T. A. Enßlin

Max Planck Institute for Astrophysics, Arching, Bavaria, Germany

Are you T. A. Enßlin?

Claim your profile

Publications (269)507.69 Total impact

  • Source
    Planck, Fermi Collaborations, P. A. R. Ade, N. Aghanim, G. Aniano, M. Arnaud, M. Ashdown, J. Aumont, C. Baccigalupi, A. J. Banday, [......], J. Valiviita, B. Van Tent, P. Vielva, F. Villa, L. A. Wade, B. D. Wandelt, I. K. Wehus, D. Yvon, A. Zacchei, A. Zonca
    [Show abstract] [Hide abstract]
    ABSTRACT: Shortened abstract: Observations of the nearby Chamaeleon clouds in gamma rays with the Fermi Large Area Telescope and in thermal dust emission with Planck and IRAS have been used with the HI and CO radio data to (i) map the gas column densities in the different phases and at the dark neutral medium (DNM) transition between the HI-bright and CO-bright media; (ii) constrain the CO-to-$H_2$ conversion factor, $X_{CO}$; (iii) probe the dust properties per gas nucleon in each gas phase and spatially across the clouds. We have separated clouds in velocity in HI and CO emission and modelled the 0.4-100 GeV intensity, the dust optical depth at 353 GHz, the thermal radiance of the large grains, and an estimate of the dust extinction empirically corrected for the starlight intensity, $A_{VQ}$. The gamma-ray emissivity spectra confirm that the GeV-TeV cosmic rays uniformly permeate all gas phases up to the CO cores. The dust and cosmic rays reveal large amounts of DNM gas, with comparable spatial distributions and twice as much mass as in the CO-bright clouds. We give constraints on the HI-DNM-CO transitions and CO-dark $H_2$ fractions for separate clouds. The corrected extinction provides the best fit to the total gas traced by the gamma rays, but we find evidence for a rise in $A_{VQ}/N_H$ and a steep rise in opacity, with increasing $N_H$ and $H_2$ fraction, and with decreasing dust temperature. We observe less variations for the specific power of the grains, except for a decline by half in the CO cores. This combined information suggests grain evolution. The gamma rays and dust radiance yield consistent $X_{CO}$ estimates near $0.7\times10^{20}$ cm$^{-2}$ (K km/s)$^{-1}$. The other dust tracers yield biased values because of the grain opacity rise in the CO clouds. These results also confirm a factor of 2 difference between $X_{CO}$ estimates at pc and kpc scales.
    09/2014;
  • Source
    P. A. R. Ade, N. Aghanim, M. I. R. Alves, G. Aniano, M. Arnaud, M. Ashdown, J. Aumont, C. Baccigalupi, A. J. Banday, R. B. Barreiro, [......], B. Van Tent, P. Vielva, F. Villa, L. A. Wade, B. D. Wandelt, I. K. Wehus, N. Ysard, D. Yvon, A. Zacchei, A. Zonca
    [Show abstract] [Hide abstract]
    ABSTRACT: We present all-sky dust modelling of the high resolution Planck, IRAS and WISE infrared (IR) observations using the physical dust model presented by Draine & Li in 2007 (DL). We study the performance of this model and present implications for future dust modelling. The present work extends to the full sky the dust modelling carried out on nearby galaxies using Herschel and Spitzer data. We employ the DL dust model to generate maps of the dust mass surface density, the dust optical extinction AV, and the starlight intensity heating the bulk of the dust, parametrized by Umin. We test the model by comparing these maps with independent estimates of the dust optical extinction AV . In molecular clouds, we compare the DL AV estimates with maps generated from stellar optical observations from the 2MASS survey. The DL AV estimates are a factor of about 3 larger than values estimated from 2MASS observations. In the diffuse interstellar medium (ISM) we compare the DL optical extinction AV estimates with optical estimates from approximately 200,000 QSOs observed in the Sloan digital sky survey. The DL AV estimates are larger than those determined from the QSOs, and this discrepancy depends on Umin. We propose an empirical renormalization of the DL AV estimate, dependent of Umin, which compensates for the systematic differences found here. This renormalization, bringing into agreement the AV estimates on QSOs, also brings into agreement the AV estimates on molecular clouds. In the diffuse ISM, the DL fitting parameter Umin, appears to trace variations in the far-IR opacity of the dust grains. Therefore, some of the physical assumptions of the DL model need to be revised. We provide a family of SEDs normalized by optical reddening, parameterized by Umin; these will be the constraints for a next generation of dust models.
