MMTF-Halpha and HST-FUV Imaging of the Filamentary Complex in Abell 1795

0.38 MB·English

by  Michael McDonald

ApJL,8 Sep 2009,in press
PreprinttypesetusingLATEXstyleemulateapjv.03/07/07
MMTF-Hα AND HST-FUV IMAGING OF THE FILAMENTARY COMPLEX IN ABELL 1795
Michael McDonald1, Sylvain Veilleux1,2
Draft version January 27, 2010
ABSTRACT
Wehaveobtaineddeep,highspatialresolutionimagesofthecentralregionofAbell1795atHαand
[N II] λ6583 with the Maryland Magellan Tunable Filter (MMTF), and in the far-ultraviolet (FUV)
0
withtheAdvancedCameraforSurveysSolarBlindChannelontheHubbleSpaceTelescope(HST).The
1
superbimagequalityofthe MMTFdatahasmade itpossible toresolvethe knownSE filamentintoa
0
pairofthin,intertwinedfilamentsextendingfor∼50kpc,withawidth<1kpc. Thepresenceofthese
2
thin, tangled strands is suggestive of a cooling wake where runaway cooling is taking place, perhaps

n aided by an enhanced magnetic field in this region. The HST data further resolve these strands into
a chains of FUV-bright stellar clusters, indicating that these filaments are indeed sites of on-going star
J formation,butatarate∼2ordersofmagnitudesmallerthanthemass-depositionratespredictedfrom

7 the X-ray data. The elevated [N II]/Hα ratio and large spatial variations of the FUV/Hα flux ratio
2
across the filaments indicate that O-star photoionization is not solely responsible for the ionization.

The data favor collisionalheating by cosmic rays either produced in-situ by magnetohydrodynamical
]
O processes or conducted in from the surrounding intracluster medium.
Subject headings: galaxies: cooling flows – galaxies: clusters: individual (Abell 1795) – galaxies:
C
elliptical and lenticular, cD – galaxies: active – ISM: jets and outflows
.
h
p
- 1. INTRODUCTION oney & Bland-Hawthorn 2001; Jaffe et al. 2005) and X-
ro Theabsenceofmassive(∼100−1000M⊙yr−1)cooling rdaoyub(Flea-bsiidanede,traald.i2o0j0e1t;(C4Cra2w6f.o4r2d;evtanalB.2r0eu05g)e,laetpaolw.1er9f8u4l,,
t flowsinthecoresofX-rayluminousgalaxyclustersisof-
s Ge & Owen, 1993) emanating from the central AGN,
tenusedasprimeevidencethatfeedbackplaysanimpor-
a
and a very disturbed, star-forming, central region (Mc-
[ tant role in regulating star formation and galaxy forma-
tionindense environments(see, e.g.,reviewby Veilleux, Namaraetal.1996;Smithetal.1997;Mittazetal.2001;
3 Cecil,&Bland-Hawthorn2005). Energiesofafew×1049 Crawfordet al. 2005).
v As part of a survey of cooling flow clusters, we have
ergspersolarmassofstarsformedareneededtoexplain
4 carried out deep, high-resolution (delivered image qual-
the sharpcutoffatthe brightend ofthe galaxyluminos-
5 ity, DIQ ∼ 0.7′′) imaging of Abell 1795 at Hα and
ity function. Starburst-driven winds are too feeble by a
5 [NII]λ6583usingtheMaryland-MagellanTunableFilter
factor of several to fully account for the cutoff, so AGN
1
(MMTF) on the Magellan-Baade 6.5-mtelescope and in
. feedback is invoked. The ubiquity of large “cavities” in
9 theX-raysurfacebrightnessofclusterswithradiogalax- thefar-ultraviolet(FUV)usingtheACSsolarblindchan-
0 nel(SBC)cameraonthe Hubble Space Telescope (HST).
