Chandra X-ray Observatory Newsletter 2009

8/4/2019 Chandra X-ray Observatory Newsletter 2009

1/48

C handra n ewsWinter 2009

(see page 3 for article )

B laCk holes through CosmiC time : e xploring the distant x- ray u niverse with extragalaCtiC C handra surveys

r yan h iCkox

Published by the Chandra X-ray Center (CXC) Issue Number 16


2/48

CXC Newsletter 2

B laCk h oles through C os -miC t ime

Ryan C. Hickox 3

p rojeCt s Cientist ' s r eportMartin C. Weisskopf, Ste-

phen O'Dell 8

CxC p rojeCt m anager ' s r eport

Roger Brissenden 9

i nstruments : aCisPaul Plucinsky, CatherineGrant, Joe DePasquale 10

i nstruments : hrCRalph Kraft, C.-Y. Ng 10

i nstruments : hetgDan Dewey 12

CxC C ontaCt p ersonnel15

i nstruments : letgJeremy Drake 16

C handra C aliBrationLarry David 19

C handra C aliBration r eviewVinay Kashyap, Jennifer

Posson-Brown 21

C handra ' s F irst d eCade oF d isCovery

22

s upernova r emnants &p ulsar w ind n eBulae in the C handra e ra

23

B ringing C osmology to the p uBliC

Megan Watzke, PeterEdmonds 24

CxC 2008 s CienCe p ress r eleases

Megan Watzke 25

u seFul C handra w eB a d -dresses

25

C handra s CienCe p uBliCa -tions

Paul Green 26

Ciao 4.1: with a w hole n ew s herpa !

Antonella Fruscione,Douglas Burke, Aneta

Siemiginowska26

n ew F unCtionality For p ro -posers : pr o v is and m ax e xpo

Paul Green 28

t he 6 th C handra /Ciaow orkshop

Antonella Fruscione 29

n ews From the C handra d ata a rChive

Arnold Rots, SherryWinkelman 30

C handra s ourCe C atalogIan Evans 31

tgC atDavid P. Huenemoerder,

Arik Mitschang, JoyNichols, Mike A. Nowak,

Norbert S. Schulz, DanDewey

33

C yCle 10 p eer r eview

Belinda Wilkes

u pComing C handra r elated m eetings

4

e instein p ostdoCtoral F el -lowship p rogram

Nancy R. Evans

e duCation and p uBliC o ut -

reaCh p roposals s eleCted in C yCle 10

Kathy Lestition

C handra in iya2009Megan Watzke

i nternational x- ray o Bser -vatory u pdate

Michael R. Garcia

C handra u sers ' C ommittee m emBership l ist

4

dg t au : e nergetiC j ets From a Budding s olar s ys -tem

reprint

i mportant d ates For C handra

4

T able of C onTenTs


3/48

Winter, 2009

I n the last decade we have seen profound advances inour understanding of the composition and evolution of the Universe. Prominent among these is the discovery thatessentially all galaxies with stellar bulges contain super-massive black holes (SMBH), which are believed to be the

relics of accretion in active galactic nuclei (AGN). Further,the masses of SMBHs are tightly correlated to the proper-ties of their host bulges, and the energy released by AGNmay have a signi cant effect on the star formation historyof galaxies. Thus it is increasingly clear that growth andevolution of black holes and galaxies are linked throughcosmic time.

X-ray surveys are exceptionally powerful tools forstudying the evolution of black holes and their host galax-ies, by detecting large numbers of AGN over a wide rangeof redshifts and cosmic environments from voids to groupsand clusters. With its superb angular resolution, low back-

ground, and sensitivity in the energy range 0.58 Chandra has been at the forefront of recent extragalasurveys. In this article we provide an overview of of the leadingChandra surveys, and describe some recresults on the composition of the cosmic X-ray backgr(CXB), the evolution of black hole accretion, the natuAGN populations, and links between AGN and theirgalaxies and environments. This is an extremely activexciting eld, with many key contributions made by XMM-

Newton , Suzaku , INTEGRAL , Swift , and other space anground-based observatories; here we will focus on j

few representative results fromChandra surveys.Breadth and depth in Chandra surveys

Chandra extragalactic surveys range from very dand narrow to shallow and very wide, allowing us to the broadest possible range in redshift and luminosityure 1). The deepest existing X-ray surveys are theChan-dra Deep Fields (CDFs) North (Alexander et al. 2003South (Luo et al. 2008), each with 2 Ms total expoOwing toChandra s unparalleled spatial resolution, thobservations are not limited by confusion and prob

depths more than 6 times fainter than is accessible witother X-ray observatory. The CDFs have yielded extranary progress in understanding faint X-ray populationresolving the CXB. However, the X-ray luminosity sity is dominated by more luminous X-ray sources thrare in the CDFs, soChandra also has undertaken severshallower surveys over wider areas to study these obThese surveys include, in order of increasing area, ELN (Manners et al. 2003), the Extended CDF-S (Leet al. 2005), AEGIS-X (Laird et al. 2009), CLASXSCLANS (Trouille et al. 2008; CLANS is comprisedata from the SWIRE/Chandra survey; Wilkes et al. 2

C-COSMOS (Elvis et al. 2009), XDEEP2 (Murray 2008) and XBotes (Murray et al. 2005). The areas anresponding ux limits for these surveys are shown in Fig -ure 2. Most surveys also have extensive multiwavelcoverage with HST , Spitzer , ground-based optical imagiand spectroscopy, radio, and other observations (Figuallowing us to understand the detailed spectral energytributions (SEDs) and environments of the X-ray souFor a more detailed review of X-ray surveys with a on the deepest elds, see Brandt & Hasinger (2005).

In addition to dedicatedChandra observations in contiguous survey elds, the Chandra Multiwavelength Proje

FIGURE 1 (full size image on cover): X-ray images andsurvey areas for a few representativeChandra surveys:XBotes (Murray et al. 2005), C-COSMOS (Elvis et al.2009), AEGIS-X (Nandra et al. 2005), and theChandra Deep Field-North (Alexander et al. 2003). The relativeareas of each eld are superposed on the XBotes image,and survey exposure times are shown. (The elds coverseparate regions on the sky, and are shown together onlyfor comparison.) The full C-COSMOS image is shownon an expanded scale. The inset illustration shows anactive galactic nucleus; the vast majority of the X-raysources detected inChandra surveys are AGN.(Figure

prepared by the author, AGN illustration credit: NASA/ JPL-Caltech/T.Pyle-SSC)

B laCk holes through CosmiC time : e xploring the distant x- ray

u niverse with extragalaCtiC C handra surveys

r yan C. h iCkox


4/48

CXC Newsletter 4

(ChaMP) uses archival data for optical imaging and spectro-scopic followup of serendipitous sources from 392 (non-con-tiguous) ACIS pointings (Green et al. 2004, Kim et al. 2007).The large samples allow removal of potentially biasing PI-tar-geted objects and provide large statistical subsamples immuneto cosmic variance. Similar follow-up of serendipitous hardX-ray (210 keV) sources has been undertaken by the Ser-endipitous Extragalactic X-ray Source Identi cation (SEXSI)program (Harrison et al. 2003).

Resolving the cosmic X-ray background

Since its discovery in rocket ights at the dawn of X-rayastronomy (Giacconi et al. 1962), the origin of the diffuse ex-

tragalactic X-ray background has been one of the leaquestions in high-energy astrophysics. With the extional sensitivity of the CDFs, it is now clear that alall of the extragalactic CXB at energies 5 keV, indicating a missing plation of hard sources, which may include AGN thahighly obscured by intervening gas (see below). Hi& Markevitch (2006) took a complementary approach,measuring the absolute ux of the unresolved CXB in theCDFs, and showed that the resolved fraction of thekeV CXB is 80%.

Hickox & Markevitch (2007b) demonstrated that only7% 3% of the 12 keV CXB remained unresolved af -ter excluding HST sources in the GOODS eld, in broadagreement with a stacking analysis by Worsley e(2006). By studying the distribution of X-ray counthe HST source positions, Hickox & Markevitch (2007a)showed that the log N -logS for faint, unresolved X-ray gaxies in the CDFs is consistent with an extension oobserved population of faint star-forming galaxies, rthan AGN. These results indicate a large populatiofaint X-ray sources that may be accessible with deobservations in the CDFs.

The cosmic evolution of black hole growth

A key question in black hole evolution is: wherewhen did black holes gain their mass? Unlike galaxiewhich we can determine ages for the stars), black hhave no memory of their formation history. Thereto determine the cosmic evolution of black hole grwe must observe that growth directly, by measuringthe space density of accreting black holes evolves cosmic time. A number of authors have used the wof spectroscopic redshifts available in X-ray survederive the X-ray luminosity function for AGN at a rof redshifts from the local Universe to z > 4 (e.g., Ueda eal. 2003; Barger et al. 2005; Hasinger et al. 2005; Siman et al. 2008a). To cover the largest possible regithe luminosity-redshift plane, these studies combinefrom narrow, deep and wide, shallowChandra surveys, aswell as data from XMM-Newton , ASCA , Rosat, and othermissions.

FIGURE2: Chandra surveys span a wide range of depthsand areas, in order to probe the widest possible ranges inredshift, luminosity, and environment. The gure showslimiting 0.52 keV ux versus area for various Chandra blank- eld and serendipitous extragalactic surveys (see textfor references). Since sensitivity varies acrossChandra

elds, for a given survey the area increases with increas -ing ux limit. Lines show the sensitivity curves between25% and 75% of the total area of each survey. Solid linesshow contiguous surveys, dotted lines show serendipitoussurveys, and dashed lines show surveys comprised of twoor three separate elds with similar depths and multiwave -length coverage. Flux limits are de ned somewhat dif -ferently for different surveys, but generally correspond toa ~50% completeness limit. For SEXSI (which is a hardX-ray selected serendipitous survey) the 210 keV limited

uxes were converted to 0.52 keV by dividing by 6.5,corresponding roughly to the relative on-axis ux limitsof the CDF-N in the soft and hard bands. The AEGIS-Xand XBotes sensitivities correspond to the 200 ks and 5 kssurveys, respectively, while the sensitivity from the 10 ksXDEEP2 exposures are estimated from XBotes.

