Filename
stringlengths 22
64
| Paragraph
stringlengths 8
5.57k
|
|---|---|
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | where Mp and M(cid:63) are the mass of the planet and star, αSS is the Shakura & Sunyaev (1973) viscosity parame- ter, H is the disk scale height and rp is the semi-major axis of the planet. Our previous simulations (Barrière- Fouchet et al. 2005, hereafter BF05) show the settling rate – and hence the thickness of the dust layer – de- pends on grain size. Since the gap criterion depends on the disk scale height, this would suggest that it is easier to create and sustain a gap in the dust layer than in the gas. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | There are a variety of observational signature of planetary gaps, including mid-infrared dips (e.g. Cal- vet et al. 2002; Rice et al. 2003) and direct scattered light (Weinberger et al. 1999; Schneider et al. 2006) and sub-millimetre observations of protoplanetary disks (Ozernoy et al. 2000; Wilner et al. 2002). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | tially as grains settle to the mid-plane and migrate ra- dially (BF05). As well as the gas-to-dust ratio varying throughout the disk, we expect that the effects of plan- etary gaps will be stronger in the dust phase than in the gas phase, which will further affect observations. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | In this paper we study the formation of a gap trig- gered by an embedded planet in the dust layer of a protoplanetary disk. We will study the effects of planet mass and grain size on gap formation and evolution in 3D dusty-gas protoplanetary disks. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | where ρg is the gas density, s is the (spherical) grain radius, c is the sound speed, and v is the velocity dif- ference between dust and gas. For details of how the equations of motion and the density of the two fluids are calculated, we refer the reader to BF05. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | The dust particles are incompressible (ρd = con- stant) and there is no grain evaporation or coagulation, nor any gas condensation. All simulations presented consider just one (spherical) grain size at a time. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | We set up a disk of gas and dust with a total mass of Mdisk around a 1M(cid:12) star with an embedded planet of mass Mp at a distance of rp. The dust phase is 1% of the total disk mass and the system is evolved for about 100 planetary orbits. The disk equation of state is isothermal with constant vertical temperature and radial profile T ∝ r−1. Code units are set by G = M(cid:63) = rp = 1 and the isothermal sound speed at r = 1 is c = H/r = 0.05. The initial density profile is constant and the dust density is ρd (see below for details). The planet is treated as a point mass particle which moves under the gravitational influence of the star on a fixed circular orbit (i.e. no migration). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Simulations start with 50,000 gas and 50,000 dust particles. Particles are removed from the simulations if they cross the Hill radius of the planet, get closer than 0.4 code units of the central star (which sets the inner disk edge), or if they escape past 4 code units (which sets the outer disk edge). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | minimum mass solar nebula (MMSN model), and (2) a classical T Tauri star disk (CTTS model). The MMSN disk has a mass of 2.9 10−3 M(cid:12) and extends from 2 AU to 20 AU. The standard MMSN model has a 1 Jupiter mass (MJup) planet on a circular orbit of radius 5.2 AU in a disk containing grains 1 m in size and ρd = 1.25 g cm−3. The CTTS disk mass is 0.01 M(cid:12) and spans 16 AU to 150 AU in radius. The standard CTTS model has a 5 MJup planet on a circular orbit of radius 40 AU in a disk containing grains 1 mm in size and ρd = 1.0 g cm−3. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | For both disks we start by running the standard model and compare the evolution of the gap in the gas and dust phases. We then run a series of experiments to study the effect of grain size in the dust disk for both models, with s = 1 cm, 10 cm and 1 m for the MMSN disk and s = 100 µm, 1 mm and 1 cm for the CTTS disk. (Because the nebula conditions and par- ticularly the density in the CTTS and MMSN models are different, the grain sizes used in the two models are different in order to obtain similar values for the gas drag and hence dust settling and migration rates.) This is followed by a series of experiments that study the effects of planetary mass on gap formation and evo- lution, with Mp = 0.05, 0.1, 0.2, 0.5 and 1.0 MJup for the MMSN model and Mp =0.1, 0.5, 1.0 and 5.0 MJup for the CTTS model. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | The results of the standard MMSN model is shown in Fig. 1. The left panel shows the top-down (x, y) and side-on (r, z) view of the gas and dust disk morpholo- gies. While the planet opens a gap in the gas, the gap in the dust layer is much more striking. The right panel of Fig. 1 compares the evolution of the azimuthally av- eraged surface density profile of the gas and dust after 8.4 and 104 planetary orbits. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Recent observations have suggested that massive planets at large distances from the star may exist, such as a 5 MJup planet 30 AU from LkCa 15 (Pietuet al. 2006) and a 12.5 MJup planet 135 AU from GG Tau (Beust & Dutrey 2005). Our standard CTTS model has a 5 MJup planet at 40 AU and the results are shown in Fig. 2. For the nebular parameters used, such a mas- sive planet almost completely empties the inner disk of both gas and dust (though this is likely due to the large inner disk radius - see Crida & Morbidelli 2007). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Fig. 1.