ies confirms that AGN outflows modify the thermody-
9 These data far surpass any previously available images
namics of the intracluster medium (ICM; see review by
0 of the filaments in this cluster in both depth and spatial
Peterson & Fabian 2006). The relativistic gas injected
: resolution;thisLetterdescribestheresultsfromouranal-
v into the ICM by the AGN has enough energy to quench
i themassaccretionofcoolingflows,buttheexactmecha- ysis of these data. The results from the survey will be
X presented in future papers. The acquisition and reduc-
nismbywhichtheenergyintheradiobubblesturnsinto
r heat is still debated. tionofthedataonAbell1795arediscussedinSection2,
a followed by a description of the results (Section 3) and
The presence of warm hydrogen in the form of line-
a discussion of the implications (Section 4). Through-
emitting filaments extending from the brightest cluster
out this Letter we assume a distance to Abell 1795 of
galaxy(hereafterBCG)hasbeenobservedinmanycool-
260 Mpc, based on a cosmology with H = 73 km s−1
ing flow clusters to date (e.g. Hu et al. 1985, Heckman 0
et al. 1989,Crawfordet al.1999;Jaffe et al.2005;John- Mpc−1, Ωmatter=0.27 and Ωvacuum=0.73.
stoneetal.2005). However,whileseveralintriguingideas
have been put forward, the origin of this gas and the
mechanism for heating it are, as yet, unknown. Of all 2. OBSERVATIONSANDDATAREDUCTION
the galaxy clusters with known optical filaments, there 2.1. Hα – Maryland-Magellan Tunable Filter
is perhaps none quite so spectacular as Abell 1795. This
MMTF has a very narrow bandpass (∼5–12˚A)
cluster has been very extensively studied at a variety of
which can be tuned to any wavelength over ∼5000-
wavelengths, leading to the discovery of a single, long
(∼50 kpc) filament seen in Hα (Cowie et al. 1983; Mal- 9200˚A(Veilleux et al.2009). Coupled withthe exquisite
imagequalityofMagellan,thisinstrumentisidealforde-
1Department of Astronomy, University of Maryland, College tectingemission-linefilamentsindistantclusters. During
Park,MD20742 April2008,weobservedAbell1795foratotalof60min-
fac2hA1l3s1o2:,MDa-x8-5P7l4a1ncGka-Irnchstinitgu,tGfu¨errmexatnryaterrestrischePhysik,Post- utes each at λHα = 6972.8 ˚A, λ[NII] = 6994.7 ˚A and
λcontinuum =7044˚A.Thesedatawerereducedusingthe
2
Fig. 1.— MMTF Hα and [N II] λ6583 and HST/SBC FUV images of Abell 1795. The three upper panels show maps of the surface
brightness,Σ,inunitsofergss−1cm−2arcsec−2. Thelowerleftandcenterpanelsare[NII]/HαandFUV/Hαratiomaps,respectively. In
thelatterimage,theredshaderepresentsregionswherethereisverylittleornoHαcoincidentwiththeFUVemission. Thebottomright
panel outlines the terminology we use in this paper, specifically in Table 1, for the different regions. The “central region” consists of all
emissionnorthof(andincluding)theSWextension. Inallpanels,the10kpcscalecorrespondsto7.9′′,andtheblackcrosshairrepresents
thecentroidofoptical emission,atα=207.2188◦,δ=26.5929◦.
3
MMTF data reduction pipeline3. The continuum image
wasthenPSFandintensitymatchedtothenarrow-band
images to allow for careful continuum subtraction.
2.2. Far UV – Hubble Space Telescope ACS/SBC
FUVimagingwasacquiredusingtheACSSBConthe
HSTintheF140LPbandpass,withatotalexposuretime
of1197seconds. With the MMTF data alreadyinhand,
we were able to choose two different pointings of the
35′′×35′′ field of view to allow full coverage of the SE
filament. Exposures with multiple filters are required to
properlyremovethe knownSBC redleak,whichmay be
substantial in the central regiondue to the fact that the
underlying galaxy is very luminous and red. However,
we were unable to obtain complementary exposures in
a redder SBC band for the filaments, due to scheduling
constraints,and thus we proceedwithout removalof the
offendinglight. ThecentralUVfluxeshaveanassociated
errorof ∼ 2.5% due to the underlying red galaxy(based
on a preliminary analysis of additional FUV data of the
centralgalaxyobtainedbyW.Sparksandcollaborators).