    09/2014;
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present the results of approximately three years of observations of Planck Sunyaev-Zeldovich (SZ) sources with the Russian-Turkish 1.5-m telescope (RTT150), as a part of the optical follow-up programme undertaken by the Planck collaboration. During this time period approximately 20% of all dark and grey clear time available at the telescope was devoted to observations of Planck objects. Some observations of distant clusters were also done at the 6-m Bolshoy Telescope Azimutal'ny (BTA) of the Special Astrophysical Observatory of the Russian Academy of Sciences. In total, deep, direct images of more than one hundred fields were obtained in multiple filters. We identified 47 previously unknown galaxy clusters, 41 of which are included in the Planck catalogue of SZ sources. The redshifts of 65 Planck clusters were measured spectroscopically and 14 more were measured photometrically. We discuss the details of cluster optical identifications and redshift measurements. We also present new spectroscopic redhifts for 39 Planck clusters that were not included in the Planck SZ source catalogue and are published here for the first time.
    07/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The Andromeda Galaxy (M31) is one of a few galaxies that has sufficient angular size on the sky to be resolved by the Planck satellite. Planck has detected M31 in all of its frequency bands, and has mapped out the dust emission with the High Frequency Instrument, clearly resolving multiple spiral arms and sub-features. We examine the morphology of this long-wavelength dust emission as seen by Planck, including a study of its outermost spiral arms, and investigate the dust heating mechanism across M31. We find that dust dominating the longer wavelength emission ($\gtrsim 0.3$ mm) is heated by the diffuse stellar population (as traced by 3.6 $\mu$m emission), with the dust dominating the shorter wavelength emission heated by a mix of the old stellar population and star-forming regions (as traced by 24 $\mu$m emission). We also fit spectral energy distributions (SEDs) for individual 5' pixels and quantify the dust properties across the galaxy, taking into account these different heating mechanisms, finding that there is a linear decrease in temperature with galactocentric distance for dust heated by the old stellar population, as would be expected, with temperatures ranging from around 22 K in the nucleus to 14 K outside of the 10 kpc ring. Finally, we measure the integrated spectrum of the whole galaxy, which we find to be well-fitted with a global dust temperature of ($18.9\pm0.9$) K with a spectral index of $1.61\pm0.11$ (assuming a single modified blackbody), and a significant amount of free-free emission at intermediate frequencies, which when converted into a star formation rate agrees well with the star formation estimate from H$\alpha$ emission of 0.4$M_\odot$ yr$^{-1}$. We see no evidence for spinning dust emission, with a 3$\sigma$ upper limit of 1.26 Jy in the 20-60 GHz band.
    07/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The grand-design spiral galaxy M51 was observed with the LOFAR High Frequency Antennas (HBA) and imaged in total intensity and polarisation. This observation covered the frequencies between 115 MHz and 175 MHz. We produced an image of total emission of M51 at the mean frequency of 151 MHz with 20 arcsec resolution and 0.3 mJy rms noise, which is the most sensitive image of a galaxy at frequencies below 300 MHz so far. The integrated spectrum of total radio emission is described well by a power law, while flat spectral indices in the central region indicate thermal absorption. We observe that the disk extends out to 16 kpc and see a break in the radial profile near the optical radius of the disk. Our main results, the scale lengths of the inner and outer disks at 151 MHz and 1.4 GHz, arm--interarm contrast, and the break scales of the radio--far-infrared correlations, can be explained consistently by CRE diffusion, leading to a longer propagation length of CRE of lower energy. The distribution of CRE sources drops sharply at about 10 kpc radius, where the star formation rate also decreases sharply. We find evidence that thermal absorption is primarily caused by HII regions. The non-detection of polarisation from M51 at 151 MHz is consistent with the estimates of Faraday depolarisation. Future searches for polarised emission in this frequency range should concentrate on regions with low star formation rates.