— Left panel: End state of the standard MMSN model after 104 orbits (1,230 years), showing top-down and side-on views of the disk. The top row shows the gas disk, while the bottom row shows the (coloured) dust overlaid on the gas (grey). Right panel: Azimuthally averaged surface density profiles for a gap created in the standard MMSN model, showing the comparison of the gas and dust surface densities after 8.4 and 104 orbits. The gas density is scaled by 0.01 for direct comparison to the dust. The vertical lines are, from left to right, the 1:2 and 2:3 internal resonances, the 1:1 planetary orbit, and the 3:2 and 2:1 external resonances respectively. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | size, we ran a series of simulations to determine the effect of grain size on the gap. Three grain sizes are tested in both disk models: 1 cm, 10 cm and 1 m for the MMSN disk and 100 µm, 1 mm and 1 cm for the CTTS disk. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Fig. 3 shows how the gap morphology of both the MMSN and CTTS disks vary with dust grains size. We find that both the width and depth of the gap increases with increasing grain size. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Structures created by planets in dusty disks are more diverse than those created in the gaseous disks. With only aerodynamic drag, we find that it is possible to create disks with a large central hole or a ring. Rice et al. (2006) also found that the presence of a planet can produce disks with a central hole for certain grain sizes. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Finally we investigate the effect of the planetary mass on the evolution of the gap. For our MMSN disk with 1 m grains, Mp varies from 0.05 MJup to 1 MJup, and for our CTTS disk containing 1 mm grains, Mp varies from 0.1, 0.5, 1 to 5 MJup. Fig. 4 shows the comparison of the surface density profiles for both models. For the MMSN disk, the gap is more striking in the dust than the gas, while in the CTTS disk the inner disk appears depleted of both gas and dust, while the dust pile-up at the outer gap edge is clearly seen. For the CTTS disk, no change is seen in the surface density for a 0.1 MJup planet in either the gas or dust phase. These results are in general agreement with the minimum planet mass required to produce a gap in the disk models of Paardekooper & Mellema (2006). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Our results have implications for observational pre- dictions of protoplanetary disks hosting planets. Wolf & D’Angelo (2005) and Varnière et al. (2006) use results of 2D hydrodynamic simulations to produce synthetic images of protoplanetary disks, but these simulations assume that the gas and dust are well mixed, which our results clearly demonstrate is not the case. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Fig. 2.— Same as Fig.1 but for the standard CTTS model, in which 104 orbits is equivalent to 26,310 years. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Fig. 3.— Gap created in the MMSN disk (left) and CTTS disk (right) for a range of dust grain sizes after 104 orbits. Top panel shows the top-down view of disk and bottom panel shows the side-on view. Grey is gas and the dust is coloured by density. For the MMSN disk (with a 1 MJup planet at 5.2 AU) from left to right shows 1 m, 10 cm and 1 cm sized grains. For the CTTS disk (with a 5 MJup planet at 40 AU) from left to right shows 1 cm, 1 mm and 100 µm sized grains. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | in MMSN disks should be visible with ALMA. For a more detailed analysis, we refer the reader to Fouchet et al. (2007). |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | outer gap edge is not due to resonant trapping. The accumulation of grains at the external edge of the gap may, however, favour the growth of planetesimals in this high density region. |
Processed_Gap_Formation_in_the_Dust_Layer_of_3D_Protoplaneta.txt | Fig. 4.— Azimuthally averaged surface density profiles for various planet masses after 104 planetary orbits. Top row: MMSN disk, Bottom row: CTTS disk. Left frames: dust profiles, Right frames: gas profiles. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | ABSTRACT This study unveils new insights into the Bullet Cluster through cm-wave observations made with the Australia Telescope Compact Array, focusing on the Sunyaev-Zel’dovich Effect (SZE). A novel SZE feature was discovered, distinctively offset from the X- ray brightness peak, challenging conventional associations between such phenomena. The findings, achieved through high-resolution interferometric imaging, reveal complex substructures within the SZE, including significantly displaced components uncorre- lated with X-ray, optical, or lensing maps. This deviation underscores the complexity within the cluster’s dynamics and suggests that detailed, high-resolution SZE imaging is crucial for understanding the physics of merging clusters. We perform several tests to ensure that one of these novel and unusual features is indeed real, and that the detection of all features in the images is robust. In addition to these displacements, we find several SZE features that are tentatively located at critical boundaries in this merging cluster - these features will be discussed in the next few manuscripts in this series. These observations pave the way for more sophisticated analyses, combining SZE and X-ray data, to decode the evolutionary mysteries of cosmic structures like the Bullet Cluster. We note here that we are discussing only one novel / peculiar feature in the observations of the cluster done in the year 2019. Further observations have revealed more features, which we shall discuss in the next few manuscripts in this series on the Bullet cluster. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The Sunyaev–Zeldovich effect (SZE) is the inverse Comp- ton scattering of CMB photons off a collection of elec- trons which causes a spectral distortion in the CMB (Sunyaev & Zel´dovich 1972). This inverse Compton scatter- ing causes a decrease in CMB brightness at frequencies be- low 217 GHz and an increase above 217 GHz. The amplitude neTedℓ1, of the decrease in brightness is given by ∆T ∝ where the integral is along the line of sight. Galaxy clusters are the largest gravitationally bound objects and the largest sources of hot, dense ionized gas, which makes them ideal targets for studying through the SZE, especially since the SZE is independent of redshift. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | last decade for a variety of reasons; namely, the existence of a strong radio halo (Liang et al. 2000), the Sunyaev-Zel’dovich effect (Halverson et al. (2009), Plagge et al. (2010) and ref- erences therein), its X-ray brightness (Markevitch et al. 2002), though most notably in providing the most direct proof of the existence of dark matter (Clowe et al. 2006). It is a cluster collision/merger event at z ∼0.296, with the larger, westward cluster being ∼10 times the mass of the smaller ‘bullet’. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | X-ray luminosities greater than 5 × 1044erg s−1 have radio halos (Venturi et al 2008). However, among clusters that are dynamically disturbed, about 75% exhibit radio halos (Cas- sano et al 2010). The total power radiated from radio halos (at 1.4 GHz) is found to scale with X-ray luminosity and temperature. In the currently favoured model of radio halo formation, the ‘turbulent re–acceleration through merger’ model, the spectra of radio halos is expected and indeed, ob- served to steepen considerably at frequencies below 10 GHz. Measurements by Liang et al (2000) suggest such a steepen- ing at ∼8.5 GHz. The radio halo in the Bullet cluster was thus not expected to be observable at 18 GHz. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Since not many clusters at redshifts greater than 0.3 were known prior to the PLANCK satellite and MACSur- vey, it has been difficult to probe the existence of a redshift evolution in the occurrence of radio halos, until now. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Two competing models exist as explanations of the ori- gins or radio halos: the so–called Primary or Turbulent Reac- celeration (see, e.g. Brunetti & Lazarian (2011); Cassano (2010); Cassano et al. (2010)) and Secondary models (e.g. Keshet et al. 2010). Primary models of radio halo forma- tion feature an exponential steepening of the spectrum, or a “knee” – the exact frequency at which the “knee” occurs depends on the epoch of the cluster merger, and the age of the radio halo. This “knee” is seen at a wide range of fre- quencies, from 1.4 GHz to 5 GHz. In secondary models, the steepening of the spectrum is followed by a flattening, usu- ally at frequencies higher than 9 GHz. Deep high–frequency observations (above 9 GHz) of radio halos are therefore cru- cial for understanding the origins of radio halos. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In this paper, we present the first 18 GHz interfer- ometer observations of the Bullet cluster from the Aus- tralia Telescope Compact Array (ATCA) using the recently installed Compact Array Broadband Backend or CABB (Wilson et al. including S.S. Malu 2011). We present our observations in §2, describe the imaging of the cluster at 18 GHz in §3, enumerate the consistency checks run by us in §4, place limits on the SZE in §5 and discuss the SZE features in §6. It may be noted here that the interferometer imaging lacks the low spatial frequencies that are required to image the SZE effect in its entirety, and misses the ex- tended SZE emission; therefore our image may not be useful for fitting gas models along with X-ray data. The imaging reveals, for the first time, the existence of a SZE component in the Bullet Cluster displaced from the X–ray peak, point- ing to the importance and need for high angular resolution imaging of SZE in such complex merging clusters in order to usefully model the cluster gas. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | South Pole Telescope (Lueker et al. 2010) show that on arcminute and smaller scales observed CMB anisotropy power is significantly less than levels expected from models of SZE from a cosmological distribution of clusters. This discrepancy may be an indication of our lack of understanding of the environmental impact of cluster mergers. This paper presents sub-arcmin angular resolution observations of the SZE in the cluster, which demonstrates that the relation between SZE and X-ray emission is complex and requires additional physics that may be related to the impact of the merger on intracluster gas. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The Australia Telescope Compact Array (ATCA) is a radio interferometer with six 22-m antennas, five of which may be positioned on stations along a ‘T’-shaped rail track that is 3-km along E-W and 214-m along N-S. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Observations were made in a pair of 2-GHz bands: a ‘17- GHz band’ covering frequencies 16-18 GHz and a ‘19-GHz band’ covering the range 18-20 GHz. Each of the 2-GHz wide bands were subdivided into 2048 frequency channels. All observations were in full polarization mode and recorded multi-channel continuum visibilities. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Observations were made in each of two ATCA array configurations: H168 and H75 that have baselines up to 168 and 75 m respectively; the former to enable subtraction of unresolved continuum sources, the latter to provide en- hanced surface brightness sensitivity for imaging of the SZE. A journal of the observations is in Table 1. In each of the ob- serving sessions, antenna pointing corrections were updated every hour using a 5-point offset pattern observation on PKS B0537−441, the unresolved calibrator PKS B0742−56 was observed every 10 min to monitor and correct for ampli- tude and phase drifts in the interferometer arms, and PKS B1934−638 was observed once every session as a primary calibrator to set the absolute flux density scale. During the 10 mins between successive calibrations, the two pointing positions were sequentially observed for 90 sec each. Visibil- ities were recorded with 10 sec averaging. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | nels at each of the band edges were excluded from analysis to avoid data in frequency domains where signal path gains are relatively low. Frequency channels that appeared to have relatively large fluctuations in visibility amplitude owing to hardware faults in the digital correlator were also rejected prior to calibration and imaging. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Adopted fluxes for the primary calibrator PKS B1934−638 were 1.146 and 0.992 Jy in the 17 and 19 GHz bands respec- tively; the spectral index α (defined as Sν ∝ ν α , where Sν is the flux density at frequency ν) was adopted to be −1.33 in both bands (see Sault (2003)). Outliers in the amplitudes of visibility data on PKS B1934−638 were rejected—removing 15% of data—and the reliable visibilities were used to set the absolute flux density scale as well as determine the instru- ment bandpass calibration. When calibrated for the band- pass, the visibility amplitudes of PKS B0742−56 showed continuity across the 4-GHz observing frequency range and a trend consistent with a single power-law: this was a check ◦ of the bandpass calibration. Drifts of up to 30 were observed in the interferometer arms over the observing sessions: cali- brations for the time-varying complex gains in the antenna signal paths as well as calibrations for polarization leak- ages were derived from the visibilities on PKS B0742−56. RMS phase variations in antenna signal paths within the ◦ 1-min calibrator scans was within 10 , indicating that short timescale atmospheric and instrumental phase cycling would not result in amplitude attenuation of more than 3%. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 1. 18 GHz mosaic image of the Bullet cluster. In this representation as well as subsequent images displayed herein the image has been tapered to make the rms noise uniform over the field of view; this results in an attenuation of unresolved sources by factor 0.02 at the edge. Image rms noise is 6.5 µJy beam−1; the beam FWHM is 39.′′8 × 32.′′1 at 87.3◦ P.A. and is shown using a filled ellipse in the lower right corner. Contour levels are at ±(3,4,5,6,7,8,9,10) times the rms noise. Locations of dis- crete sources detected by Liang et al. (2000) and in this image are marked with cross symbols. This image, and the two images that follow have been deconvolved. Errors due to sidelobes are not significant, as calculated and discussed in the text. In this image, and all subsequent images, light regions and solid contours indi- cate −ve features; dark regions and dashed contours +ve features. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Since no circular polarization is expected, Stokes V is expected to be consistent with thermal noise. Therefore, at times and frequency channels where Stokes V visibilities de- viate more than four times rms thermal noise in the cali- brated visibilities acquired towards the Bullet cluster point- ings, data in all Stokes parameters were rejected. Stokes–V based clipping was therefore done, aimed at automated re- jection of self-generated low level interference. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Multi-channel continuum visibilities in both bands and both pointings were imaged together to form a wide-field multi- frequency synthesis mosaic image of the Bullet cluster shown in Fig. 1. The image displayed has been truncated where the attenuation in the mosaic image owing to the pri- mary beam—and consequently the response to source flux density—falls below 10%. The mosaic image was made with a beam of FWHM 20.′′ position angle (P.A.), −1 at the centre and has an rms noise that is 5.5 µJy beam and increases by factor five to the edges of the field of view. The rms noise in the image was estimated from Stokes V data and is consistent with expectations based on the sys- tem temperature during the observations, bandwidth and integration time. The image displayed has been tapered to make the rms noise uniform over the field of view. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 2. 18 GHz mosaic image of the bullet cluster with unresolved continuum sources subtracted. The beam FWHM and rms noise are same as for figure 1. Contour levels are at ±(3,4,5,6,7,8,9,10) times the rms noise. Locations of discrete sources detected by Liang et al. (2000) and in our 18 GHz data are marked with cross symbols. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | RA: 06h58m37.s9, DEC: −55◦57′25′′ and is marked ‘A’ in the figure; this is a LIRG high-z galaxy (Rex et al. 2009) and is detected with a flux density of 1.69 ± 0.04 mJy at 18 GHz. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Radio images of the cluster field include discrete ra- dio sources, a radio halo (Liang et al. 2000), and SZE (Halverson et al. 2009; Plagge et al. 2010) that is expected to appear as a decrement at the ATCA observing frequen- cies. For the characterization of the SZE decrement and the Radio Halo, it is essential to accurately subtract all discrete sources. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The image in Fig. 1 was made with visibility weighting close to natural in order to have low rms noise and good sur- face brightness sensitivity. To characterize first the unre- solved continuum sources in the field, we constructed a sep- arate image with uniform weighting of the visibilities that has a factor of two greater rms noise but, importantly, de- emphasizes extended halo emission and SZE. Sky regions that have image intensity exceeding four times the rms noise in Fig. 1 were examined for unresolved sources. Apart from sources ‘A’ through ‘H’, three sources were detected with flux density exceeding four times the rms noise in the uni- formly weighted image; these are marked ‘1’ through ‘3’ in Fig. 1. Of these, sources ‘1’ and ‘2’ are undetected in the 1.344 GHz image and are likely to have relatively flatter spectra. The flux densities of the 11 unresolved continuum sources in the field were estimated from the image made with uniform weighting and subtracted from the visibility data. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | A mosaic image of the cluster made with natural weight- ing and using visibilities from which unresolved continuum sources were subtracted is in Fig. 2. The beam FWHM and rms noise distribution and taper are same as for Fig. 1. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Positions of all the sources subtracted have been marked in Fig. 1 and in Table 3 we list the positions and 18-GHz flux densities. A negative feature appears close to the loca- tion where source ‘F’ was subtracted; this is an extended negative feature that is absent in the higher resolution im- ages made with uniform visibility weighting. The brightest positive feature in this residual image is extended emission with a peak of 40 µJy beam−1 at RA: 06h58m31.s1, DEC: ◦ −55 ; this is 18-GHz emission from the radio halo, which was previously discovered by Liang et al. (2000). At RA: 06h58m31.s7, DEC: −55◦57′04′′ is the peak negative fea- ture; this is the centre of the SZ feature, discussed below. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | of the cluster. These SZE features appear to be confined within, and are bounded by the lowest intensity contours of the Radio Halo component, as can be seen in the 18- GHz image (Fig. 3) prior to any subtraction of the halo component, which would only enhance these SZE features. In this image, which has the unresolved continuum sources subtracted, there is a peak in flux density at the peak of the Radio Halo and positive 18-GHz emission towards the eastern parts of the Radio Halo. Most noteworthy is the negative peak, which may be a deep SZE feature, to the south of the peak in the expected halo contribution, at RA: ′′ ′ ◦ 06h58m30.s285, DEC: −55 . It may be noted that this 08 57 interferometer image lacks the low spatial frequencies to re- produce the extended SZE emission, which may be missing. The 18 GHz image made from ATCA (Fig. 3) exhibits −1 at not just the SZE (peak decrement −84.7µJy beam −10.7σ), but also, surprisingly, the radio halo (peak 7.8σ). Radio halos are known to have spectra that steepen at fre- quencies less than 5 GHz. The spectral index of the radio halo in the Bullet cluster does not show any sign of steep- ening at frequencies lower than 8 GHz (Liang et al. 2000), and the apparent steepening at 8.8 GHz may well be due to the SZE. To the best of our knowledge, this is the first positive detection of a radio halo at any frequency above 10 GHz; further discussion of the radio halo at 18 GHz and its implications is left to a companion manuscript. Therefore, in conclusion, the SZE shown in Fig. 3 is the least that can be expected at 18 GHz, with a peak SZE dip of −10.7σ. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 5. An image of the radio halo in the Bullet cluster. This image was obtained by convolving the 1.344 GHz radio halo image with the 18 GHz synthesized beam and then applying a gain taper for the primary beam at 18 GHz. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | notes– Letters indicate discrete sources listed in Liang et al. (2000) and have the same letters as in their Table 2; numbers in- dicate additional discrete sources from our 18-GHz observations. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In the next section, we describe consistency checks on our data and 18 GHz images, to prove that the SZ feature discussed above is not an artifact of improper deconvolution or incorrect point–source removal. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Sidelobe level at the SZE peak dip and the Radio Halo peak sites due to each unresolved continuum source was es- timated, from the beam pattern corresponding to the image in Fig. 8. These values are stated as fractions in Table 5, in columns 4 and 5.Next, the residuals at the point–source sites are estimated; these are listed in column 6 in Table 5 as fractions of the noise rms. Estimates of the errors caused due to sidelobes of these residuals were then estimated; these are listed in the last two columns in the same table. A worst case error in flux density at the position of the hole is es- timated by summing these error contributions; the value of −1, this error estimate is less than 0.06σ or 0.38µJy beam for both the SZE peak dip and the RH peak. The contri- bution to uncertainty from point-source subtraction is thus not significant. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Our 18 GHz data was collected in two different arrays of the ATCA: H168 and H75, in 2009 and 2010. Both the 2009 and 2010 datasets contain data from both the arrays. We can therefore test the validity of the features in the 18 GHz im- age by comparing images from the 2009 and 2010 datasets. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 6. The difference between images made with the 2009 and 2010 datasets. Noise RMS is 20µJy beam−1, and this differ- ence image is consistent with gaussian noise. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | A difference of the images made from the two datasets was , to taken – these images were made with a resolution of 16 ensure that the pixels are independent data samples. The difference is shown in Fig.(6). This difference image was −1, us- compared with a gaussian with RMS 20µJy beam ing the non–parametric Mann–Whitney Wilcoxon test, and it was found that z–value=0.7 and 2–tailed p–value=0.43. The difference image is therefore consistent with gaussian noise. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The shape and size of the SZ effect in our 18 GHz may be compared to the 90 GHz APEX single–dish image, provided that a reliable model of the radio halo at 18 GHz exists. It is not possible to separate the radio halo and the SZ given observations at a single frequency, but a few reasonable as- sumptions may be made about the radio halo, in order to estimate the SZ at 18 GHz. These are described in §5.1 and §5.2. The resulting 18 GHz SZ images were convolved to a 90 beam, as shown in Figs.(9&11). These are qualita- tively similar to Fig.