3. RESULTS
Figures1–3showthenewlyacquiredMMTFandHST
Fig. 2.—ColorcompositeimageofAbell1795. Thethreecolors
data on Abell 1795. The most striking result is that representtheredcontinuum(yellow),Hα(red)andfarUV(blue).
the “SE filament”, which has long been known (Cowie ThelongSEfilamentisresolvedintotwointertwinedfilaments in
etal.1983)is,infact,apairofthin,intertwinedfilaments HαandFUV.
inHα. Thesefilamentsare∼42′′ and35′′ (52.9and44.1
kpc) in length, and their widths in Hα are unresolved inconsistentwithphotoionizationbyhotstars. [NII]/Hα
(<0′.′7∼1kpc). ThediscoveryofthinstrandsintheSE is even higher in the nucleus, approaching unity, likely
filamentisreminiscentofmagneticfieldlines. Astronger due to heating by the AGN, and drops northeast of the
than average magnetic field in the ICM could prevent nucleus, where the jet has apparently cleared out a cav-
the filament material from evaporating due to thermal ity. Given the MMTF bandwidth (∼10˚A), we note that
conduction. The BCG in Abell 1795 is a bright double- the [N II] image will also contain any Hα with relative
jetradio-loudcDgalaxy(4C26.42)withthereforestrong velocity>785kms−1,andviceversa. However,sincethe
extended magnetic field (Ge & Owen 1993), but radio typicalvelocity widths are∼300kms−1 in the filaments
emissionisnotdetectedbeyond±6kpcfromthenucleus. (e.g. Crawford et al. 2005), we do not expect interline
We return to this issue in Section 4. contamination to be important.
In the central region of Abell 1795, the Hα emission In the FUV (Figure 1, upper right panel, and Figure
formsaring,withacavityslightlynortheastoftheBCG 3), both strands are visible and, with the added resolu-
center(Figure1). Themorphologyisverysimilartothat tion of HST, are resolved in some regions into chains of
seen in the X-ray (Fabian et al. 2001). The cavity sur- bright, compact sources – the sites of recent star forma-
rounds one of the radio jets emanating from the AGN, tion (see also Crawford et al. 2005). The two brightest
suggesting that it was created by recent AGN activity blobs seen in the filaments at Hα (blobs #1 and #2 in
(Crawford et al. 2005). The very sharp extension di- the nomenclature ofFigure1)break upinto FUV point-
rectlytothesouthwestofthecentralregioniscoincident source and diffuse emission, but spatial offsets of ∼1–2
with the counter-jet. The jet appears to be efficiently kpc are visible between the Hα and FUV emission cen-
heating the gas in this region (van Breugel et al. 1984). troids. This shift is particularlyobviousinthe FUV/Hα
Ingeneral,theHαemissionspatiallycorrelateswellwith map in the lower middle panel of Figure 1. We do not
the X-ray emission in both the central regions and on believe this relative offset is due to errors in astrometry,
largerscalestowithinthe spatialresolutionofthe X-ray since it was also noted in the earlier data of Crawford
data. et al. (2005).
The upper middle panel of Figure 1 reveals the distri- Most of the bright Hα features in the central region
bution of [N II] in Abell 1795. All of the features that emit strong FUV, such as the nucleus, the ring and the
are seen in Hα are also visible in [N II], including both SWextension. However,therearefeaturesinthecentral
filaments, the short SW jet, the nucleus and the cav- region that are not common between the FUV and Hα
ity northeast of the nucleus. The ratio of [N II]/Hα, a data. IntheFUVmap,therearetwonorthernextensions
measure of the relative importance of heating and ion- to the east and west (the NE and NW “extensions” in
ization, is high and fairly uniform throughout the fila- Figure 1),as well asa curvedextensionsouthwestof the
ments (∼0.35–0.55,see Table 1 and the lower left panel centralregion(the “arc”). The emissioninthese regions
of Figure 1); this elevated ratio confirms earlier results is lumpy but largely unresolved into points sources. In
(Crawford et al. 2005 and references therein) and seems Hα, there is a very bright and thick extension at the
base of the SE filament which is not nearly as bright
3 http://www.astro.umd.edu/∼veilleux/mmtf/datared.html in the FUV. This region and other anomalous features
4
TABLE 1
Luminosities, fluxratios andinferredstar formationrates inthe centralregion ofAbell1795.