    07/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The presence of multiple fields during inflation might seed a detectable amount of non-Gaussianity in the curvature perturbations, which in turn becomes observable in present data sets like the cosmic microwave background (CMB) or the large scale structure (LSS). Within this proceeding we present a fully analytic method to infer inflationary parameters from observations by exploiting higher-order statistics of the curvature perturbations. To keep this analyticity, and thereby to dispense with numerically expensive sampling techniques, a saddle-point approximation is introduced whose precision has been validated for a numerical toy example. Applied to real data, this approach might enable to discriminate among the still viable models of inflation.
    07/2014;
  • Source
    Niels Oppermann, Torsten A. Enßlin
    [Show abstract] [Hide abstract]
    ABSTRACT: The extraction of foreground and CMB maps from multi-frequency observations relies mostly on the different frequency behavior of the different components. Existing Bayesian methods additionally make use of a Gaussian prior for the CMB whose correlation structure is described by an unknown angular power spectrum. We argue for the natural extension of this by using non-trivial priors also for the foreground components. Focusing on diffuse Galactic foregrounds, we propose a log-normal model including unknown spatial correlations within each component and cross-correlations between the different foreground components. We present case studies at low resolution that demonstrate the superior performance of this model when compared to an analysis with flat priors for all components.
    07/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Any variation of the fundamental physical constants, and more particularly of the fine structure constant, $\alpha$, or of the mass of the electron, $m_e$, would affect the recombination history of the Universe and cause an imprint on the cosmic microwave background angular power spectra. We show that the Planck data allow one to improve the constraint on the time variation of the fine structure constant at redshift $z\sim 10^3$ by about a factor of 5 compared to WMAP data, as well as to break the degeneracy with the Hubble constant, $H_0$. In addition to $\alpha$, we can set a constraint on the variation of the mass of the electron, $m_{\rm e}$, and on the simultaneous variation of the two constants. We examine in detail the degeneracies between fundamental constants and the cosmological parameters, in order to compare the limits obtained from Planck and WMAP and to determine the constraining power gained by including other cosmological probes. We conclude that independent time variations of the fine structure constant and of the mass of the electron are constrained by Planck to ${\Delta\alpha}/{\alpha}= (3.6\pm 3.7)\times10^{-3}$ and ${\Delta m_{\rm e}}/{m_{\rm e}}= (4 \pm 11)\times10^{-3}$ at the 68% confidence level. We also investigate the possibility of a spatial variation of the fine structure constant. The relative amplitude of a dipolar spatial variation of $\alpha$ (corresponding to a gradient across our Hubble volume) is constrained to be $\delta\alpha/\alpha=(-2.4\pm 3.7)\times 10^{-2}$.
    06/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Planck data when combined with ancillary data provide a unique opportunity to separate the diffuse emission components of the inner Galaxy. The purpose of the paper is to elucidate the morphology of the various emission components in the strong star-formation region lying inside the solar radius and to clarify the relationship between the various components. The region of the Galactic plane covered is l=300-0-60deg where star-formation is highest and the emission is strong enough to make meaningful component separation. The latitude widths in this longitude range lie between 1deg and 2deg, which correspond to FWHM z-widths of 100-200pc at a typical distance of 6kpc. The four emission components studied here are synchrotron, free-free, anomalous microwave emission (AME), and thermal (vibrational) dust emission. These components are identified by constructing spectral energy distributions (SEDs) at positions along the Galactic plane using the wide frequency coverage of Planck (28.4-857GHz) in combination with low-frequency radio data at 0.408-2.3GHz plus WMAP data at 23-94GHz, along with far-infrared (FIR) data from DIRBE and IRAS. The free-free component is determined from radio recombination line (RRL) data. AME is found to be comparable in brightness to the free-free emission on the Galactic plane in the frequency range 20-40GHz with a width in latitude similar to that of the thermal dust; it comprises 45+/-1% of the total 28.4GHz emission in the longitude range l=300-0-60deg. The free-free component is the narrowest, reflecting the fact that it is produced by current star-formation as traced by the narrow distribution of OB stars. It is the dominant emission on the plane between 60 and 100GHz. RRLs from this ionized gas are used to assess its distance, leading to a free-free z-width of FWHM ~100pc...(abridged)
