The Fermi bubbles are two large structures in the gamma-ray sky extending to 55 • above and below the Galactic center. We analyze 50 months of Fermi Large Area Telescope data between 100 MeV and 500 GeV above 10 • in Galactic latitude to derive the spectrum and morphology of the Fermi bubbles. We thoroughly explore the systematic uncertainties that arise when modeling the Galactic diffuse emission through two separate approaches. The gamma-ray spectrum is well described by either a log parabola or a power law with an exponential cutoff. We exclude a simple power law with more than 7σ significance. The power law with an exponential cutoff has an index of 1.9 ± 0.2 and a cutoff energy of 110 ± 50 GeV. We find that the gamma-ray luminosity of the bubbles is 4.4 +2.4 −0.9 × 10 37 erg s −1. We confirm a significant enhancement of gamma-ray emission in the southeastern part of the bubbles, but we do not find significant evidence for a jet. No significant variation of the spectrum across the bubbles is detected. The width of the boundary of the bubbles is estimated to be 3.4 +3.7 −2.6 deg. Both inverse Compton (IC) models and hadronic models including IC emission from secondary leptons fit the gamma-ray data well. In the IC scenario, synchrotron emission from the same population of electrons can also explain the WMAP and Planck microwave haze with a magnetic field between 5 and 20 μG.

We present Suzaku X-ray observations along two edge regions of the Fermi Bubbles, with eight 20 ks pointings across the northern part of the North Polar Spur (NPS) surrounding the north bubble and six across the southernmost edge of the south bubble. After removing compact X-ray features, diffuse X-ray emission is clearly detected and is well reproduced by a three-component spectral model consisting of unabsorbed thermal emission (temperature kT 0.1 keV) from the Local Bubble, absorbed kT 0.3 keV thermal emission related to the NPS and/or Galactic halo (GH), and a power-law component at a level consistent with the cosmic X-ray background. The emission measure (EM) of the 0.3 keV plasma decreases by 50% toward the inner regions of the northeast bubble, with no accompanying temperature change. However, such a jump in the EM is not clearly seen in the south bubble data. While it is unclear whether the NPS originates from a nearby supernova remnant or is related to previous activity within or around the Galactic center, our Suzaku observations provide evidence that suggests the latter scenario. In the latter framework, the presence of a large amount of neutral matter absorbing the X-ray emission as well as the existence of the kT 0.3 keV gas can be naturally interpreted as a weak shock driven by the bubbles' expansion in the surrounding medium, with velocity v exp ∼ 300 km s −1 (corresponding to shock Mach number M 1.5), compressing the GH gas to form the NPS feature. We also derived an upper limit for any non-thermal X-ray emission component associated with the bubbles and demonstrate that, in agreement with the aforementioned findings, the non-thermal pressure and energy estimated from a one-zone leptonic model of its broadband spectrum, are in rough equilibrium with that of the surrounding thermal plasma.

In our previous works (Kataoka et al. 2013, we found absorbed thermal X-ray plasma with kT ≃ 0.3 keV observed ubiquitously near the edges of the Fermi bubbles and interpreted this emission as weakly shock-heated Galactic halo (GH) gas. Here we present a systematic and uniform analysis of archival Suzaku (29 pointings; 6 newly presented) and Swift (68 pointings; 49 newly presented) data within Galactic longitudes |l| < 20 • and latitude 5 • |b| < 60 • , covering the whole extent of the Fermi bubbles. We show that the plasma temperature is constant at kT ≃ 0.30±0.07 keV, while the emission measure (EM) varies by an order of magnitude, increasing toward the Galactic center (i.e., low |b|) with enhancements at the north polar spur (NPS), SE-claw and NW-clump features. Moreover, the EM distribution of kT ≃ 0.30 keV plasma is highly asymmetric in the northern and southern bubbles. Although the association of the X-ray emission with the bubbles is not conclusive, we compare the observed EM properties with simple models assuming (i) a filled halo without bubbles, whose gas density follows a hydrostatic isothermal model (King profile) and (ii) a bubble-in-halo in which two identical bubbles expand into the halo forming thick shells of swept halo gas. We argue that the EM profile in the north (b > 0 • ) favors (ii), whereas that of the south (b < 0 • ) is rather close to (i), but weak excess signature is clearly detected also in the south like NPS (South Polar Spur; SPS). Such an asymmetry, if due to the bubbles, cannot be fully understood only by the inclination of bubbles' axis against the Galactic disk normal, thus suggesting asymmetric outflow due to different environmental/initial condition.

No abstract

We present multi-wavelength monitoring results for the broad-line radio galaxy 3C 120 in the MeV/GeV, submillimeter, and 43 GHz bands over 6 yr. Over the past 2 yr, the Fermi-Large Area Telescope sporadically detected 3C 120 with high significance and the 230 GHz data also suggest an enhanced activity of the source. After the MeV/GeV detection from 3C 120 in MJD 56240-56300, 43 GHz Very Long Baseline Array (VLBA) monitoring revealed a brightening of the radio core, followed by the ejection of a superluminal knot. Since we observed the γ-ray and VLBA phenomena in temporal proximity to each other, it is naturally assumed that they are physically connected. This assumption was further supported by the subsequent observation that the 43 GHz core brightened again after a γ-ray flare occurred around MJD 56560. We can then infer that the MeV/GeV emission took place inside an unresolved 43 GHz core of 3C 120 and that the jet dissipation occurred at sub-parsec distances from the central black hole (BH), if we take the distance of the 43 GHz core from the central BH as ∼0.5 pc, as previously estimated from the time lag between X-ray dips and knot ejections. Based on our constraints on the relative locations of the emission regions and energetic arguments, we conclude that the γ rays are more favorably produced via the synchrotron self-Compton process, rather than inverse Compton scattering of external photons coming from the broad line region or hot dusty torus. We also derived the electron distribution and magnetic field by modeling the simultaneous broadband spectrum.