Region LHα L[NII] LUV [NII]/Hα UV/Hα SFR(Haα) SFR(UaV)
[107 L⊙] [107 L⊙] [108 L⊙] [M⊙/yr] [M⊙/yr]
Total 5.49 1.97 8.07 0.36 14.7 <1.67 <2.66
SEfilaments 0.72 0.26 1.47 0.36 20.3 0.22 0.48
Blob1 0.25 0.09 0.70 0.38 28.4 0.08 0.23
Blob2 0.16 0.09 0.16 0.53 10.0 0.05 0.05
Centralregion 4.77 1.72 6.61 0.36 13.8 <1.45 <2.18
Nucleus 0.12 0.10 0.06 0.81 4.4 <0.04 <0.02
NEextension 0.05 0.00 0.36 0.00 73.0 0.02 0.12
NWextension 0.05 0.00 0.16 0.00 33.3 0.02 0.05
SEextension 0.50 0.13 0.44 0.26 8.9 0.15 0.15
SWextension 1.00 0.48 1.85 0.48 18.5 0.30 0.61
Arc 0.03 0.00 0.28 0.00 102.8 0.01 0.09
(a)StarformationratesderivedusingtheprescriptionsinKennicutt
(1998)assumingalloftheHαandUVemission,includingthatfrom
the nucleus, is due to star formation; the value in the nucleus is
thereforeanupperlimit.
are easily identified in the FUV/Hα map. (More details plungingthroughtheICMwouldresultincooling,rather
will be given in an upcoming paper where images of the than heating, of the multiphase medium. Moreover we
central region of Abell 1795 in all three FUV bands of see no obvious connection in the data between the radio
ACS/SBC are considered. This multiband analysis is jets of 4C 6.42 and the long filaments.
beyond the scope of the present Letter.) By process of elimination, these arguments seem to
favor scenario (a). The very thin geometry of the Hα
4. DISCUSSION filaments points to highly non-linear runaway cooling in
this region. Potentially contributing heating/ionization
4.1. Origin and Power Source of Filaments
sourcesin the filaments include (i) the centralAGN, (ii)
A number of very specific scenarios were put forward hot young stellar population outside the cD galaxy, (iii)
by Fabian et al. (2001) to explain the origin of the SE X-rays from the ICM itself, (iv) heat conduction from
filament, taking into account that 4C 26.42 is in motion the ICM to the colder filaments, (v) shocks and turbu-
withinthe gravitationalpotentialofthe Abell 1795clus- lent mixing layers,and(vi) collisionalheating by cosmic
ter (Oegerle & Hill 2001; see also discussion in Craw- rays. Arguments based on the AGN energetics and the
ford et al. 2005 and Rodriguez-Martinez et al. 2006): line ratios in and out of the central region suggest that
(a) a cooling wake, produced by a cooling flow occur- theAGNisnotstronglyinfluencingthe ionizationofthe
ring around the moving cD galaxy (as in NGC 5044; material much beyond the central region (R . 6 kpc).
David et al. 1994), (b) a “condensation trail” produced The long, thin geometry of the filaments and their qui-
by the ram pressure of the radio source passing through escent velocity field (Crawford et al. 2005) seem to rule
the multiphase medium of the ICM and ISM of the host outionizationbyshocksand/orturbulentmixing layers,
galaxy,(c)evaporationofcoldgasram-pressurestripped while the relative weakness of high-ionization lines sug-
fromthe cD galaxy,and(d) anaccretionwakeproduced gests that ionization by the X-ray ICM is also a small
by the gravitational focussing effects of the moving cD contributor. This leaves scenarios (ii), (iv), and (vi) as
galaxy on the ICM (e.g. Sakelliou et al. 1996). themostplausibleexplanationsforthedominantheating
Scenario(d) requiresthat the accretingICM be gravi- mechanism in the filaments.