0.01 0.10 1.00 10.00Area (deg 2)

1017

1016

1015

1014

L i m i t i n g

0 . 5 2 k e V f l u x

( e r g s c m

2

s

1 )

Chandra extragalactic surveys

CCOSMOS

XBotes

ChaMP

ECDFSAEGISX

CDFS

CDFN

SEXSI

CLANS +CLASXSELAISN

XDEEP2


5/48

Winter, 2009

These studies have shown that AGN activity has a rela-tively complex and interesting evolution with redshift.Chandra results favor a model of luminosity-dependent density evolution (LDDE) , in which the number densityof AGN evolves differently for sources of varying lumi-nosities. These results provide evidence fordownsizing , inwhich the density of the most luminous AGN peaks earlierin cosmic time than for less luminous objects (e.g., Steffenet al. 2003; Hasinger et al. 2005, see Figure 4), which canbe shown to imply that large black holes are formed earlierthan their low-mass counterparts (e.g., Merloni & Heinz2008; Shankar et al. 2009). Qualitatively similar downsiz-ing has been observed for star formation in galaxies (e.g.,Cowie et al. 1996), providing a circumstantial link betweenSMBH and galaxy evolution.

Understanding X-ray source populations

The large numbers of AGN detected in extragalactic sur-veys allows for robust statistical studies of AGN popula-tions. Particular effort has been focused on measurementsof X-ray spectra, which provide insights into the nature of AGN accretion. Tozzi et al. (2006) performed spectral tsfor hundreds of sources in the CDF-S, while Green et al.(2009) measured the spectra for >1000 SDSS quasars inChaMP, including 56 with z > 3. These studies show that ingeneral, the unabsorbed spectra of AGN have remarkablyuniform power-law continua. However,Chandra studiesalso have shown that X-ray spectra and X-ray to UV SEDsof AGN get harder with decreasing Eddington ratios1 (e.g.,Steffen et al. 2006; Kelly et al. 2008), similarly to blackhole X-ray binaries (Remillard & McClintock 2006).

X-ray surveys also can constrain the numbers of AGNthat are absorbed by intervening gas, which preferentiallyabsorbs low-energy X-rays and so hardens the observedspectrum. The total spectrum of the CXB is harder than theemission from a typical unabsorbed AGN, indicating a sig-ni cant contribution from absorbed sources. In the localUniverse, 75% or more of optically selected Seyfert gal -axies are obscured by dust (Maiolino & Rieke 1995), andChandra surveys suggest a similar fraction are absorbed inX-rays. Further, there is evidence that the obscured frac-tion rises at lower luminosities and may increase at higherredshifts (e.g., Ueda et al. 2003; Steffen et al. 2003; Has-inger 2008), which may constrain models in which AGNprovide radiation pressure feedback on surrounding gas. Inaddition, X-ray studies have con rmed the identi cationof obscured AGN detected in optical and IR observations(e.g., Hickox et al. 2007; Polletta et al. 2008; Donley et al.2008; Alexander et al. 2008b).

Using the observed luminosity functions, spectralshapes, and absorbing columns derived fromChandra andother surveys, it has been possible to model the spectrumof the total cosmic X-ray background (e.g., Treister & Urry

2005; Gilli et al. 2007). While these models still havni cant uncertainties (particularly in the number of high -ly obscured AGN), their success implies that we maconverging on a coherent picture for the cosmic evolof black hole growth. One caveat however, is that thenumbers of X-ray counts in surveys make it dif cult to dis -tinguish absorption from intrinsically hard spectra forsources. If the X-ray spectra of AGN become harder aEddington ratio, this could produce a large number ofluminosity, X-ray hard AGN that would be classi ed asabsorbed in current analyses (Hopkins et al. 2009)ture deep surveys or detailed stacking of X-ray spectrabe able to break the degeneracy between intrinsic speshape and absorbing column.

Links between AGN and galaxy evolution

Finally,Chandra surveys have allowed for detailed amination of the host galaxies and environments of XAGN, providing insight on the role of AGN in the evolof galaxies. One powerful diagnostic is the color-lumity distribution for the galaxies that host AGN. Galare known to be divided into two types in color-magnspace: the red sequence of luminous, passively evogalaxies, and the blue cloud of less luminous, star-forsystems. Interestingly, a number of Chandra studies havefound that at z 1, luminous X-ray AGN (unlike radio,tical, or infrared-selected AGN) are preferentially fou

FIGURE 3: Optical counterparts of X-ray sources inCDF-N as detected in the Hubble Deep Field (HDF)Shown is the full HDF image, and circles show objmatched with X-ray sources in the 2 Ms CDF-N.(FromChandra press release; Credit: NASA/Penn State .)


6/48

CXC Newsletter 6

the green valley, with colors intermediate between blueand red galaxies (e.g., Nandra et al. 2007; Georgakakis etal. 2008; Silverman et al. 2008c; Hickox et al. 2009, seeFigure 5), indicating that X-ray AGN may be associatedwith the transition of galaxies from the blue cloud to thered sequence, and may be responsible for quenching thestar formation in galaxies through feedback (Bundy et al.2008; Georgakakis et al. 2008). Further, studies of the en-vironments and clustering of AGN nd that Chandra X-rayAGN are preferentially found in overdense regions charac-teristic of galaxy groups (e.g., Yang et al. 2006; Georgaka-kis et al. 2007; Silverman et al. 2008b; Hickox et al. 2009;Coil et al. 2009). Models suggest it is these environmentswhere star formation shuts off in massive galaxies; AGNmay play a role in the initial quenching of star formation,or in subsequent heating that prevents gas from cooling andfurther forming stars (e.g., Croton et al. 2006; Bower et al.2006; Hopkins et al. 2008).

Chandra surveys also have explored the presence of AGN associated with massive, vigorously star-forminggalaxies in the distant Universe ( z ~ 2). X-ray studies of starburst galaxies selected in the CDFs with observationsin the submillimeter (e.g., Alexander et al. 2005) and in-frared (Daddi et al. 2007; Fiore et al. 2008) have providedevidence for a large density of highly obscured AGN ingalaxies co-eval with the formation of the bulk of theirstellar mass. Alexander et al. (2008a) found that the AGNin submm galaxies have black hole masses that are rough-ly consistent with those expected from the local relationbetween black hole mass and bulge mass, indicating thatthere may be continuous feedback between star formation

and accretion in these systems. While the precise nof the AGN population associated with luminous starbis not yet clear, these studies point further towards imtant links between the growth of galaxies and their ceSMBHs.

The future

Future X-ray surveys will provide more sensitive ovations and larger AGN samples to study the characterand evolution of SMBH accretion in greater detail. future prospect withChandra is even deeper observatioin the CDFs.Chandra s high angular resolution will low it to observe signi cantly deeper (up to 8 Ms or more)without reaching the confusion limit in the central reg(Alexander et al. 2003). This would allow the detectia new population of extremely faint star-forming galaas well as providing better photon statistics for the sothat have already been resolved. The upcoming NuST2mission (to launch 2011) will provide an unprecedeall-sky survey at hard X-ray (680 keV) energies, wthe eROSITA3 instrument on the Spectrum X-Gamma servatory (scheduled for launch in 2011-2012) will suthe sky at 0.212 keV energies and will provide enorsamples of X-ray AGN. Among missions proposed ffuture, Simbol-X4 (target launch 2014) would provide hangular resolution (better than 30'') and high sensitivthe ~0.5-80 keV range, while EXIST5 would conduct large-area X-ray survey at the very hard (5-600 keV) gies. The Wide-Field X-ray Telescope6 would provide aanalog to SDSS in the X-ray band, detecting >107 X-ray

FIGURE 4: The comoving space density of X-ray AGN as a function of redshift, shown in the (a) 0.52band (Hasinger et al. 2005) and (b) 210 keV band (Ueda et al. 2003). Lines show the evolution of Aspace density in different bins of luminosity. The data favor LDDE models, in which the space density ofluminosity AGN peaks at a lower redshift compared to high-luminosity objects. Figure compiled by Brandt &Hasinger (2005).


7/48

Winter, 2009

FIGURE 5: (a) Optical colors and absolute magnitudes of AGNs at 0.25


8/48

CXC Newsletter 8

AGN over >10,000 deg2 in the energy range 0.14 keV(possibly extending to 6 keV). Further, the enormous sen-sitivity of the International X-ray Observatory7 would al-low us to study in detail the spectra of large numbers of faint AGN. These future missions will be essential to buildon our understanding of the growth of black holes and theirplace in the larger picture of galaxy and structure formationin the Universe.

Acknowledgements : Thanks to David Alexander, Frances-ca Civano, Bill Forman, Paul Green, Christine Jones, andSteve Murray for input, discussions, and comments on thisarticle.

