spray, A. (2005).Robust Gamma Ray Signature of WIMP Dark Matter. ArXiv High Energy Physics -Phenomenology e-prints.

    [Blandford and Ostriker, 1978] Blandford, R. D. and Ostriker, J. P. (1978). Particleacceleration by astrophysical shocks. ApJL, 221:L29–L32.

    [Bringmann et al., 2008] Bringmann, T., Bergström, L., and Edsjö, J. (2008). New gamma-ray contributions to supersymmetric dark matter annihilation. Journal of High EnergyPhysics, 1:49.

    [Bringmann et al., 2012] Bringmann, T., Huang, X., Ibarra, A., Vogl, S., and Weniger,C. (2012). Fermi LAT search for internal bremsstrahlung signatures from dark matterannihilation. JCAP, 7:54.

    [Bringmann and Weniger, 2012] Bringmann, T. and Weniger, C. (2012). Gamma raysignals from dark matter: Concepts, status and prospects. Physics of the Dark Universe,1:194–217.

    [Caraveo, 2014] Caraveo, P. A. (2014). Gamma-Ray Pulsar Revolution. ARA&A, 52:211–250.

    [Carretti et al., 2013] Carretti, E., Crocker, R. M., Staveley-Smith, L., Haverkorn, M.,Purcell, C., Gaensler, B. M., Bernardi, G., Kesteven, M. J., and Poppi, S. (2013). Giantmagnetized outflows from the centre of the Milky Way. Nature, 493:66–69.

    [Carroll and Ostlie, 2007] Carroll, B. and Ostlie, D. (2007). An Introduction to ModernAstrophysics. Pearson Addison-Wesley.

    [Clayton and Jin, 1995] Clayton, D. D. and Jin, L. (1995). Gamma Rays, Cosmic Rays,and Extinct Radioactivity in Molecular Clouds. ApJ, 451:681.

    [Harding and Muslimov, 1998] Harding, A. K. and Muslimov, A. G. (1998). Polar CAPModel for Pulsar High-Energy Emission. In Shibazaki, N., editor, Neutron Stars andPulsars: Thirty Years after the Discovery, page 311.

    [Harding et al., 2005] Harding, A. K., Usov, V. V., and Muslimov, A. G. (2005). High-Energy Emission from Millisecond Pulsars. ApJ, 622:531–543.

    [Hooper and Goodenough, 2011] Hooper, D. and Goodenough, L. (2011). Dark matterannihilation in the Galactic Center as seen by the Fermi Gamma Ray Space Telescope.Physics Letters B, 697:412–428.

    [Hooper and Slatyer, 2013] Hooper, D. and Slatyer, T. R. (2013). Two emission mecha-nisms in the Fermi Bubbles: A possible signal of annihilating dark matter. Physics ofthe Dark Universe, 2:118–138.

    [Hubbell, 2006] Hubbell, J. (2006). Electron–positron pair production by photons: Ahistorical overview. Radiation Physics and Chemistry, 75(6):614 – 623. Pair Production.

    [Kiziltan and Thorsett, 2009] Kiziltan, B. and Thorsett, S. E. (2009). Constraints onPulsar Evolution: The Joint Period-Spin-Down Distribution of Millisecond Pulsars.ApJL, 693:L109–L112.

    [Knödlseder, 2010] Knödlseder, J. (2010). Studies of Gamma Ray Sources with the FermiLarge Area Telescope. ArXiv e-prints.

    13

  • [Konopelko, 1999] Konopelko, A. K. (1999). Design studies for the future 50 GeV arraysof imaging air Čerenkov telescopes. Astroparticle Physics, 11:263–266.

    [Laurent et al., 2011] Laurent, P., Rodriguez, J., Wilms, J., Cadolle Bel, M., Pottschmidt,K., and Grinberg, V. (2011). Polarized Gamma-Ray Emission from the Galactic BlackHole Cygnus X-1. Science, 332:438–.

    [MAGIC Collaboration et al., 2014] MAGIC Collaboration, Aleksić, J., Ansoldi, S., andAntonelli, L. A. t. (2014). Measurement of the Crab Nebula spectrum over three decadesin energy with the MAGIC telescopes. ArXiv e-prints.

    [Navarro et al., 1996] Navarro, J. F., Frenk, C. S., and White, S. D. M. (1996). TheStructure of Cold Dark Matter Halos. ApJ, 462:563.

    [Ramaty and Mandzhavidze, 1998] Ramaty, R. and Mandzhavidze, N. (1998). Solar Flares:Gamma Rays. ArXiv Astrophysics e-prints.

    [Rando et al., 2013] Rando, R., Charles, E., and Digel, S. (2013). Fermi lat performance.

    [Ruffini et al., 2010] Ruffini, R., Vereshchagin, G., and Xue, S.-S. (2010). Electron-positronpairs in physics and astrophysics: From heavy nuclei to black holes. PhysREP, 487:1–140.

    [Su et al., 2010] Su, M., Slatyer, T. R., and Finkbeiner, D. P. (2010). Giant Gamma-rayBubbles from Fermi-LAT: Active Galactic Nucleus Activity or Bipolar Galactic Wind?ApJ, 724:1044–1082.

    [Tam et al., 2012] Tam, P. H. T., Kong, A. K. H., and Hui, C. Y. (2012). Gamma-rayemission from globular clusters. ArXiv e-prints.

    [Tavani et al., 2009] Tavani, M., Barbiellini, G., Argan, A., Boffelli, F., and Bulgarelli,A. et al . (2009). The AGILE Mission. A&A, 502:995–1013.

    [Teegarden and Sturner, 1999] Teegarden, B. J. and Sturner, S. J. (1999). INTEGRALObservations of Gamma-Ray Bursts. In AAS/High Energy Astrophysics Division #4,volume 31 of Bulletin of the American Astronomical Society, page 717.

    [Tempel et al., 2012] Tempel, E., Hektor, A., and Raidal, M. (2012). Fermi 130 GeVgamma-ray excess and dark matter annihilation in sub-haloes and in the Galactic centre.JCAP, 9:32.

    [Weniger, 2012] Weniger, C. (2012). A tentative gamma-ray line from Dark Matter anni-hilation at the Fermi Large Area Telescope. JCAP, 8:7.

    [Wijnands and van der Klis, 1998] Wijnands, R. and van der Klis, M. (1998). A millisec-ond pulsar in an X-ray binary system. Nature, 394:344–346.

    [Zubovas et al., 2011] Zubovas, K., King, A. R., and Nayakshin, S. (2011). The MilkyWay’s Fermi bubbles: echoes of the last quasar outburst? NMRAS, 415:L21–L25.

    14