tationallyfocusedbythecDgalaxyandcoolintoawake. The presence of FUV point sources in the filaments
However, the relatively large sound speed of the hot gas (Figure 3) lends support to scenario (ii), but is there
makes it unlikely that this process alone can accountfor enough star formation to account for the observed Hα
the presence of the long, thin, X-ray filament. In sce- emission? To try to answer this question, we compare
nario (c), the material making up the filaments is ISM the FUV and Hα emission. Figure 4 shows the aver-
stripped from the cD galaxy. The large mass of hot gas age Hα and FUV surface brightnesses of the brighter
in the filaments, ∼5×109 M⊙ (Fabian et al.2001),and features defined in Figure 1. The Hα and FUV lumi-
thingeometryofthefilamentsarehardtoexplaininthis nositiesofthese featuresare listedinTable 1. Assuming
scenario. However, this does not exclude the possibility that the FUV and Hα star formation rate prescriptions
that some of the gas outside the cD galaxy is produced ofKennicutt(1998)derivedfromtheglobalpropertiesof
in this way (e.g. the broad base of the Hα filaments). star-forminggalaxiesalsoapplyindividuallytothesefea-
Scenario (b) also presents some problems. As argued by tures (we return to this assumptionbelow), we find that
Fabianetal.(2001),itseemsunlikelythatthecDgalaxy
5
all ∼2 orders of magnitude smaller than the predicted
Chandra- and XMM-derived integrated mass deposition
rates from the inner ICM (Ettori et al. 2002; Peterson
et al. 2003).
TheintegratedHαandFUVquantitiesdiscussedsofar
do not tell the full story: Large spatial variations of the
FUV/Hα ratio are seen on smaller scale (see lower mid-
dlepanelofFigure1andpixel-to-pixelsurfacebrightness
measurementsofFigure4). Thesespatialvariationsmay
be due to a number of effects, including differential dust
extinction,variationsinstarformationhistory(SFH:age
ofburst,decaytimescale)orinitialmassfunction(IMF),
complexgeometryofthegasrelativetotheionizingstars,
andcontributionsfromnon-stellarprocessestothe FUV
and Hα emission. Extinction corrections would boost
the FUV fluxes relative to Hα and therefore could ac-
count for regions with anomalously small FUV/Hα ra-
tios. Published data on the filaments suggest relatively
modest extinctions, however (e.g. Crawford et al. 2005).
VariationsintheSFHorIMFchangetherelativeimpor-
tanceofionizingandnon-ionizingstarsandmayaccount
for variations in both directions of the FUV/Hα flux ra-
tio. Our data do not providestrongconstraints onthese
parameters. Geometricaleffects are undoubtedly impor-
tant in some regions, particularly in blobs #1 and #2,
wherespatialoffsets of∼1–2kpc arevisible betweenthe
HαandFUVemissioncentroids. Intheseblobs,theKen-
nicutt (1998) prescriptions may severely underestimate
the number of FUV-bright stars needed to account for
the Hα emission; processes other than photoionization
by hot stars appear needed to account for the observed
Hαattheselocations(the recombinationtimescaleisat
least an order of magnitude shorter than the dynamical
time scale to move ∼1–2 kpc, unless the density of the
Hα filaments is much less than ∼1 cm−3).
As mentioned in Section 3, additional heating sources
also appear needed to explain the unusually strong low-
ionization lines detected in the blobs and in between
them(e.g. Huetal.1985;Crawfordetal.2005;Figure1).