FOOTNOTES

1The Eddington ratio is de ned as the ratio of bolometricaccretion luminosity to the Eddington limit, whereLEdd = 1.3x 1038( M BH/ M ) ergs s-1.2http://www.nustar.caltech.edu/3

http://www.mpe.mpg.de/projects.html#erosita4http://smsc.cnes.fr/SIMBOLX/5http://hea-www.harvard.edu/EXIST/6http://wfxt.pha.jhu.edu/7http://ixo.gsfc.nasa.gov/

REFERENCES

Alexander, D. M., et al. 2003, AJ, 126, 539Alexander, D. M., Bauer, F. E., Chapman, S. C., Smail, I., Blain, A. W.,Brandt, W. N., & Ivison, R. J. 2005, ApJ, 632, 736Alexander, D. M., et al. 2008a, AJ, 135, 1968Alexander, D. M., et al. 2008b, ApJ, 687, 835Barger, A. J., Cowie, L. L., Mushotzky, R. F., Yang, Y., Wang, W.-H., Stef-

fen, A. T., & Capak, P. 2005, AJ, 129, 578Bower, R. G., Benson, A. J., Malbon, R., Helly, J. C., Frenk, C. S., Baugh, C.M., Cole, S., & Lacey, C. G. 2006, MNRAS, 370, 645Brandt, W. N. & Hasinger, G. 2005, ARA&A, 43, 827Bundy, K., et al. 2008, ApJ, 681, 931Coil, A. L., et al. 2009, submitted to ApJ (arXiv:0902.0363)Cowie, L. L., Songaila, A., Hu, E. M., & Cohen, J. G. 1996, AJ, 112, 839Croton, D. J., et al. 2006, MNRAS, 365, 11Daddi, E., et al. 2007, ApJ, 670, 173Donley, J. L., Rieke, G. H., Prez-Gonzlez, P. G., & Barro, G. 2008, ApJ,687, 111Elvis, M., et al. 2009, submitted to ApJFiore, F., et al. 2008, ApJ, 672, 94Georgakakis, A., et al. 2007, ApJ, 660, L15Georgakakis, A., et al. 2008, MNRAS, 385, 2049Giacconi, R., Gursky, H., Paolini, F. R., & Rossi, B. B. 1962, Physical Re -view Letters, 9, 439Gilli, R., Comastri, A., & Hasinger, G. 2007, A&A, 463, 79Green, P. J., et al. 2009, ApJ, 690, 644Green, P. J., et al. 2004, ApJS, 150, 43Harrison, F. A., Eckart, M. E., Mao, P. H., Helfand, D. J., & Stern, D. 2003,ApJ, 596, 944Hasinger, G. 2008, A&A, 490, 905Hasinger, G., Miyaji, T., & Schmidt, M. 2005, A&A, 441, 417Hickox, R. C., et al. 2007, ApJ, 671, 1365Hickox, R. C., et al. 2009, ApJ in press (arXiv:0901.4121)Hickox, R. C. & Markevitch, M. 2006, ApJ, 645, 95. 2007a, ApJ, 671, 1523. 2007b, ApJ, 661, L117

Hopkins, P. F., Hernquist, L., Cox, T. J., & Kere, D. 2008, ApJS, 175, 356Hopkins, P. F., Hickox, R., Quataert, E., & Hernquist, L. 2009, submitted toMNRAS (arXiv:0901.2936)Kelly, B. C., Bechtold, J., Trump, J. R., Vestergaard, M., & Siemiginowska,A. 2008, ApJS, 176, 355Kim, M. et al 2007, ApJS, 169, 401Laird, E.S. et al 2009, ApJS 180, 102Lehmer, B. D., et al. 2005, ApJS, 161, 21Luo, B., et al. 2008, ApJS, 179, 19Maiolino, R. & Rieke, G. H. 1995, ApJ, 454, 95Manners, J. C., et al. 2003, MNRAS, 343, 293Merloni, A. & Heinz, S. 2008, MNRAS, 388, 1011Moretti, A., Campana, S., Lazzati, D., & Tagliaferri, G. 2003, ApJ, 588, 696Murray, S. S., Forman, W. R., Jones, C., Hickox, R., Kenter, A., & Willner,S. 2008, AAS HEAD Meeting, Los Angeles, CA, 32.03Murray, S. S., et al. 2005, ApJS, 161, 1Nandra, K., et al. 2007, ApJ, 660, L11Polletta, M., Weedman, D., Hnig, S., Lonsdale, C. J., Smith, H. E., &Houck, J. 2008, ApJ, 675, 960Remillard, R. A. & McClintock, J. E. 2006, ARA&A, 44, 49Shankar, F., Weinberg, D. H., & Miralda-Escud, J. 2009, ApJ, 690, 20Silverman, J. D., et al. 2008a, ApJ, 679, 118Silverman, J. D., et al. 2008b, ApJ in press (arXiv:0810.3653)Silverman, J. D., et al. 2008c, ApJ, 675, 1025

Steffen, A. T., Barger, A. J., Cowie, L. L., Mushotzky, R. F., & Yang, Y. 2003,ApJ, 596, L23Steffen, A. T., Strateva, I., Brandt, W. N., Alexander, D. M., KoekA. M., Lehmer, B. D., Schneider, D. P., & Vignali, C. 2006, AJ, 131, 2826Tozzi, P., et al. 2006, A&A, 451, 457Treister, E. & Urry, C. M. 2005, ApJ, 630, 115Trouille, L., Barger, A. J., Cowie, L. L., Yang, Y., & Mushotzky, R. F. 2008,ApJS, 179, 1Ueda, Y., Akiyama, M., Ohta, K., & Miyaji, T. 2003, ApJ, 598, 886Wilkes, B. et al. 2009, submitted to ApJSWorsley, M. A., Fabian, A. C., Bauer, F. E., Alexander, D. M., Brandt,& Lehmer, B. D. 2006, MNRAS, 368, 1735Worsley, M. A., et al. 2005, MNRAS, 357, 1281Yang, Y., Mushotzky, R. F., Barger, A. J., & Cowie, L. L. 2006, ApJ, 645, 68

T his year we celebrateChandra 's 10th anniversary! On 1999 July 23, the space shuttle Columlaunched theChandra X-ray Observatory. Then on 19August 12,Chandra obtained its rst-light image of an x-ray source we dubbed "Leon X-1", in honor of TelesScientist Leon VanSpeybroeck. We fondly remember

and sincerely appreciate his contributions to the succethis mission. We also are grateful to the hundreds of stists, engineers, technicians, support personnel, and (sic) managers who contributed to the developmentoperation of Chandra .

On behalf of theChandra Team, we invite you to jous in Boston, September 22-25, for the Symposium Years of Science withChandra Chandra 's First Decadof Discovery (http://cxc.harvard.edu/symposium_20

Here's to another 10 years!

p rojeCt s Cientist s r eport

m artin C. w eisskopF , s tephen od ell


9/48

Winter, 2009

C handra marked nine years of successful mission op-erations with continued excellent operational andscienti c performance. Telescope time remained in highdemand, with signi cant oversubscription in the Cycle 10peer review held in June. In December 2008 the observingprogram transitioned on schedule from Cycle 9 to Cycle10, and we look forward to the Cycle 11 peer review inJune 2009.

NASA has decided to merge theChandra and GLAST(now Fermi) fellowship programs, together with a num-ber of additional fellowships, into a combined programcalled the Einstein Fellowships, which includes researchareas related to the science goals of the Physics of the Cos-mos program and its missions. NASA chose the CXC tosolicit, evaluate and select proposals and administer theawards. In August the CXC issued a call for proposals forten Einstein fellowships and received 156 proposals. (SeetheEinstein Postdoctoral Fellowship Program article).

The team worked hard to prepare for NASAs SeniorReview of operating missions held in April. For the rsttime, the review included two large missions,Chandra andSpitzer . Chandra ranked second of the ten missions re-viewed, with a score of 9.1 out of 10. The review commit-tees report said, Chandra has recently produced - and isexpected to continue to produce - results which change ourview in areas as diverse as dark matter, dark energy, andclusters of galaxies. As a result of the ranking,Chandra sfunding level is expected to remain at approximately thepresent level for the next two years.

The CXC mission planning staff continued to maximizeobserving ef ciency in spite of temperature constraints onspacecraft pointing. Competing thermal constraints con-tinue to require some longer observations to be split intomultiple short duration segments, to allow the spacecraftto cool at preferred attitudes. Following careful study andthermal modeling, mission planning complexity eased af-ter an EPHIN radiation detector thermal constraint was re-laxed in December 2006 and selected heaters on the ACISinstrument and the science instrument module were turnedoff in 2008 (see theInstruments: ACIS article). Overallthe average observing ef ciency in 2008 was 68%, com -pared with 67% in the prior year and a maximum possibleef ciency of ~70%. Operational highlights over the pastyear included 9 requests to observe targets of opportunitythat required the mission planning and ight teams to re -schedule and interrupt the on-board command loads. Thesun was quiet during the year, causing no observing inter-

ruptions due to solar activity.Chandra passed through th2008 summer and winter eclipse seasons, as well as alunar eclipse in February, with nominal power thermal performance.

The mission continued without a signi cant anomalyand had no safe mode transitions in the last year. Onoccasions the spacecraft transitioned to normal sun due, it is believed, to single event upsets in electroniccuits. In all cases the operations teams returned the scraft to normal status with no adverse consequenceminimal loss of observing time.

Both focal plane instruments, the ACIS (Advanced Imaging Spectrometer) and the HRC (High ResolCamera), have continued to operate well and have hamajor problems. ACIS, along with the overall spacehas continued to warm gradually. Following tests coned during 2007 and 2008 to determine the effect of tuoff the ACIS detector housing heater, the heater was tuoff permanently in 2008, with an immediate bene cial re -duction in the average ACIS focal plane temperaturetheInstruments:ACIS article).

All systems at theChandra Operations Control Centcontinued to perform well in supporting ight operations.

Chandra data processing and distribution to obsers continued smoothly, with the average time fromservation to delivery of data averaging less than 2 TheChandra archive holdings grew by 1.6 TB to 6.6(compressed) and now contain 25 million les. 1.2 TB of the increase representsChandra Source Catalog data proucts.

The Data System team released software updatesupport the submission deadline for Cycle 10 observproposals (March 2008), the Cycle 10 Peer Review (and the Cycle 11 Call for Proposals (December). In tion, the team released production versions of theChandra Source Catalog (CSC) software and began processingfor the catalog in the Fall. Virtually all publicly avaiACIS data representing compact sources have processed, representing on the order of 100,000 souand will be available when the catalog is of cially re -leased in early 2009.

The Education and Public Outreach (EPO) grousued 11Chandra press releases, 17 image releases, anpress postings and advisories during 2008. The EPOorganized two press conferences and two NASA mteleconferences, one accompanied by a separate brie ngfor NASAs Museum Alliance. The team continurelease Chandra podcasts through the CXC website, w22 new standard de nition and 20 high de nition podcastson science topics this year. They also released an podcast of the text of the multi-wavelength Braille Touch the Invisible Sky, and a video plus audio poof the Universe Forums Incredible Two Inch Univwith American Sign Language translation. In addi

CxC p rojeCt m anager s r eport

r oger B rissenden


10/48

CXC Newsletter 10

the Chandra podcasts were converted to the Adobe MediaPlayer (AMP) format for the launch of the AMP channelon NASAs website. The team received a NASA ROSESgrant to implement the International Year of AstronomyCornerstone Project, From Earth to the Universe exhib-its at two semi-permanent and six traveling sites aroundthe United States (see theChandra in IYA2009 article).

