U.S. XMM-Newton GOF Senior Review Proposal

On behalf of the US Newton X-ray Multi-Mirror Observatory (XMM-Newton) users community, the US Instrument Teams, the NASA/GSFC Guest Observer Facility, and the NASA XMM-Newton Education and Public Outreach program, we request funding for the continued support of US participation in the European Space Agency XMM-Newton mission. The reasons for doing so are compelling: the exceptional scientific return, the complementarity of XMM-Newton and Chandra observations, and the particularly low cost to NASA for enabling access by US astronomers to Great-Observatory class XMM-Newton observations. The response by the US astronomical community to the first two opportunities for submitting XMM-Newton Guest Observer proposals was extensive, with roughly one third of the proposals in both AO's having US PIs. In the first Announcement of Opportunity their success was spectacular, with two thirds of the accepted proposals having either US PIs or Co-Is. ESA does not have the resources to provide support to the large base of XMM-Newton users, and is looking to the US GOF to provide support to the US community. XMM-Newton observations have obtained excellent data for the study of a wide variety of astrophysical phenomena primarily supporting the NASA Structure and Evolution of the Universe theme. The following pages detail the scientific return that XMM-Newton has provided, and will continue to provide, the activities of the US instrument teams, the services provided by the NASA/GSFC GOF, NASA XMM-Newton E/PO activities, and the proposed budget.

XMM-Newton is the second cornerstone in the Horizon 2000 program of the European Space Agency (ESA). XMM-Newton was launched on 1999 December 10, and is in full operation. Over 1500 targets have been observed, with now greater than 70% observing efficiency on average. Over 125 refereed papers have been published. All science data are made public after the expiration of the proprietary period - one year for Guest Observer (GO) data. The XMM-Newton archive opened on 2002 April 15, and there are over 310 observations publicly available as of May 1. For maximum benefit to the science community, ESA is sponsoring the XID program through the Survey Science Centre (SSC). The XID program is designed to identify and perform optical follow-up of serendipitous X-ray sources detected by the EPIC CCD cameras. All XID results will be publicly available through the science archive at the XMM-Newton SOC.

XMM-Newton has lived up to its expectations of high throughput X-ray spectroscopy for the widest variety of astrophysical sources, from comets to quasars. The data obtained in the first two years of operation are just a taste of what is possible to obtain. The enormous archive of serendipitous sources has not yet been fully utilized, and the US GO community is just getting started on the scientific analysis of their data. Like ASCA and ROSAT before it, XMM-Newton has many productive years ahead. The combination of Chandra, Astro-E2, and XMM-Newton will give the world community the best possible combination of high angular resolution, high throughput, and high spectral resolution. Given the success of the US community in the first AO and the high proposal pressure (roughly eight times oversubscribed), XMM-Newton is clearly perceived as one of the world's pre-eminent astronomical observatories, and will continue to be so for the foreseeable future. Without US community access to XMM-Newton, US scientists will be at a severe disadvantage in many areas of research.

The science goals and achievements of XMM-Newton are directly responsive to the NASA Office of Space Science Strategic Plan. With respect to Table II of the plan http://www.hq.nasa.gov/office/codez/plans/SSE00plan.pdf (page 15), XMM-Newton provides unique or important data for all of the first three elements, and part of the element on solar variability. XMM-Newton allows studies of the fundamental processes of neutron stars and black holes, the creation of the elements in supernova explosions, the dispersal of the elements in supernova remnants and starburst galaxies, the evolution of the elements on the largest scale in clusters and groups of galaxies, and the distribution of dark matter in clusters, groups, and elliptical galaxies. The study of the nature of active stars allows for direct comparison with the early solar system and star forming regions for understanding the origin and evolution of stellar systems. XMM-Newton has determined the positions and spectral characteristics of gamma-ray bursts and examined relativistic processes from neutron stars to quasars. XMM-Newton also provides the unique capabilities of the Optical Monitor (OM) which obtains UV and optical data simultaneously with the X-ray instruments. While there is strong potential overlap in the science areas of Chandra and XMM-Newton, each mission has been optimized differently in the five dimensional space of angular resolution, bandpass, collecting area, spectral resolution, and timing ability. It is thus difficult to precisely define the areas in which one mission is superior or inferior to the other, but the world wide astronomical community has decided that both are important with each mission receiving over 800 proposals per AO cycle.