    06/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents the large-scale polarized sky as seen by Planck HFI at 353 GHz, which is the most sensitive Planck channel for dust polarization. We construct and analyse large-scale maps of dust polarization fraction and polarization direction, while taking account of noise bias and possible systematic effects. We find that the maximum observed dust polarization fraction is high (pmax > 18%), in particular in some of the intermediate dust column density (AV < 1mag) regions. There is a systematic decrease in the dust polarization fraction with increasing dust column density, and we interpret the features of this correlation in light of both radiative grain alignment predictions and fluctuations in the magnetic field orientation. We also characterize the spatial structure of the polarization angle using the angle dispersion function and find that, in nearby fields at intermediate latitudes, the polarization angle is ordered over extended areas that are separated by filamentary structures, which appear as interfaces where the magnetic field sky projection rotates abruptly without apparent variations in the dust column density. The polarization fraction is found to be anti-correlated with the dispersion of the polarization angle, implying that the variations are likely due to fluctuations in the 3D magnetic field orientation along the line of sight sampling the diffuse interstellar medium.We also compare the dust emission with the polarized synchrotron emission measured with the Planck LFI, with low-frequency radio data, and with Faraday rotation measurements of extragalactic sources. The two polarized components are globally similar in structure along the plane and notably in the Fan and North Polar Spur regions. A detailed comparison of these three tracers shows, however, that dust and cosmic rays generally sample different parts of the line of sight and confirms that much of the variation observed in the Planck data is due to the 3D structure of the magnetic field.
    05/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Polarized emission observed by Planck HFI at 353 GHz towards a sample of nearby fields is presented, focusing on the statistics of polarization fractions $p$ and angles $\psi$. The polarization fractions and column densities in these nearby fields are representative of the range of values obtained over the whole sky. We find that: (i) the largest polarization fractions are reached in the most diffuse fields; (ii) the maximum polarization fraction $p_\mathrm{max}$ decreases with column density $N_\mathrm{H}$ in the more opaque fields with $N_\mathrm{H} > 10^{21}\,\mathrm{cm}^{-2}$; and (iii) the polarization fraction along a given line of sight is correlated with the local spatial coherence of the polarization angle. These observations are compared to polarized emission maps computed in simulations of anisotropic magnetohydrodynamical (MHD) turbulence in which we assume a uniform intrinsic polarization fraction of the dust grains. We find that an estimate of this parameter may be recovered from the maximum polarization fraction $p_\mathrm{max}$ in diffuse regions where the magnetic field is ordered on large scales and perpendicular to the line of sight. This emphasizes the impact of anisotropies of the magnetic field on the emerging polarization signal. The decrease of the polarization fraction with column density in nearby molecular clouds is well reproduced in the simulations, indicating that it is essentially due to the turbulent structure of the magnetic field: an accumulation of variously polarized structures along the line of sight leads to such an anti-correlation. In the simulations, polarization fractions are also found to anti-correlate with the angle dispersion function $\Delta\psi$. [abridged]
    05/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Planck has mapped the intensity and polarization of the sky at microwave frequencies with unprecedented sensitivity. We make use of the Planck 353 GHz I, Q, and U Stokes maps as dust templates, and cross-correlate them with the Planck and WMAP data at 12 frequencies from 23 to 353 GHz, over circular patches with 10 degree radius. The cross-correlation analysis is performed for both intensity and polarization data in a consistent manner. We use a mask that focuses our analysis on the diffuse interstellar medium at intermediate Galactic latitudes. We determine the spectral indices of dust emission in intensity and polarization between 100 and 353 GHz, for each sky-patch. The mean values, $1.63\pm0.03$ for polarization and $1.52\pm0.02$ for intensity, for a mean dust temperature of 18.7 K, are close, but significantly different. We determine the mean spectral energy distribution (SED) of the microwave emission, correlated with the 353 GHz dust templates, by averaging the results of the correlation over all sky-patches. We find that the mean SED increases for decreasing frequencies at $\nu < 60$ GHz, for both intensity and polarization. The rise of the polarization SED towards low frequencies may be accounted for by a synchrotron component correlated with dust, with no need for any polarization of the anomalous microwave emission. We use a spectral model to separate the synchrotron and dust polarization and to characterize the spectral dependence of the dust polarization fraction. The polarization fraction ($p$) of the dust emission decreases by $(34\pm10)$ % from 353 to 70 GHz. The decrease of $p$ could indicate differences in polarization efficiency among components of interstellar dust (e.g., carbon versus silicate grains), or, alternatively, it could be a signature of magnetic dipole emission from ferromagnetic inclusions within interstellar grains.