We look forward to a new year of continued smoothoperations and exciting science results. Please join us tocelebrateChandra s discoveries at the seminar Ten Yearsof Science withChandra , to be held in Boston in Septem-ber, 2009.

i nstruments : aCis

p aul p luCinsky , C atherine g rant , j oe d e p asquale

FIGURE 6: The average ACIS focal plane tempeture for science observations from April 2007 unJanuary 2009. The dashed vertical line at 2008.27 dicates the day on which the ACIS detector housiheater was turned off. The green dashed horizonline indicates the desired FP temperature of -119.7 The red dashed line indicates the temperature (-119C) at which the gain of the FI CCDs changes by 0.3%and the blue dashed line indicates the temperatu(-118.2 C) at which the gain of the BI CCDs changby 0.3% .

T he ACIS instrument continued to perform well overthe past year with no failures or unexpected degra-dations. The charge-transfer inef ciency (CTI) of the FIand BI CCDs is increasing at the expected rate. The CIAOsoftware and associated calibration les correct for thisslow increase in CTI over time. The contamination layercontinues to accumulate on the ACIS optical-blocking l -ter. Recent measurements indicate that a revision of thetemporal model for the contaminant may be necessary.The CXC calibration group is investigating the issue andmay release a revision later this year.

The only signi cant change in ACIS performance over

the last year regards the thermal control of the ACIS fo-cal plane (FP). As discussed in last year's newsletter ar-ticle, a higher percentage of ACIS observations have beenexperiencing warmer than desired FP temperatures as thespacecraft ages and components inside the spacecraft getwarmer in general. In response to this trend the ACIS Op-erations team turned off the ACIS Detector Housing (DH)heater on April 7, 2008. Before the DH heater was turnedoff, approximately 1/3 of observations had an average FPtemperature of -119.2 C or warmer. After the DH heaterwas turned off, only about 2% of observations had an aver -age FP temperature of -119.2 C or warmer. The desired FP

temperature is -119.7 C. The FI CCDs have a narrowertolerance on the FP temperature than the BI CCDs. For theFI CCDs, the gain changes by 0.3% at 1.5 keV if the FPtemperature warms to -119.2 C and for the BI CCDs, thegain changes by 0.3% if the FP temperature increases to-118.2 C. Figure 6 shows the average FP temperature fromApril 2007 until January 2009. The red line shows the FICCD limit of -119.2 C and the blue line shows the BI CCDlimit of -118.2 C. The gure clearly demonstrates the im -provement in control of the ACIS FP temperature since theDH heater was turned off.

As a consequence of this change, the temperatuthe ACIS Camera Body (CB) is no longer actively trolled. This created a complication for the aspect restruction since the ACIS ducial lights are mounted on theACIS CB. As the CB temperature increases/decreaseCB expands/contracts, producing an apparent motiothe ducial lights. The CXC Aspect team has calibratedthis motion by relating the brightness of the ducial lightsto the CB temperature and has modi ed the aspect recon -struction software to account for this effect. After appthis change, the Aspect team has con rmed that the aspectreconstruction is as accurate after the DH heater was tuoff as it was before.

H RC operations continue smoothly with no mproblems, anomalies, or interruptions. Roumonitoring observations show no signi cant charge ex -traction from the detectors. There may be some evidof a decrease in the low energy (below 400 eV) QE oHRC-S, probably indicative of the chemical evolutio

i nstruments : hrCr alph k raFt , C.-y. n g


11/48

Winter, 2009

the CsI photocathode. This is being monitored by the CXCCal team and the HRC instrument team, but this phenom-enon currently is not signi cant for scienti c observations.There has been no signi cant change in the HRC-I quan -tum ef ciency during the past year. One HRC observationwas made using one of the shutters during the past year, anHRC+LETG observation of the Crab Nebula. The shut-ter was used to block the zeroth order image in order toreduce the overall instrument rate from this bright sourcebelow the telemetry limit. There were some anomalies ininserting and withdrawing the shutter in the past. Overallanother quiet year from an HRC perspective.

A wide variety of scienti c investigations have beencarried out over the past year with the HRC instruments.This year we highlight an HRC-I observation of the Mousenebula, a pulsar wind nebula, demonstrating the HRC's im-aging and timing capabilities.

High Resolution X-ray Imaging of the MouseC.-Y. Ng, for the Mouse Team

16.2 16.0 15.8 15.6 15.4 15.2 17:47:15.0 14.8

57:56.0

58.0

- 29:58:00.0

02.0

04.0

06.0

Right ascension

D e c l i n a t i o n

16.2 16.0 15.8 15.6 15.4 15.2 17:47:15.0 14.8

57 :56. 0

58.0

- 29:58:00.0

02.0

04.0

06.0

Right ascension


FIGURE 7: Raw (top) and smoothed (bottom) HRimages of the Mouse PWN.

N eutron stars lose a signi cant amount of their ro -tational energy through the relativistic winds. Theconsequent interactions with the surrounding materialsresult in broadband synchrotron emission, collectively re-ferred to as pulsar wind nebulae (PWNe). The propertiesof PWNe depend strongly on the evolutionary state andenvironment. Since neutron stars are typically born withspace velocities of a few hundred kilometers per second,they eventually escape the natal supernova remnants andtravel supersonically through the interstellar medium. Thisresults in bow shock nebulae, in which a pulsar's relativis-tic out ow is con ned by the ram pressure. As comparedto PWNe con ned within supernova remnants, bow shocksare governed by a much simpler set of boundary condi-tions, offering ideal cases to re ne our understanding of relativistic shocks and pulsar winds electrodynamics.

The best example of a bow shock PWN is G359.23-0.82

16.2 16.0 15.8 15.6 15.4 15.2 17:47:15.0 14.8

57 :56 .0

58 .0

- 29:58:00.0

02 .0

04 .0

06 .0

Right ascension


FIGURE 8: ACIS-S image of the Mouse PWN.

(the Mouse) powered by the 0.1s period pulsar J12958. This is the brightest bow shock system in X-located near the direction of the Galactic center. M

wavelength studies show that the Mouse is well-moby a bright 'head' coincident with the pulsar, a tongueresponding to the surface of the wind termination shand an elongated tail produced by material in the shock ow. We have recently obtained new HRC observa -tions of this bow shock system, which offers the hig

angular resolution images. Figure 7 showsHRC image (raw and smoothed, respectivedemonstrating the unique capability of the Hinstrument. The new HRC observation suggsome hints of small-scale features, possibly or knots, near the backward termination sh

The results will allow a comparison to detamagnetohydrodynamic simulations. For cparison, an archival ACIS images of the sregion is shown in Figure 8. For timing ansis, we have carried out simultaneous radio ing observations of the central pulsar in ordefold the HRC data. The lightcurve of the pulsshown in Figure 9 folded on the frequency oradio pulsations (100 ms). We found no stacally signi cant X-ray pulsation at the radio pulsefrequency.


12/48

CXC Newsletter 12

FIGURE 9: HRC-I lightcurve of the central pulsar of the Mouse PWN folded on the frequency of the radiopulsations.

HETG Status and Calibration

T he HETG continues to perform nomi-nally with no speci c issues. Of note inthe past year are the efforts of the "InternationalAstronomical Consortium for High Energy Cal-ibration" (http://www.iachec.org/) to assess thecross-calibration of several current X-ray mis-sions. HETG observations contributed to thelow-energy results given in: "The SMC SNR1E0102.2-7219 as a Calibration Standard for

X-ray Astronomy in the 0.3-2.5 keV Bandpass"by Plucinsky et al. 2008. The missions (andinstruments) included in this paper are: XMM-

Newton (RGS1, MOS1, MOS2, pn),Chandra (HETG-MEG, ACIS-S3),Suzaku (XIS0, XIS1)and Swift (XRT). Relative ux measurementswere compared at four low-energy spectral re-gions, corresponding to the bright lines of OVII, O VIII, Ne IX and Ne X (with rough ener-gies of 0.57, 0.65, 0.92, and 1.02 keV.) Thesemid-2008 results demonstrate an absolute cali-bration agreement that is largely within 10%across the instruments. The effort is on-going

with yearly IACHEC meetings, the next is schedulethe end of April 2009.

HETG Technique: "Look Ma! No Angstroms!"

HETG spectra are often presented in counts versus Astrom space and plotted to emphasize narrow spectratures, and so they look very different when comparedX-ray spectra from CCD instruments (e.g., ACIS, EXIS). However, there is nothing about the HETG speproducts that prevents viewing them in the more fam

ux versus keV space. Likewise, as the following exam -ples show, HETG observations can be used to: meacontinuum sources, create light curves, generate hardratios, etc.

The broadband MEG spectrum of the Galactic Xbinary 4U 1957+11 is shown in Figure 10; this binarcludes what may be the most rapidly spinning black in the Galaxy. Here the data are t with a model consist -ing of a disk blackbody plus a Comptonizing corona wcontributes at higher energies. Besides studying blackemission models through the broadband spectra, the also allowed for a very accurate (and very low) determtion of the N_H toward 4U 1957. Of course one canzoom in on the spectrum and study narrow featuresmay be present. For example, here Ne IX absorptioseen at about 0.92 keV. The measured depth of this feis consistent with the idea that most of the hot gas w

i nstruments : hetg

d an d ewey

Be sure to also see the TGCat article in this Newsletter !

FIGURE 10: HETG spectrum of the low-mass X-ray binary 4U 1See text for discussion. (from Nowak et al. 2008)


13/48

Winter, 2009

along sightlines to distant objects (e.g.,the AGN Mkn 421) is fairly local to,and above the disk of, our Galaxy.

In perhaps an even more uncommonpresentation of HETG data, Figure 11from Schulz et al. (2008) shows a lightcurve and color-color diagram from anobservation of the neutron star X-raybinary Cyg X-2. Note that the count rateis equivalent to almost 0.5 Crab duringthe observations. To observe such abright source, even given the reducedeffective area and reduced pileup of theHETG instrument, it was necessary tooperate the ACIS readout in the contin-uous clocking (CC) mode as well. TheHETG spectral data in 500 s time inter-vals were coarsely binned in three spec-tral bands (soft: 0.5-2.5, medium: 2.5-4.5, and hard: 4.5-8 keV) to create thecolor ratios plotted (Hard Ratio = hard/medium, Soft Ratio = medium/soft).With the branches of the "Z-pattern"established, it was possible to measurevariations of the ux of broad emissionlines of Mg, Si, S, and Fe in the differ-ent "branches" of the Z-track, leading toan understanding of the parameters of the accretion disk corona in Cyg X-2.