Two heating processes remain viable: (iv) heat conduc-
tion from the ICM to the colder filaments and (vi) colli-
sionalheatingbycosmicrays. Therelativeimportanceof
these processes critically depends on the strength of the
Fig. 3.— Close-up view of the criss-crossing filaments in the magnetic field in the filaments. Strong magnetic fields
FUV showing both of the bright blobs. The brightest portions of couldshieldthecoolingICMgasfromre-heatingbycon-
thefilaments breakupintoyoungUV-brightsuperstarclusters. duction with the hot ICM (scenario (iv)), creating long
tubes of cool, dense gas. Runaway star formation would
theHαandFUVsurfacebrightnessesandluminositiesof takeplacealongthesemagneticfieldlines,wherethegas
thesefeaturesimplystarformationratesurfacedensities is cooling and condensing to high enough density to be-
and star formation rates which are generally consistent come Jeans unstable. The long, thin geometry of the
to within a factor of ∼2 of each other. (The only excep- HαSEfilamentsanddetectionofembeddedFUV-bright
tions are the arc and the NE and NW extensions, which stellarclustersinthesefilamentsareconsistentwiththis
are underluminous in Hα; this could be due to a lack of picture.
gasintheseregionsorotherformsofcontinuumemission Ferlandet al.(2008,2009)haverecently examinedthe
process. A detailed analysis of the central region using questionoftheimportanceofheatingbyenergeticparti-
all three FUV bands of ACS/SBC (PI: W. Sparks) is in cles or dissipative magnetohydrodynamic (MHD) waves
preparation.) Under this assumption, the total Hα and in the centralnebulae of massive clusters. The energetic
UV-determined star formation rates are .1.7 – 2.7 M⊙ particles contributing to the heating of the filaments in
yr−1,with∼0.2–0.5M⊙ yr−1 (∼15–25%)containedin this scenario may either be produced in-situ by MHD
the filaments (Table 1). These numbers are lower than processes or conducted in from the surrounding intra-
previous measurements of the total, integrated star for- clustermedium. Givenourpreviousdiscussion,wespec-
mationrateof∼5-20M⊙yr−1(Smithetal.1997;Mittaz ulate that the optical emission in the SE filaments of
etal.2001)andthe estimateof. 1M⊙ yr−1 ofstarfor- Abell 1795 may naturally be explained by these heating
mationinthe filament(Crawfordetal.2005). Theseare processes. The magnetic field in the filaments may rep-
6
streams seen feeding galaxies in recent high-resolution
numericalsimulationsoftheearlyuniverse(e.g.Ceverino
et al. 2009).
5. CONCLUDINGREMARKS
Using deep, high-resolution Hα, [N II] λ6584, and
FUV data, we have discovered that the SE filament in
Abell 1795 is in fact two intertwined filaments of ion-
ized hydrogen. The most plausible origin for these fila-
ments is a wake of cooling ICM behind the moving cD
galaxyinAbell1795. Thenarrownessofthestrandssug-
gesthighlynon-linearrunawaycoolingoftheICM.Their
tangled morphology suggests that the infalling gas may
be interacting with enhanced magnetic fields, allowing
for less efficient energy conduction and faster cooling in
thisregion. WeobserveknotsofUV-brightpointsources
alongthese filaments,indicatingstarformationata rate
of∼0.5M⊙ yr−1 inthefilaments. Thelargespatialvari-
ationsoftheFUV/Hαratioandenhancedlowionization
lines suggest that O-star photoionization is not the sole
source of heating of this gas; collisional heating by en-
ergetic particles is another likely contributor. A deeper
understandingoftheoriginofthefilamentsofAbell1795
will require detailed MHD modeling and observationsto
Fig. 4.— FUV surface brightness, ΣFUV, versus Hα surface constrainthepurportedmagneticfield,bothofwhichare
brightness,ΣHα,inAbell1795. Thedashedlinerepresentstheex- beyondthescopeofthepresentLetter. Itisalsonotclear
pectedrelationbetweentheglobalΣHαandΣFUV forstar-forming
galaxies (Kennicutt 1998). The larger symbols show the average yet whether this scenarioapplies to cooling flow clusters
surface brightness over the brighter features defined in Figure 1. in general. We plan to address this issue in upcoming
The background distribution of points shows the 2×2 pixel-by- papers using a representative set of massive clusters.
pixelsurfacebrightnessmeasurementsforthecentralregion(grey)
andfilaments(lightred).
resent residual magnetic field originally associated with Support for this work was provided to M.M. and S.V.