    05/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Planck survey provides unprecedented full-sky coverage of the submillimetre polarized emission from Galactic dust, bringing new constraints on the properties of dust. The dust grains that emit the radiation seen by Planck in the submillimetre also extinguish and polarize starlight in the optical. Using ancillary catalogues of interstellar polarization and extinction of starlight, we obtain the degree of polarization, $p_V$, and the optical depth in the $V$ band to the star, $\tau_V$. We extract the submillimetre polarized intensity, $P_S$, and total intensity, $I_S$, measured toward these stars in the Planck 353 GHz channel. We compare the position angle measured in the optical with that measured at 353 GHz, and the column density measure $E(B - V)$ with that inferred from the Planck product map of the submillimetre dust optical depth. For those lines of sight suitable for this comparison, we measure the polarization ratios $R_{S/V} = (P_S/I_S)/(p_V/\tau_V)$ and $R_{P/p} = P_S / p_V$ through a correlation analysis. We find $R_{S/V} = 4.3$ with statistical and systematic uncertainties 0.2 and 0.4, respectively, and $R_{P/p} = 5.6$ MJy sr$^{-1}$, with statistical and systematic uncertainties 0.2 and 0.4 MJy sr$^{-1}$, respectively. Our estimate of $R_{S/V}$ is reasonably compatible with current dust models, not yet very discriminating among them. However, the observed $R_{P/p}$ is a more discriminating diagnostic for the polarizing grain population and is not compatible with predictions of dust models, the observations being higher by a factor of about 2.5. These new diagnostics from Planck, including the spectral dependence in the submillimetre, will be important for constraining and understanding the full complexity of the grain models, and for further interpretation of the Planck thermal dust polarization.
    05/2014;
  • Source
    ArXiv e-prints. 05/2014;
  • ArXiv e-prints. 05/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: (abridged) Observations of Faraday rotation for extragalactic sources probe magnetic fields both inside and outside the Milky Way. Building on our earlier estimate of the Galactic foreground (Oppermann et al., 2012), we set out to estimate the extragalactic contributions. We discuss different strategies and the problems involved. In particular, we point out that taking the difference between the observed values and the Galactic foreground reconstruction is not a good estimate for the extragalactic contributions. We present a few possibilities for improved estimates using the existing foreground map, allowing for imperfectly described observational noise. In this context, we point out a degeneracy between the contributions to the observed values due to extragalactic magnetic fields and observational noise and comment on the dangers of over-interpreting an estimate without taking into account its uncertainty information. Finally, we develop a reconstruction algorithm based on the assumption that the observational uncertainties are accurately described for a subset of the data, which can overcome this degeneracy. We demonstrate its performance in a simulation, yielding a high quality reconstruction of the Galactic Faraday depth, a precise estimate of the typical extragalactic contribution, and a well-defined probabilistic description of the extragalactic contribution for each source. We apply this reconstruction technique to a catalog of Faraday rotation observations. We vary our assumptions about the data, showing that the dispersion of extragalactic contributions to observed Faraday depths is likely lower than 7 rad/m^2, in agreement with earlier results, and that the extragalactic contribution to an individual data point is poorly constrained by the data in most cases. Posterior samples for the extragalactic contributions and all results of our fiducial model are provided online.