HETG Science: "3D Clues in NovaOutbursts"

Black holes and neutron stars cap-ture our imagination and interest partlybecause they are so "extreme". But theexciting phenomena surrounding theirlives, accretion disks, jets, outbursts,and explosions, are also seen in more"vanilla" systems. Koerding et al.(2008) recently noted: "there may be acommon disc/jet coupling in all accret-

4 5 7321 6 8

0.32 0.34 0.36 0.38 0.40 0.42

0 . 3 2

H

a r d R a t i o

Time [10 k sec ]

S oft Ratio

0 . 2 4

0 . 2 6

0 . 2 8

0 . 3 0

2 0 0

2 5 0

3 0 0

C o u n t R a t e

FIGURE 11: (Top) Light curve from an HETG observation of Cyg X-2tom) The light-curve points are plotted in a color-color diagram showinhorizontal branch (blue), normal branch (red), and aring branch (green)(fromSchulz et al. 2008).

ing objects from young stellar objects (YSOs) to gammaray bursts (GRBs)." This statement was made in the con-text of one example of such systems, the cataclysmic vari-ables (CVs), binary systems consisting of a donor star andan accreting white dwarf (WD).

As artistically shown in Figure 12 (courtesy A. Beard-more, http://www.astro.keele.ac.uk/~apb/OGL_CV/), in a"dwarf nova" the material from the donor star accumulatesin the cold outer region of an accretion disk around a non-magnetic WD. On occasion a viscous instability will trig-ger the matter to fall into the boundary layer around theWD, creating an optical outbust. In a recent paper,Okada

et al. (2008) present the X-ray view of an outburst oCyg using archival HETG data, Figure 13. Althougoverall X-ray ux decreases during outburst, the uxes of the emission lines increase by factors of 2 to 10.

Of particular interest is the broadening of the linesing the outburst. This broadening, likely due to Dovelocities, gives clues to determine the geometry oemitting material. Okada et al. consider material in awind, a shell of matter falling onto the white dwarf,or azimuthal motion of the plasma around the white dThey lean toward this latter mechanism based on ts to theset of bright K-alpha lines of O, Ne, Mg and Si.


14/48

CXC Newsletter 14

REFERENCES

Drake et al. (2009), ApJ, 691, 418.Koerding et al. (2008), Science, 320, 1318.Nowak et al. (2008), ApJ, 689, 1199.Okada et al. (2008), ApJ, 680, 695.Orlando, S., Drake, J.J., Laming J.M. (2009),A&A 493, 1049.Plucinsky et al. (2008), Proc SPIE 7011 (arX-iv:0807.2176).Schulz et al. (2009), ApJL submitted (arX-iv:0812.2264).

FIGURE 12: An artistic scientists view of the dwnova SS Cyg (courtesy A. Beardmore, http://www.tro.keele.ac.uk/~apb/OGL_CV/).

FIGURE 13: HETG spectra of SS Cyg in quiescence outburst (from Okada et al. 2008)

FIGURE 14: HETG-measured line pro-les of RS Oph at 13.9 days into its Feb -

ruary 2006 outburst; the data (blue) havea detailed structure differing from thesimple spherical shell model (backgroundgray; from Drake et al. 2009).

A similar but more energetic system is RS Oph, a fulledged nova in which matter accretes from a red giant

(RG) onto the surface of the WD where it explodes in athermonuclear runaway, Figure 15 (left) (http://www.swift.ac.uk/RSOph.shtml). Observed with Director Discretion-ary Time in 2006, the HETG-measured line shapes, Figure14 (Drake et al. 2009), encode not only aspects of the sourcegeometry and velocities, but they are sculpted by differ-ential absorption which removes the red-shifted (positivevelocity) side of the pro les. Amazingly, a detailed hydro -dynamic model including X-ray emission synthesis, Figure15 (right), produces asymmetric, blue-shifted line pro lesremarkably similar to those observed by the HETG. Toreproduce a comparable emssion-measure-versus-temper-ature curve as the data, the model includes an additionaldensity enhancement in the equatorial plane of the binarysystem (consistent with non-X-ray observations). This en-hancement combined with the collimation produced by thesteep density variation in the RGs wind environment pro-duces the highly asymmetric remnant.

Clearly, combining high-resolution X-ray observationsover a wide wavelength range (i.e., the whole range of abundant ions: C, N, O, Ne, Mg, Si, S, Fe ) with detailed3D hydrodynamic simulations (and bigger computers)holds the promise of letting us see the inner workings of X-ray sources with high delity in the years ahead.


15/48

Winter, 2009

FIGURE 15: (left) An artists rendering shthe blastwave leaving the WD hours afterexplosion (Credit: STFC/David Hardy). cause of the 1/r2 wind density pro le around thered giant, the blast wave is far from spherand will be effectively collimated away frthe red giant. (above) A synthetic X-ray

age of RSOph two weeks after explosion bon a detailed hydrodynamic model shows the X-ray emission mainly originates fro(relatively) small region in the equatorial p(from Orlando et al. 2009). Note the scalthis image: the WD and RG are the tiny wand red dots at far right!

CXC Contact Personnel

Director: Harvey Tananbaum Calibration: Christine Jones

Associate Director: Claude Canizares Development andOperations:

Dan Schwartz

Manager: Roger Brissenden Mission Planning: Pat Slane

Systems Engineering: Jeff Holmes Science Data Systems:Deputy:

Jonathan McDowellMike Nowak

Data Systems: Pepi Fabbiano Directors Of ce: Belinda Wilkes

Education & Outreach: Kathy Lestition Media Relations: Megan Watzke

Note: E-mail address is usually of the form: < rst-initial-lastname>@cfa.harvard.edu(addresses you may already know for nodes head.cfa.harvard.edu or cfa.harvard.edu should work also)


16/48

CXC Newsletter 16

I t was with a bittersweet tear of nostalgia in the cornerof my eye that I watched my aging Linux box and

its mountain of dusty whirring 9Gb appendages being un-plugged and wheeled away, like a beloved institutionalizedolder relative at the end of visiting time. I didn't really meanit, of course, that time, or the other time, or that time either,and that other one, or ten, when I threatened to hurl it fromthe roof of the Center for Astrophysics during one of itssenior moments of frustrating idiosyncratic X-dysfunction.So it was with some measure of guilt-ridden trepidationthat I started to punch the keys of the new model: two rip-pling quad-cores of seductive, brushed aluminium-encasedMac Pro that seemed to whisper try me... It was for thegood of the project, after all. We must be pragmatic. Senti-mentality must be cast aside.Chandra needs it. It needs itbecause it has a new virtual mirror. Its new HRMA modelmakes it even better than it was before. It helps a lot of people understand their observations more easily. It im-proves the accuracy of our astrophysics. It silently easesunspoken international tensions with its European sibling

XMM-Newton . It is progress. And it damn well breaks thewhole of the LETGS calibration and we have to do it allover again.

I mention the new computer so the reader will have com-plete faith in our wherewithal to meet this slightly involvedtask: the reprocessing and analysis of about two score andten differentChandra observations. Already I have got togrips with some powerful new spectral data analysis soft-ware on the Mac, and after only a few weeks we have madegreat progress. I haven't quite worked out the details of how to do all the spectral tting and things like thatthecool street nomenclature of the Mac is quite different towhat I'm used to and all the X-ray models and things haverather strange and sometimes colorful namesbut I amcon dent I can work it all out soon. Even the softwareitself has an imaginative name: its called Garage Band,or something like thatpresumably a reference to the earlyyears of AS&E.

Why does the LETGS have to be recalibrated? UnlikeChandra 's ACIS detector, the response of the primary de-tector for the LETGthe HRC-Srelies rather heavily onusing cosmic sources of X-rays for calibration. At shorterwavelengths, 50 or so, the supply of photons from thesteady photospheres of white dwarfs that we use as stan-dard candles1 effectively runs out, and we have to turn to,shall we say, less well-understood X-ray sources. Naturehas handed us one possible option in the form of BL Lacobjectsperhaps not a poisoned chalice, but a rather un-

pleasant cocktail you would rather pour down the while no one is looking. (This might explain some obehavior of our colleagues studying BL Lac's.)

BL Lac's beckon our calibration like sirens of themos with their apparently silky smooth pure powercontinua, only to dash it on the rocks of spectral rewhen we fail to notice the bit of curvature here or a bin index there, or that the thing has decided to leancompletely different angle to when we looked at ittime. So in order to use bright BL Lac's for calibrationthe adoptedChandra standard PKS 2155-304, we needknow how they are behaving and what spectral modattribute to them during our observations. As a refewe have used contemporaneous observations obtained

i nstruments : letg

j eremy d rake

FIGURE 16: Sky map (top) and wedge diagram (botof the region of the Sculptor Wall where the blazar H 2309 is located. Each dot correspond to a different galThe blue dashed lines in the upper panel de ne the range of declination represented by the galaxies in the lower pThe blue dashed lines in the lower panel show the red range of the galaxies in the top panel. The WHIM strucorresponding to these galaxies was detected in combChandra LETG and XMM-Newton RGS spectra. (FromBuote et al. 2009).


17/48

Winter, 2009

the HETG+ACIS combination to constrain the models forthe LETG. Since the HRMA model throughput has nowchanged, the models one derives from the HETG observa-tions also change: all the data (HETG and LETG) have tobe reanalyzed. This is of course where Garage Band and allits spectral models comes in. It appears that a combinationof U rban models based onV intage F Unk and e lectric S lap with =120 bpm provide a pretty good match to thedata. Expect the revised LETG+HRC-S calibration releaseduring the spring.

The Writing on the Sculptor Wall

Randall Smith once made the ingenious suggestion thatthe mysterious disappearance of socks experienced hereon earth could account for the missing baryons in the uni-verse.2 While the details of the theory remain to be eshedout, it is perhaps not too far-fetched in the realm of cos-mology to hypothesise that socks are somehow evaporatedfrom tumble driers into the intergalactic medium and formlow density lamentary structures similar to those predict -ed by cosmological hydrodynamic simulations (e.g. Cen& Ostriker 2006 and references therein). If so, the sock-baryon theory might be tested by absorption spectroscopyof the warm-hot intergalactic medium (WHIM).