(3) in (Halverson et al. 2009). This is discussed in further detail in §5.1. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In this section we enumerate and examine different possibil- ities for the properties of the radio halo and the SZE and provide upper and lower limits on the peak of the SZE from our observations, combined with those made by Liang et al. (2000). Our ATCA image misses the largest scales (i.e. low- est uv–spacings), and all discussion below pertains to the angular scales covered by our image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Table 4. A comparison of the compact SZ feature properties before and after point–source subtraction. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | notes– The shape of the SZ feature remains unchanged; positional differences between 3σ contours of the SZ feature before and after point–source subtraction are <1′′. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | notes– In this table, as later, we denote the peak SZE dip position by ‘SZM’ and the peak of the radio halo by ‘RHM’. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | We first made an image of the Radio Halo from Liang et al. (2000), extrapolated to 18 GHz, using the spec- tral index quoted by them. In order to ensure the same uv–coverage, we we convolved the Liang et al. (2000) im- age with the synthesized beam at 18 GHz. We then weighed this image with a taper representing the primary beam of the ATCA at 18 GHz. This results in a radio halo image that has the same uv–coverage as our 18 GHz image, and is weighed with a taper representing the beam at 18 GHz. This image can now be used to put limits on the SZE at 18 GHz, as described below. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In what follows, we denote the peak SZE dip position by ‘SZM’ and the peak of the radio halo by ‘RHM’. We enu- merate and discuss the possibilities for radio halo properties below, assuming that the radio halo has a spectral index of -1.3 between 1.344 and 6 GHz, as measured by Liang et al. (2000). Any spectral index variations mentioned in the fol- lowing discussion are between 6 and 18 GHz. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 7. 18 GHz mosaic image of the SZE in the Bullet clus- ter, with the radio halo subtracted assuming α = −1.3 between 1.344 and 18 GHz, as measured by (Liang et al. 2000); the im- age has been smoothed to a beam of FWHM 30′′. Contours are at (−2,−4,−6,−8,−10,−12,−14,−16) times the image rms noise of 8.0 µJy beam−1. Brightness centres of X–ray emission are marked. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 8. 18 GHz mosaic image of the SZE in the Bul- let cluster, with the radio halo subtracted assuming α = −1.3, as measured by (Liang et al. 2000); the image has been smoothed to a beam of FWHM 45′′. Contours are at (−28,−24,−20,−16,−14,−12,−8,−4,−2,2) times the image rms noise of 8.0 µJy beam−1. Brightness centres of X–ray emission are marked, as in the previous image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 9. 18 GHz mosaic image of the SZE in the Bul- let cluster, with the radio halo subtracted assuming α = −1.3, as measured by (Liang et al. 2000); the image has been smoothed to a beam of FWHM 90′′. Contours are at (2,−2,−4,−8,−12,−16,−20,−24,−28,−32,−36,−40) times the im- age rms noise of 14.0 µJy beam−1. Brightness centres of X–ray emission are marked, as in the previous image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | is therefore an upper limit to the SZE in the Bullet cluster at 18 GHz, given the ATCA uv–coverage. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 10. 18 GHz mosaic image of the SZE in the Bullet cluster, with the radio halo subtracted assuming α = −1.3 be- tween 1.344 and 6 GHz and α = −2.0 between 6 and 18 GHz; the image has been smoothed to a beam of FWHM 45′′. Con- tours are at (2,−2,−4,−8,−12,−15) times the image rms noise of 8.0 µJy beam−1. Brightness centres of X–ray emission are marked, as in the previous image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 11. 18 GHz mosaic image of the SZE in the Bullet cluster, with the radio halo subtracted assuming α = −1.3 be- tween 1.344 and 6 GHz and α = −2.0 between 6 and 18 GHz; the image has been smoothed to a beam of FWHM 90′′. Con- tours are at (−2,−4,−8,−11,−15,−20) times the image rms noise of 14.0 µJy beam−1. Brightness centres of X–ray emission are marked, as in the previous image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | SZ is probably a smooth envelope that encompasses multi- ple compact features that are detected in the ATCA image and are spread over the SZE extent. In this scenario, our image detects substructure and this is distributed over the SZE extent suggesting that turbulence in gas pressure owing to the merger is widespread. However, it is well known that Fourier synthesis imaging of extended emission with high angular resolution tends to represent smooth structure as a patchy image arising from calibration errors and deconvo- lution instabilities. We present in Fig. 4 a non–deconvolved source–subtracted image of the Bullet cluster, which is sim- ilar to the deconvolved image in Fig. 3. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Additionally, we have compared two independent datasets in §... above and confirmed that the substructure is identical within the expected errors associated with image noise. We have also shown in above that the contribution from point–source residuals to the error in the peaks of the radio halo and SZE is insignificant; being a small fraction of the noise rms (<0.06σ). |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 12. A model of the radio halo at 18 GHz, given the spec- tral index model which de–emphasizes the compact SZE feature, shown as a contour. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Though Liang et al. (2000) rule out spatial variations in the radio halo spectral index between 1.344 and 9 GHz, halo formation models do not rule out spatial variations in the spectral index above 10 GHz. In general, a flat index at RHM and a steep spectrum at SZM would minimize the SZE. We therefore simulated such a spectral index which would de– emphasize the SZE, in order to explore the effect of extreme variations of the spectral index on the nature of the SZE, as follows. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 13. The result of subtracting the extreme halo in Fig.(12) above, with contours of the radio halo model at 18 GHz (shown in the previous figure). The subtraction clearly leads to several compact features, most of which are caused by the compact fea- tures in the radio halo model. However, the compact SZ feature seen in our original 18 GHz feature is still present. As discussed in the text, this implies that a simple spatial variation that de– emphasizes the radio halo at the SZM is not sufficient to make the compact SZE feature go away. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 14. A slice connecting SZM and RHM, and including the entire extent of the SZE feature as well as the radio halo. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | given the simulated spectral index, and was subtracted from our 18 GHz image, resulting in the image shown in Fig.(13). The resulting SZE image in Fig.(13) clearly has many compact SZE features. This implies that the compact SZE feature seen in Fig.(2) is not merely an artifact of simple (smooth) spatial spectral index variations. The figure clearly suggests that a different approach is needed in order to de– emphasize the compact SZE feature – the radio halo must have a different variation in order for the compact feature to be absent. We explore the simplest such variation below. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Since a simple de–emphasizing of the SZE in the previous section did not yield an SZE image with no compact fea- tures, we now attempt to find the radio halo distribution at 18 GHz, which, when subtracted from our 18 GHz image, would yield a uniform SZE over the entire region of the clus- ter. We do this analysis first along a line connecting the SZM and the RHM, and then for the entire image, as described below. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | A slice connecting the SZM and the RHM was taken, as shown in Fig.(14). A plot of intensity values along the slice is shown in Fig.(15). If the SZE was a large, diffuse feature, it would be a constant over this slice. The least value of this constant would be the same as the lowest value in the plot in Fig.(15), i.e. −90µJy beam−1. We therefore assume that the diffuse SZE will have this value – this will yield the minimum spatial variations in the spectral index needed to have a non–compact SZE. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 15. A plot of intensity along the slice shown in Fig. 14. This plot starts from the bottom right corner of the slice in Fig. 14. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 16. The same slice as in the previous figure, on the radio halo image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 17. A plot of intensity along the slice shown in Fig. 16. This plot starts from the bottom right corner of the slice in Fig. 16. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 19. Variation of the spectral index (between 6 and 18 GHz) as a function of distance along slice which is needed to make the SZ feature in the 18 GHz image non–compact. This spectral index is derived from the minimum radio halo level needed to make the SZ feature non–compact, shown in Fig. 18. The area that has 2 > α > 1 is marked in the figure. Notice that only ∼25% of the plot has a spectral index 2 > α > 1. The region around SZM needs to have a very steep spectrum in order for the SZE feature to be non–compact. This kind of extreme variation in the spectral index (4 > α > 0.1) has never been observed in any radio halo, because radio halos are produced by synchrotron. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 18. A plot of the minimum Radio Halo intensity needed to make the SZ in the 18 GHz image non–compact. This is ob- tained by subtracting the peak dip of the SZ feature from the radio halo intensity along the slice defined in the above figures. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 20. An image of the spatial variation of the spectral index (between 6 and 18 GHz) as a function of position. The area that has 2 > α > 1 is marked between the two green contours in the figure. Notice that only ∼25% of the plot has a spectral index 2 > α > 1. The region around SZM needs to have a very steep spectrum in order for the SZE feature to be non–compact. This kind of extreme variation in the spectral index (4 > α > 0.1) has never been observed in any radio halo, because radio halos are produced by synchrotron. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 21. Our 18 GHz image, with an overlay of contours of the spectral index needed between 6 and 18 GHz in order to make the SZ feature non–compact. Regions with different ranges are marked in the figure as follows. Region 1: 46 α >2; Region 2: 2> α >1; Region 3: 1> α >0; Region 4: α >0. Notice that region 2 has <25% of the area in the index contour. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 22. 1.344 GHz image from (Liang et al. 2000), with an overlay of contours of the spectral index needed between 6 and 18 GHz in order to make the SZ feature non–compact. Regions with different ranges are marked in the figure as follows. Region 1: 46 α >2; Region 2: 2> α >1; Region 3: 1> α >0; Region 4: α >0. Notice that region 2 has <25% of the area in the index contour. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | spectral index map required for the SZ feature to be non– compact. This spectral index map is shown in Fig.(21&22). It is more useful to overplot the contours of this spatial variation of the spectral index on our 18 GHz image as well as the radio halo; these overplots are shown in Figs.(21&22) respectively. From Fig.