    04/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: (abridged) Observations of Faraday rotation for extragalactic sources probe magnetic fields both inside and outside the Milky Way. Building on our earlier estimate of the Galactic foreground (Oppermann et al., 2012), we set out to estimate the extragalactic contributions. We discuss different strategies and the problems involved. In particular, we point out that taking the difference between the observed values and the Galactic foreground reconstruction is not a good estimate for the extragalactic contributions. We present a few possibilities for improved estimates using the existing foreground map, allowing for imperfectly described observational noise. In this context, we point out a degeneracy between the contributions to the observed values due to extragalactic magnetic fields and observational noise and comment on the dangers of over-interpreting an estimate without taking into account its uncertainty information. Finally, we develop a reconstruction algorithm based on the assumption that the observational uncertainties are accurately described for a subset of the data, which can overcome this degeneracy. We demonstrate its performance in a simulation, yielding a high quality reconstruction of the Galactic Faraday depth, a precise estimate of the typical extragalactic contribution, and a well-defined probabilistic description of the extragalactic contribution for each source. We apply this reconstruction technique to a catalog of Faraday rotation observations. We vary our assumptions about the data, showing that the dispersion of extragalactic contributions to observed Faraday depths is likely lower than 7 rad/m^2, in agreement with earlier results, and that the extragalactic contribution to an individual data point is poorly constrained by the data in most cases. Posterior samples for the extragalactic contributions and all results of our fiducial model are provided online.
    03/2014;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a generic inference method for inflation models from observational data by the usage of higher-order statistics of the curvature perturbation on uniform density hypersurfaces. This method is based on the calculation of the posterior for the primordial non-Gaussianity parameters $f_\text{NL}$ and $g_\text{NL}$, which in general depend on specific parameters of inflation and reheating models, and enables to discriminate among the still viable inflation models. To keep analyticity as far as possible to dispense with numerically expensive sampling techniques a saddle-point approximation is introduced, whose precision is validated for a numerical toy example. The mathematical formulation is done in a generic way so that the approach remains applicable to cosmic microwave background data as well as to large scale structure data. Additionally, we review a few currently interesting inflation models and present numerical toy examples thereof in two and three dimensions to demonstrate the efficiency of the higher-order statistics method. A second quantity of interest is the primordial power spectrum. Here, we present two Bayesian methods to infer it from observational data, the so called critical filter and an extension thereof with smoothness prior, both allowing for a non-parametric spectrum reconstruction. These methods are able to reconstruct the spectra of the observed perturbations and the primordial ones of curvature perturbation even in case of non-Gaussianity and partial sky coverage. We argue that observables like $T-$ and $B-$modes permit to measure both spectra. This also allows to infer the level of non-Gaussianity generated since inflation.
    Journal of Cosmology and Astroparticle Physics 03/2014; 2014(06). · 6.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Late stages of stellar evolution are characterized by copious mass-loss events whose signature is the formation of circumstellar envelopes (CSE). Planck multi frequency measurements have provided relevant information on a sample of Galactic planetary nebulae (PNe) in this important and relatively unexplored observational band between 30 and 857GHz. Planck enables the assembly of comprehensive PNe spectral energy distributions (SEDs) from radio through to far-infrared frequencies. Modelling of the derived SEDs provides us with information on physical properties of CSEs and the mass content of both main components: ionized gas, traced by the free-free emission at cm mm waves; and thermal dust, traced by the millimetre and far IR emission. In particular, the amount of ionized gas and dust has been derived here. Such quantities have also been estimated for the very young PN CRL618, where the strong variability observed in its radio and millimetre emission has previously prevented the construction of its SED. A morphological study of the Helix Nebula has also been performed. Planck maps reveal, for the first time, the spatial distribution of the dust inside the envelope, allowing us to identify different components, the most interesting of which is a very extended component (up to 1 pc) that may be related to a region where the slow expanding envelope is interacting with the surrounding interstellar medium. Moreover, in this object, thermal dust and H2 emission appear to be spatially correlated, providing hints of H2 formation on dust grain surfaces.
    03/2014;

Publication Stats

3k Citations
507.69 Total Impact Points

Institutions

  • 2002–2013
    • Max Planck Institute for Astrophysics
      Arching, Bavaria, Germany
  • 2010
    • Scuola Internazionale Superiore di Studi Avanzati di Trieste
      Trst, Friuli Venezia Giulia, Italy
  • 2004
    • Harvard-Smithsonian Center for Astrophysics
      • Smithsonian Astrophysical Observatory
      Cambridge, Massachusetts, United States
  • 2000
    • University of Toronto
      Toronto, Ontario, Canada
  • 1996–1997
    • Max Planck Institute for Radio Astronomy
      Bonn, North Rhine-Westphalia, Germany