X-ray WHIM spectroscopy was pioneered by FabrizioNicastro and coworkers based on LETG spectra of brightbackground blazars (yes, that cocktail again! Nicastro etal. 2002, Nicastro et al. 2005). It is tricky because the ab-sorption signatures are rather weak, and either an unusu-ally bright background source or a very long exposure (andpreferably both) is needed to see any convincing absorptionfeatures. In a recent study using both theChandra LETGand RGS, Buote et al. (2009) have circumvented one of themajor problems that greatly affects the sensitivity of theexperiment. In searching for WHIM signatures at arbitraryred shift, ablind search must be done, testing for spectralfeatures at all plausible red shifts from zero up to that of thebackground source. If, instead, the red shift of a particularWHIM feature is known, features can be sought at theirexpected wavelengths, thereby reducing the chance of co-incidence with statistical uctuations through the numberof trials avoided in the corresponding blind search.

Buote et al. (2009) looked for WHIM absorption signa-tures in the spectra of the blazar H 2356-309 located be-hind a large structure in the Sculptor region of the southernsky containing a conglomeration of groups and clusters of

galaxies known as theSculptor Wall . This region and thline of sight to H 2356-309 is shown in Figure 16. LETG and RGS spectra of H 2356-309 are illustratFigure 17, where features attributed to absorption by Oat z=0, likely corresponding to local Galactic gas oLocal Group WHIM, and at z=0.032, corresponding texpected Sculptor Wall red shift, are detected. NeitheChandra or XMM-Newton observations provide signi cantdetections alone, but when combined both lines attasigni cance of ~ 3. The Buote et al. (2009) work providesstrong evidence in support of the theoretical simulathat predict the missing baryons reside in the WHIM

The most common constituent of sockscottocomprises mostly bres of cellulose (C 6H12O6) with traceamounts of heavier elements such as Si, Ca and Al. socks instead are made of more complex proteins fromkeratin family and contain much more nitrogen. Whildetection of O VII lines is perfectly consistent with hionized socks of any type, future deeper studies enpassing lines of C, N and O might also help constraidifferent sock populations of the universe.

No Gain Map without Pain Map

In Newsletter 15, my good colleague Brad Wargbravely contravened the Writers Guild of America sand penned an excellent report of the latest work tode ning a new gain map for use in shovelling out thosebackground events that accumulate like snow in New land over the HRC-S during observations. In 200implemented a background lter developed by Brad that

removes background based on detector pulse heightsentially the strength of the electron cascade resultinga photon or particle event on the illuminated side odetector. Photon events on the HRC-S have a fairly narrowly-peaked pulse height distribution whereas paevents giving rise to the snowy background are much

FIGURE 17: Background-subtracted XMM-Newton RGS(black) and Chandra LETG (red) spectra of blazar H 2356-309 in the 2122.5 range. The solid curves illustrate spec-tral models comprising a power law continuum and absorp-tion lines corresponding to the local WHIM and WHIM atthe red shift of the Sculptor Wall. (From Buote et al. 2009).


18/48

CXC Newsletter 18

uniformly distributed. By drawing a line around the pho-ton distribution and throwing out the events outside it wecan exclude some of the background.

This lter is still quite effective and gets rid of abouthalf the background with minimal X-ray event loss. How-ever, two aspects of the detector performance prevent usdrawing a general line around photon events as tightly aswe would like to exclude the most background possible.Firstly, we had to allow some slop for a secular drop inthe detector gainand the peak of the X-ray pulse heightdistributionthat has continued since launch. Secondly,

the detector pulse height response for a photon of a givenenergy strongly varies over the detector. To avoid theseproblems, Brad and Pete Ratzlaff went to great pains todevised a time-dependent gain map that utilizes the energy-dependent event position signals telemetered by the detec-tor. Unlike simple pulse height, this scaled SUMAMPSparameter shows very little spatial variation over the detec-tor. The result is a background lter that can cut around realphoton events much more tightly and reject more back-ground (Figure 18). The overall ef cacy of the new lteris shown in Figure 19.

As you can gather from Figure 18, the improvemover the current lter is larger toward longer wavelengths.Anyone attempting to analyze spectra longward of ~that are not completely source-dominated (and they rare) is urged to apply the new ltering algorithm in order toreduce the background. Even for 50 these reprocesing steps should prove worthwhile. Since there reallygain without pain, what is the catch? The algorithm hto be implemented in CIAO and its application is sliless direct than the current approach. However, Petezlaff has whipped up a ripping Perl program that doe

job and is easy to apply (though I have not yet workehow to do this in Garage Band). The lter is more fullydescribed with links to the ltering program and gain mapon http://cxc.harvard.edu/contrib/letg/GainFilter/. ACIAO implementation of the improved lter is expectedlater this year.

For the HRC-S Pain Map, see our faces at the upcoCalibration Workshop.

FIGURE 18: Comparison of the current and new lters, applied to data from 2000 and2008. The old PI lter becomes less effective over time because it does not account fordecreasing gain. Less background is removed at short wavelengths because the lteringthreshold must be higher to avoid excluding X-ray events. The raw background rate ishigher in 2008 than 2000 because of the solar cycle, but relatively more background isremoved in 2008 because the BG pulse-height distribution has relatively more high-channel events. (From http://cxc.harvard.edu/contrib/letg/GainFilter/)


19/48

Winter, 2009

FOOTNOTES

1 e.g. Newsletter no. 132 In a presentation to the Constellation-X Facility ScienceTeam, 2006 December

REFERENCES

Buote, D. A., Zappacosta, L., Fang, T., Humphrey, P. J., Gastaldello, F.,& Tagliaferri, G. 2009, arXiv:0901.3802

Cen, R., & Ostriker, J. P. 2006, ApJ, 650, 560

Nicastro, F., et al. 2005, Nature, 433, 495

FIGURE 19: The 2008 LETG+HRC-S backgroundrate with and without ltering, using the standard'bow-tie' spectral extraction region. The right-handaxis units refer to the ~ 0.07 FWHM resolution of the LETGS. (From http://cxc.harvard.edu/contrib/letg/GainFilter/)

Update to the HRMA Effective Area

An updated version of the High Resolution MirrorAssembly (HRMA) effective area was released

to the public in CALDB 4.1.1 on Jan. 21, 2009. This re-lease was based on a re-analysis of ground-based calibra-tion data taken at the X-ray Calibration Facility (XRCF)at the Marshall Space Flight Center (MSFC). The newHRMA effective area has the most signi cant affect onthe spectroscopic analysis of hot clusters of galaxies (kT> 4 keV) and ux estimates for very soft sources. The newHRMA effective area was tested with a variety of astro-nomical sources (e.g., clusters of galaxies, AGNs, thermal

C handra C aliBration

l arry d avid

supernova remnants, synchrotron dominated remnantsofter thermal sources) and the differences in the despectral parameters (e.g., temperatures and power-ladices) between the new and old HRMA effective areatypically less than 3%, except for hot clusters of galaxies.For hot clusters, temperatures determined with theHRMA effective area are up to 10% less compared to thetemperatures derived with the previously released veof the HRMA effective area. While the spectral paramof soft sources do not change signi cantly with the newHRMA effective area, uxes derived with the new HRMAeffective area can be up to 8% higher for very soft sourcesin the 0.5-2.0 keV energy band. This article gives adiscussion of the recently released update to the HReffective area.

Chandra /XMM-Newton Cross-Calibration with Clus-ters of Galaxies

As part of an on-goingChandra / XMM-Newton cross-calibration project with the Astronomical InternatCouncil for High Energy Calibration (IACHEC), a saof 12 X-ray bright clusters were chosen to compare gasperatures obtained from ACIS, MOS and PN data. Wthis project began, revisions to the low energy responthe two MOS detectors were imminent, so the compawas restricted to the 2.0-7.0 keV band pass. The resucomparison showed that temperatures derived from Aand EPIC (MOS and PN) data were in good agreefor cool clusters (kT < 4 keV). For hotter clusters, Atemperatures were up to 10-15% higher than EPIC derivedtemperatures. Since it is impossible to determine such a comparison whether the ACIS or EPIC derivedperatures are closer to the true temperatures, a studyundertaken to check the internal consistency of the Acalibration. This study showed that ACIS temperaobtained by tting the continuum emission in hot clustersfrom 2.0-6.0 keV were systematically higher than peratures obtained from tting a broad energy band from0.5-7.0 keV and also the temperature calculated fromH-like to He-like Fe-k line ratio.

HRMA Effective Area Calibration

Prior to this release, there have been two major adments to the effective area of the HRMA since the comtion of ground-based calibration. The rst was an empiricalcorrection based on the results of extensive ground-btests at the XRCF, which was incorporated in versionof the HRMA effective area. The second was the addto the model of a contamination layer on the optics, bupon analyses of on-orbit grating spectra of blazars wexhibited residuals near the Ir-M edge, around 2 keV.


20/48

CXC Newsletter 20

residuals could be reduced if the mirrors had a hydrocarboncontamination layer of thickness 22 . The addition of thislayer led to release of version 007 of the HRMA effectivearea in CALDB 3.2.1 on Dec. 15, 2005.

Re-Analysis of the XRCF Data

A recent re-analysis of the XRCF data showed that themolecular contamination was already present on the mir-rors during testing at the XRCF. The initial empiricalXRCF correction essentially corrected for the gross ef-fects of the molecular contamination, but had insuf cientdetail to resolve the Ir-M edge discrepancy. The additionof the explicit 22 thick contamination layer provided therequired detail to x the Ir edge problems, but essentiallydoubled the contamination layer, over-correcting the grossfeatures.