the radio galaxy but now entrained in the ICM flow. If by NSF through contract AST 0606932 and by NASA
this is the case, the crisscrossing geometry of the SE fil- throughcontractHSTGO-1198001A. S.V.alsoacknowl-
aments may reflect precession of the radio jets or the edges support from a Senior Award from the Alexander
orbital motion of the cD galaxy in the cluster potential. von Humboldt Foundation and thanks the host insti-
Finally, we end with a cautionary note: Our data do tution, MPE Garching, where this paper was written.
not provide quantitative constraints on the strength of We thank D.S.N. Rupke, C.S. Reynolds, E. Ostriker, H.
the purported ICM magnetic field. The possibility of Netzer, and M.C. Miller for useful discussions, and are
runawaystarformationunaidedbymagneticfieldcannot particularly grateful to W. Sparks for letting us exam-
be ruled out. In fact, the filaments seen in Abell 1795 ine F150LP and F165LP images of the central region of
sharesamorphologicalresemblancewiththenarrowcold Abell 1795 obtained under HST/PID #11681.
REFERENCES
Ceverino,D.,Dekel,A.,&Bournaud,F.2009,MNRAS,000,000 Maloney,P.R.,&Bland-Hawthorn,J.2001,ApJ,553,L129
Cowie,L.L.,Hu,E.M.,Jenkins,E.B.,&York,D.G.1983,ApJ, McNamara, B. R., Wise, M., Sarazin, C. L., Jannuzi, B. T., &
272,29 Elston,R.1996,ApJ,466,L9
Crawford,C.S.,Allen,S.W.,Ebeling,H.,Edge,A.C.,&Fabian, Mittaz,J.P.D.,etal.2001,A&A,365,L93
A.C.1999, MNRAS,306,85 Oegerle,W.R.,&Hill,J.M.2001,AJ,122,2858
Crawford, C. S., Sanders, J. S., & Fabian, A. C. 2005, MNRAS, Peterson, J. R., Kahn, S. M., Paerels, F. B. S., Kaastra, J. S.,
361,17 Tamura, T., Bleeker, J. A. M., Ferrigno, C., & Jernigan, J. G.
David,L.P.,Jones, C.,Forman,W,&Daines,S.1994, ApJ,428, 2003,ApJ,590,207
544 Peterson,J.R.,&Fabian,A.C.2006,Phys.Rep.,427,1
Ettori, S., Fabian, A. C., Allen, S. W., & Johnstone, R. M.2002, Rodr´ıguez-Mart´ınez, M., Vela´zquez, P. F., Binette, L., & Raga,
MNRAS,331,635 A.C.2006,A&A,448,15
Fabian,A.C.,Sanders,J.S.,Ettori,S.,Taylor,G.B.,Allen,S.W., Sakelliou, I., Merrifield,M. R., & McHardy, I. M. 1996, MNRAS,
Crawford,C.S.,Iwasawa,K.,&Johnstone,R.M.2001,MNRAS, 283,673
321,L33 Smith, E. P., Bohlin, R. C., Bothun, G. D., O’Connell, R. W.,
Ferland,G.J.,etal.2009,MNRAS,386,L72 Roberts,M.S.,Neff,S.G.,Smith,A.M.,&Stecher,T.P.1997,
Ferland,G.J.,etal.2008,MNRAS,392,1475 ApJ,478,516
Ge,J.P.,&Owen,F.N.1993,AJ,105,778 vanBreugel,W.,Heckman,T.,&Miley,G.1984,ApJ,276,79
Heckman,T.M.,Baum,S.A.,vanBreugel,W.J.M.,&McCarthy, Veilleux, S., Cecil, G., & Bland-Hawthorn, J. 2005, ARA&A, 43,
P.1989,ApJ,338,48 769
Hu,E.M.,Cowie,L.L.,&Wang, Z.1985,ApJS,59,447 Veilleux,S.,etal.2009,arXiv:0908.1629
Jaffe,W.,Bremer,M.N.,&Baker,K.2005,MNRAS,360,748
Johnstone, R.M.;etal.2005,MNRAS,382,1246
Kennicutt, R.C.,Jr.1998,ARA&A,36,189