(21), it is clear that the region with the deepest SZE coincides with the region that has the steep- est spectral index, 4.0> α >2.0. This range for the spectral index is far beyond the limits obtained from observations of radio halos thus far. The reason that spectral indices this steep have never been observed (and are not likely to be ob- served) is that radio halos are caused by synchrotron emis- sion. A steepening in radio halo emission from the region with a deep SZE implies a steepening in the Energy spec- trum of electrons in that region. This is unlikely, given that SZE∝ neTedℓ. Therefore, a spectral index steepening in a deep SZE region is physically inconsistent. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Additionally, spectral indices of α <1 have not been ob- served, whereas we find large regions with this spectral index in Fig.(21). A relatively flat spectral index of 1> α >0.7 may be expected in the hottest and most turbulent regions in the ICM of clusters that have undergone recent mergers. How- ever, flattening is not likely to occur a significant distance away from the most turbulent and hottest regions of the ICM, which is what we observe in the spectral index map in Fig.(20). |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | We may therefore conclude that in trying to rule out a compact SZE feature, we have admitted physically inconsis- tent ranges of spectral indices in a large (>75%) region of the ICM in the Bullet cluster. This attempt to rule out a compact SZE feature is also, therefore, a failure. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In order to obtain further constraints on the SZE and the radio halo, the two components need to be separated through multi–frequency observations; a comparison of 18 GHz and 150 GHz images can lead to SZE–halo separation; however, uv–coverages of the ATCA and APEX instruments is complementary (since APEX is a single–dish experiment). Single–dish measurements of the Bullet cluster at 18 GHz are therefore needed to separate radio halo and SZE. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The sky distribution in the SZE does indeed coincide with the radio halo distribution in the western and southen parts; additionally, in our high-angular resolution interferometer image, SZE is also detected towards the western end of the radio halo in two distinct features in the NW and SW parts of the image. The significance of the peak dips in these features is 4.0σ and 5.5σ (times the noise RMS of 8.0µJy −1) respectively, in Fig.(2) – without radio halo sub- beam traction. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Figure 23. X–ray greyscale overlaid with contours of SZE ob- tained after subtracting the radio halo assuming a spectral in- dex of α =1.3 from 1.344 to 18 GHz. Contour levels are at (−2,−4,−6,−8,−10,−12,−14,−16) times the noise RMS of 6.5µJy beam−1. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | scales of 2.′6 and larger, which is resolved out and missing in the ATCA inteferometer image. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Clearly, the data suggest the existence of a range of scales in the SZE structure in this merging cluster, requir- ing imaging with sensitivity to a range of spatial scales to produce a holistic image of the SZE. In Fig. 23 we show contours of the SZE overlaid on Chandra X-ray intensity distribution shown using grey scales. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Massardi et al. (2010) have recently found a similar dis- placement of SZE dip from X–ray brightness peaks in the merging cluster CL0152−1357. In addition, this effect has been known to exist in the merging cluster RXJ1347−1145 (Komatsu et al. 2001; Mason et al. 2010). |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Future ALMA observations with a range of spatial frequen- cies would be expected to provide SZE images that re- produce the compact and extended structure together and hence be amenable to modeling of the gas in such merg- ing clusters, where the pressure and temperature may have complex distribution structures. However, it may be noted that the synthesized of even the most compact sub–array in ALMA is too small, even at the lowest frequencies at which ALMA operates, to characterize the structures in the Bullet cluster (e.g. ALMA field–of–view at 100 GHz, the lowest op- erating frequency, is only 60′′; the smallest baselines of ∼18 ′′ m yield an angular scale of 35 ). The total–power mode of ALMA is the only way that the large–scale structure of the SZE in the Bullet cluster can be characterized. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | In order to enable a proper difference between the large– scale SZE at 150 GHz from APEX and the SZE at 18 GHz, leading to a characterization of the radio halo, a single–dish observation of the Bullet cluster at 18 GHz is needed. This measurement is possible using the 64 m Parkes Telescope in Australia. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | The Australia Telescope Compact Array is part of the Aus- tralia Telescope which is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Greyscale in Figure 23 courtesy of the Chandra X–ray Observatory Center, which is operated by the Smithsonian Astrophysical Observatory on behalf of NASA. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Travel to Australia was made possible by two travel grants from the Raman Research Institute, Bangalore, India and one from IUCAA, Pune, India. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | We gratefully acknowledge a part of the observations being carried out by Dr. R. Subrahmanyan and Dr. M. Wieringa. We also thank Dr. D. Narasimha for motivating the search for SZ polarization. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | SSM gratefully acknowledges the support provided by CSIR (Council for Scientific and Industrial Re- search), Government of India, through their grant number 03(1462)19 EMR−II. |
Processed_Bullet_Cluster:_Remarkable_Revelations_and_Endurin.txt | Lueker, M. et al., 2010, ApJ, 719, 2 Sunyaev, R. A., Zeldovich, Ya. B., 1972, Comm. Astrophys. |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.