During XRCF testing, a system of shutters was placed

behind the HRMA so that the response of the 4 shells couldbe measured independently. Two instruments were usedat the focal plane; a ow proportional counter (FPC) anda solid state detector (SSD). A re-analysis of single-shellSSD measurements of a continuum source, this time in-cluding molecular contamination, led to estimates of thecontamination layer depths on the individual shells (18-28A) which are in good agreement with the result derivedfrom the on-orbit blazar gratings spectra. After incorporat-ing this effect into the model, an additional gray correctionwas required to bring the model's prediction for the XRCFmeasured effective area in line with that measured. This

correction is similar in purpose to the XRCF correctionfactor used in earlier models, but is derived from a differentweighting of the FPC and SSD measurements. There arestatistically signi cant differences between those measure -ments; the current method of weighting is, we believe, animprovement over the previous method. The resolution of the discrepancies between the two detectors is the subjectof an ongoing investigation, which may lead to improve-ments in our understanding and subsequent improvementsto the HRMA effective area.

The new version of the HRMA effective area (CALDB4.1.1) is based on the predictions of the new model, in-cluding the as-measured contamination depths and the graycorrection. Since the depth of the molecular contaminantvaries from shell-to-shell, an updated version of the HETGef ciency was also released in CALDB 4.1.1 to maintaincross-calibration consistency between the HEG (which in-tercepts photons from the two smallest mirror shells) andMEG (which intercepts photons from the two largest mir-ror shells).

FIGURE 20: Comparison of cluster tempera-

tures derived from the analysis of XMM-NewtonMOS andChandra ACIS data using two versionsof the HRMA effective area. All temperatureswere derived by tting the data in the 2.0-7.0 keVenergy band.

FIGURE 21: Comparison of cluster temperaturesderived from the analysis of XMM-Newton MOSandChandra ACIS data using two versions of theHRMA effective area. All temperatures were de-rived by tting the data in the 0.5-7.0 keV energyband.


21/48

Winter, 2009

Comparison of Cluster Temperatures with UpdatedHRMA Effective Area

Figures 20 and 21 show a comparison of ACIS andMOS derived cluster temperatures with the old and newversions of the HRMA effective area. Temperatures weredetermined in both a hard (2.0-7.0 keV) and broad (0.5-7.0keV) energy band. These gures show that there is nowgood agreement between ACIS and MOS derived tempera-tures in the hard energy band for all cluster temperatures.In the broader energy band, ACIS derived temperatures areslightly higher than MOS temperatures. This results fromlower MOS temperatures derived in the broad energy bandcompared to the hard energy band. Our studies have alsoshow that the MOS and PN produce consistent tempera-tures in both energy bands.

While the new HRMA effective area is a signi -cant improvement over the previous version, theChan-dra optics team is still investigating a few unresolvedissues with the XRCF data which may lead to fur-ther improvements in the HRMA effective area.

C handra C aliBration r eview

v inay k ashyap , j enniFer p osson -B rown

T he Chandra Calibration Review (CCR) is held re

larly for the dissemination and advancement ofunderstanding of the performance and capabilities oChandra X-ray Observatory. It is intended to share Chandra teams' knowledge of the detectors, gratings, rors, and aspect system with the community while enaging participation and feedback in the process of caling the observatory. Abstracts are solicited on calibrrelated issues.

This year, we will again be holding the Calibrationview in conjunction with an anniversary symposiumTen Years of Science withChandra Symposium in BostoMassachusetts. Calibration talks will be presented on M

day September 21, prior to the Symposium. Calibratiolated posters will be displayed throughout the Sympowhich runs from Tuesday, September 22 through FrSeptember 25.

For more information, see http://cxc.harvard.edu/ccr/

http://cxc.harvard.edu/cc r /

September 21, 2009

Chandra Calibration Review

Preceding the Ten Years of Science with Chandra symposium

Boston, Massachusetts


22/48

CXC Newsletter 22


23/48

Winter, 2009

WORKSHOP ABSTRACT


24/48

CXC Newsletter 24

T iming for the publicity of science results can be chal-lenging. There are many factors to juggle: accep-tance of the paper into a journal, schedules of the mainscientists, constraints from NASA, and others. We try toavoid holding press conferences near the holidays, sincewe don't want to hear that half of the world's science re-porters are heading off for vacation instead of writingabout our latest result. But, mid-December can work just

ne, since for the second year in a row we have held a suc -cessful press conference at this time. Last year, the workon 3C321 by Dan Evans and others (which became knownas the 'death star galaxy') was a big hit in the media. http://

chandra.harvard.edu/photo/2007/3c321/. This year, a pressconference was held on December 16th, 2008 to announce"Dark Energy Found Sti ing Growth in Universe", workled by Alexey Vikhlinin at CfA. http://chandra.harvard.edu/chronicle/0408/darkenergy/. This press conferencegenerated widespread press coverage, including the NewYork Times, the Washington Post, USA Today, NPR, theAssociated Press, Reuters and the Economist, and manyothers.

Why did this result generate so much attention? Thereare several reasons. First, it represents a signi cant advancein astrophysics, and that is important to experienced sci-

ence journalists. Also, several well-regarded astronomerswho were not involved with the research gave positivesupporting comments, and this helped reporters appreciatethe signi cance of the ndings. Alexey's work gives thebest results so far from studying the effects of cosmic ac-celeration on the growth of large-scale structure, which isfundamentally different from studying cosmic accelerationitself. Not only does it allow important properties of darkenergy to be independently determined, but it permits con-straints on the behavior of gravity over large scales, whencompared with distance measurements. In other words,this work was unlike anything else that reporters had heard

before: a key to getting and keeping their attention.It is worth noting that Alexey and his colleagues did notpublish their cosmological constraints until their surveywas complete and they had carried out detailed simula-tions and cross-checks of their results. This patience wasrewarded with the recognition by cosmology experts thattheir results represented more than an incremental ad-vance. This may not have been the case if the results hadbeen allowed to trickle out. With this approach, there wasa danger that the basic results would be "scooped" by a

different team, for example a group using weak lensia probe of growth of structure, but a competitive resunot emerge.

Another important reason for the success of this stothat dark energy is now one of the hottest research eldsin science, with new papers appearing almost everyon astro-ph. According to the Dark Energy Task F"nothing short of a revolution in our understanding odamental physics will be required to achieve a fullderstanding of the cosmic acceleration." As a result oimportance, an aggressive and expensive set of ovational programs are either underway or planned to

dark energy in more detail. This has reached the where there has even been a backlash with, for examwritten and verbal debates between a prominent astmer and physicist about whether such attention is for the eld of astronomy. These developments are familiarto many astronomers, of course, but they have also noticed by science reporters.

Finally, and perhaps most importantly, cosmology iof the most exciting and stimulating elds in all of sciencefor the public. Questions like "What is the Universe of?," "What did the Universe look like in the past?,"What is the destiny of the Universe?" have been a

for millennia. If a result can give interesting new ansto any of these big questions, it is compelling to the pand to reporters. If it can address all of these questionthe recent work does, that is even better. Also, studidark energy address intriguing questions about the nof the vacuum, such as whether it is possible for noto weigh something.

Given the strength of this story, it did not matter thaChandra images were not the most eye-catching oneshave ever released, or that the indirect effects of repugravity were not trivial to explain, or that this result dcontain any major surprises. The advantages overwhe

the disadvantages, resulting in the excellent press cage.OtherChandra results on cosmology have also receiv

a signi cant amount of press coverage, including the workon the Bullet Cluster by Doug Clowe and collaborato2006, and independent work on dark energy by Stevlen and collaborators in 2004. We continue to look foresults in this eld to publicize. Maybe we'll do anothermid- December press conference!

B ringing C osmology to the p uBliC

m egan w atzke , p eter e dmonds


25/48

Winter, 2009

Date PI Objects Title

January 9 Ralph Kraft (CfA) Cen AJet Power and Black Hole AssortmentRevealed in NewChandra Image

January 10 Rodrigo Nemmen (Penn State) 9 galaxies Chandra Data Reveal Rapidly WhirlingBlack Holes

February 13 Rasmus Voss(MPE)Gijs Nelemans (Radboud Univ.) SN 2007onPossible Progenitor of Special SupernoType Detected

March 20 Armin Rest (Harvard) SNR 0509-67.5Action Replay of Powerful Stellar Expsion

April 28 John Fregeau (Northwestern) 13 globular clustersOldest Known Objects May Be Surprisingly Immature

May 14 Steven Reynolds (NorthCarolina State) G1.9+0.3Discovery of Most Recent Supernova iOur Galaxy

June 18 Sera Markoff (Univ. Amster-dam) M81 Black Holes Have Simple Feeding Hab

July 16 Philip Humphrey (Univ. Cali-fornia Irvine) NGC 4649 A New Way to Weigh Giant Black Ho

September 25 Franz Bauer (Columbia) SN 1996cr Powerful Nearby Supernova Caught byWeb

October 30 Gary Steigman (Ohio State) Bullet Cluster Searching for Primordial Antimat

December 16 Alexey Vikhlinin (CfA) 86 galaxy clustersDark Energy Found Sti ing Growth in theUniverse

CxC 2008 s CienCe p ress r eleases

m egan w atzke

To Change Your Mailing Address:http://cxc.harvard.edu/cdo/udb/userdat.html

CXC: http://chandra.harvard.edu/

CXC Science Support:http://cxc.harvard.edu/ CXC Education and Outreach:

http://chandra.harvard.edu/pub.html

ACIS: Penn Statehttp://www.astro.psu.edu/xray/axaf/

High Resolution Camera:http://hea-www.harvard.edu/HRC/HomePage.html

HETG: MIThttp://space.mit.edu/HETG/

LETG: MPEhttp://www.mpe.mpg.de/xray/wave/axaf/index.php

LETG: SRONhttp://www.sron.nl/divisions/hea/chandra/ CIAO:

http://cxc.harvard.edu/ciao/

Chandra Calibration:http://cxc.harvard.edu/cal/

MARX simulatorhttp://space.mit.edu/ASC/MARX/

Useful Chandra Web Addresses

MSFC: Project Science:http://wwwastro.msfc.nasa.gov/xray/axafps.html


26/48

CXC Newsletter 26

C handra s CienCe p uBliCations

p aul g reen , For Cdo

C handra is nearing a decade of robustly successfulscience observations. The number of Chandra -re-

lated science publications continues to grow strongly. Thegure here shows the number of refereed Chandra Science

Papers by publication year.The Chandra Data Archive (CDA) Group at the CXC

tracks such publications by a daily query to the ADS. Allpublications are scanned and reviewed by hand, catego-rized by several parameters and entered into the Bibliog-raphy Database. The website http://cxc.harvard.edu/cda/bibstats/ is updated monthly with fresh plots and datatables. A full list of the articles tallied is produced, withlinks to the ADS abstracts. The pages also contain detaileddescriptions of the classi cations we use, for instance,what do we de ne as a "Refereed Chandra Science Paper".The CXC also compiles statistics onChandra -related PhDdissertations.

There is some unavoidable lag time for papers to ap-pear in the ADS after they are published, and then the CXCBibliography Database is updated within a couple of weeksof these new ADS entries. From experience, the asymp-totic paper count for a given year is thus not available untilabout April of the year following, which explains why wemark the last bar differently in Figure 22.

Keep those papers coming!

FIGURE 22: The number of refereedChandra Science Papers by publication year. Based on experienwe expect that the full tally of 2008 papers will not be complete until sometime in Spring 2009.

V ersion 4.1 of theChandra Interactive Analysis oObservation (CIAO) software was released incember 2008, followed immediately by a small patch to include data les for proposal planning purposes.

CIAO 4.1 includes several improvements and bug xesto some of the CIAO tools, libraries and GUIs (the renotes describe all these in detail and Figure 23 showexample), but the main feature of this release is an allnon-beta version of Sherpa, the CIAO modeling and ttingpackage. A re-designed, re-written Sherpa was releasethe rst time as beta version in December 2007. Now, thefully functional package is available for the communi

Sherpa supports forward tting of 1-D X-ray spectrafrom Chandra and other X-ray missions, tting of 1-Dnon-X-ray data, including ASCII data arrays, radial

les, and lightcurves. The options for grating data analysisinclude tting the spectrum with multiple response lesrequired for example for overlapping orders inChandra HRC/LETG observations. Modeling of 2-D spatial dfully supported, including PSF and exposure maps ef

The main new feature of the Sherpa redesign concthe Sherpa's environment: it can be run in a S-Lang

Ciao 4.1: with a whole new s herpa !

a ntonella F rusCione , d ouglas Burke and a neta s iemiginowska For the

Ciao team


27/48

Winter, 2009

(http://www.s-lang.org/) or in Python (http://www.python.org/) with only very minor syntax differences. Users canselect the language they prefer and in general will choosethe language they have the most experience with, that theirfriends or colleagues use, or for which they can get the bestsupport. In both cases they will gain scripting and program-ming capabilities, easier access to mathematical calcula-tions, and access to other libraries written by the commu-nity. This new infrastructure greatly enhances the defaultSherpa functions, and provides users with an environmentfor developing complex and sophisticated analysis.

Sherpa is designed to be used as a user-interactive ap-plication or in a batch mode. It is an importable module forthe S-Lang and Python scripting and is available as a C/C++ library for software developers.

Sherpa internal data structures are exposed and fully ac-cessible through high level user interface functions. This

FIGURE 23: An example of thew new PRISM application and its interactive plotting/histogramming capabilitiele loaded into PRISM contains the results from 50,000 iterations of a Metropolis-Hastings sampler (see http://hea-www.

harvard.edu/AstroStat/statjargon.html) used to evaluate the distribution of t parameters for the best- t model found bySherpa. In this case the data is an image of a galaxy cluster with a bright point source close to the center. The data was twith a gaussian2d component to represent the point source, a beta2d component for the cluster emission, and a at back -ground, all convolved with a PSF. Histograms of the cluster and point source center values have been overlaiInteractive Histogram option in PRISM; one window compares the X-axis values and one the Y-axis values. Talso allows the histogram properties (in this case the ll style and color) to be changed before plotting. As can be seen,although the x values are similar, the y values do not overlap which shows that the point source is offset fromcenter. A third plot window shows a sample scatter plot of two t parameters and was obtained with the Interactive Plotoption in PRISM.

accessibility allows users to develop their own comanalysis routines. For example users can input theirmodels, can write speci c analysis procedures not includ -ed in the current Sherpa package or run many simularequired for the planning of future observations.

One example of such a user package is DeprojecPython module for X-ray clusters and other diffuse Xdata that require 3D to 2D model deprojection. This Pextension module was developed by Tom Aldcroft andnow available on the CXC contributed software web http://cxc.harvard.edu/contrib/deproject/

Possibly the main strength of Sherpa is its reliabiliconvergence. There are three optimization methods: mar, a modi cation of the Levenberg-Marquardt algo -rithm which uses the LMDIF algorithm (More 1978);dermead, A Nelder-Mead Simplex direct search (N& Mead 1965); and moncar, a Monte Carlo method


28/48

CXC Newsletter 28

T he CXC announces the availability of PRoPRoVis, at http://cxc.harvard.edu/soft/provis vides interactive plotting of Chandra spacecraft roll, pitchand target visibility for a selected celestial target, usrecent projectedChandra ephemeris. PRoVis replaces told WebVis tool, and adds interactive functionality,cursor readout, revised and expanded output data foand a plethora of plotting format choices. For instancdefault, hatched regions highlight the thermal restricimposed by theChandra pitch ranges available for yocelestial target.

Spacecraft thermal constraints translate into restricon the maximum uninterrupted exposure time for a ta

n ew F unCtionality For p roposers : pr o v is and

m ax e xpo

p aul g reen

FIGURE 24: A snapshot of the PRoVis target visibility interface showing curves of roll, pitch, and target vitarget across the year 2010. The restrictive character of the pitch ranges are optionally shown by the hatched mouse cursor position creates a vertical line and intersecting curve points marked by Xs, whose values for rovisibility are displayed above the plot.

based on the paper by Storn and Price (1997). These morerobust algorithms are a complete replacement of the SherpaOPTIM routines in CIAO 3.4.

Several statistics are available in Sherpa including 2 with various weight options, C statistics and the maxi-mum likelihood statistics as de ned by Cash. Users need tonote that some options of 2 statistics shows bias in tting

low count data, so they have a choice of appropriate statis-tics for low count Poisson X-ray data.

More information on CIAO can be found at http://cxc.harvard.edu/ciao/. Sherpa documentation and sample anal-ysis thread are at http://cxc.harvard.edu/sherpa/

REFERENCES

J.J. More, "The Levenberg Marquardt algorithm: implementation andtheory," in Lecture Notes in Mathematics 630: Numerical Analysis,G.A. Watson (Ed.), Springer-Verlag: Berlin, 1978, pp.105-116).J.A. Nelder and R. Mead (Computer Journal, 1965, vol 7, pp 308-313)

Storn, R. and Price, K. "Differential Evolution: A Simpleand Ef cient Adaptive Scheme for Global Optimization over Continu -ous Spaces." J. Global Optimization 11, 341-359, 1997.


29/48

Winter, 2009

depending on the available pitch angle and season. CXCnow also provides the MaxExpo page http://cxc.harvard.edu/proposer/maxexpo.html with up-to-date static plotsand tables showing the maximumChandra dwell time as afunction of pitch angle and season. The information hereis most useful for proposers whose observations requireconstraints for scienti c reasons, to determine the maxi -mum uninterrupted exposure time for a given target, onceits pitch angle is known, e.g., using PRoVis!

t he 6 th C handra /Ciaow orkshop

a ntonella F rusCione , For the Ciao group

A fter a pause of a few years (the 5th workshopoccurred in 2003) and by popular demand, the

Chandra X-Ray Center (CXC) resumed theChandra /CIAO workshops with a 6th one during 3 days on Octo-ber of 2008. TheChandra /CIAO workshops aim at help-ing users to work with theChandra Interactive Analysisof Observations (CIAO) software.

The workshop included oral presentations by CXCexperts and several hours of hands-on session wherestudents could try to use CIAO "for real" with constantsupport from CIAO team members. Every presentationfor this - and all previous - workshops can be found fromhttp://cxc.harvard.edu/ciao/workshop/index.html Inmany cases these presentations are a very good startingpoint for new and oldChandra /CIAO users looking fora summary of a speci c subject (e.g. Pileup Modelling,Source Detection, etc.).

About 30 people from all of the world and with a va-riety of backgrounds and expertise in X-ray data analysisattended the talks and the hands-on session, putting theCIAO tools and documentation to a real test!

A feedback survey distributed at the end showed uswhat we did well and where we could improve in thefuture. Many students expressed strong interest in an"Advanced CIAO Workshop" which we may try in thefuture. Everyone seemed to like the hands-on session andthe availability of immediate on-site support. One surveyanswer for them all:4. Did you get enough support dur-ing the hands-on session? That was simply perfect, andpeople were so nice!

Thanks to all for coming and look out for the next an-nouncement.

FIGURE 25: Douglas Burke on "MergingChandra Observations"

FIGURE 26: Workshop participants hard at workduring the hands-on session

FIGURE 27: Helping students!

FIGURE 28: Even the DS9 expert was called in...


30/48

CXC Newsletter 30

Chandra Deep Fields

I n 2008 we released new custom processed datasets forChandra Deep Fields, North and South. These datasets contain the merged 2Ms observations for CDFN andCDFS. The credit for the reprocessing of these data goes toall of CXC Data Systems. Descriptions of and access to thedatasets can be found from our Contributed Dataset Pages

http://cxc.harvard.edu/cda/Contrib/CDFN.htmlhttp://cxc.harvard.edu/cda/Contrib/CDFS.html

Bibliographic Database

Our Chandra Bibliography Database and classi cationsystem is undergoing a major overhaul and expansion. Thedatabase has been redesigned to be mission independentand to work with a new suite of tools for querying ADS,populating the database, downloading papers, and scan-ning papers for content in an automated fashion. Becauseof our improvements in scanning the content of papers,we now look at every astronomy paper in ADS, thus mak-ing our database much more complete. However, becauseof the high volume, we still limit physics and instrumentpapers to those which containChandra and its instrumentsin the title or abstract.

The new system also has a plug-in feature which givesa data center the opportunity to connect mission-speci cinformation to paper content. We use this feature to aginstrument con guration information in the bibliographydatabase from theChandra Observation Catalog. The sys-tem also keeps a complete history of edits made to bib-codes and allows for multiple reviewers

Documents

Chandra X-ray Observatory Newsletter 2009