XMM-Newton was designed to observe astrophysical objects over the

keV band. Its large collecting area and highly elliptical orbit allow for long, uninterrupted observations of X-ray sources with an unprecedented sensitivity. The observatory consists of three co-aligned high-throughput 7.5 m-focal-length telescopes with

HPD (

FWHM) angular resolution. XMM-Newton provides images over a

field of view using the European Photon Imaging Camera (EPIC) detectors, which are CCD arrays (two MOS type and one PN type). High-resolution spectra ( $E/{\Delta}E\sim200-800$ ) are provided by the Reflection Grating Spectrometers (RGS) that deflect half of the beam from two X-ray telescopes (those with the EPIC MOS detectors). The sixth instrument is the Optical Monitor (OM), a co-aligned 30 cm optical/UV telescope sensitive in the 1600-6500 Å band. All scientific instruments operate simultaneously, providing exceptionally rich data sets. All of the detectors can be run in a variety of modes, allowing them to be ``tuned'' to the desired science needs, angular, spectral, and temporal, of a given observation. Source positions for faint XMM-Newton sources have an RMS uncertainty of $\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}3''$ over the entire field. The calibration accuracy has reached the 5% level both within and between the EPIC MOS, PN, and RGS, and promises to improve significantly.

During the first 18 months after launch, a variety of software and ground segment problems prevented speedy delivery of data to the GO community. These problems were solved by the fall of 2001, and over 93% of all observations are now delivered within six weeks of observation. The instruments are functioning perfectly except for three anomalies: in the first nine months of operations, two of the 18 chips in the RGS failed and the OM has a stray light problem and a somewhat reduced UV sensitivity. However after these initial difficulties no other lasting problems have occurred. The spacecraft is operating normally, and is more stable than expected before launch. Detailed analysis of the instrument performance shows that the anticipated life of the instruments is consistent with pre-launch estimates of more than 10 years.

Since the launch of XMM-Newton, the relative allocation of time between the Calibration and Performance Verification (Cal/PV), Guaranteed Time (GT), and GO time was heavily modified from the original plan due to the mission operation difficulties. These resulted in much lower than anticipated observing efficiency during the first year of operation. ESA, on the advice of its Science Policy Committee, has decided to complete the GT program by 2003 January. This policy effectively extends the period of AO-1 for the GO community.

The XMM-Newton GO program is fully open to US participation (including US participation in peer reviews). Proposals in response to the AO for the first cycle of open observations (AO-1) were due in 1999 April and those for AO-2 were due in 2001 October. Subsequent AOs will occur annually (with AO-3 due about 2003 March). The peer review of AO-2 has been delayed because of the ESA decision concerning the completion of the GT program, and it is anticipated that the results will be known by mid-July.

The EPIC instruments provide the X-ray imaging capabilities for the XMM-Newton observatory, and are complimentary to Chandra's imagers with significantly larger effective area and field of view but lower angular resolution. The EPIC MOS and PN instruments are functioning nominally, with angular and energy resolution as designed. In periods of low background, the background rates are consistent with preflight estimates. As with Chandra, there are intervals when the background rate increases significantly due to soft protons. Roughly 10% of the total time is contaminated by such flares, similar to Chandra.

All EPIC operating modes have been verified, including the burst and timing modes of the PN and the windowing modes of the MOS and PN. The degradation of the energy resolution by solar particles (the increase in the CCD charge transfer inefficiency, CTI) is as expected before launch, and the instrument should have at least a ten-year lifetime.

Total Energy Band	keV
Field of View	$\sim30'$ diameter
PSF	FWHM, $\sim15''$ HPD
Timing Resolution	7 $\mu$ s to 2.5 s, mode dependent
Spectral Resolution	55 eV at 1 keV
Sensitivity	$\sim10^{-15}$ ergs cm $^{-2}$ s $^{-1}$
Effective Area	2484 cm at 1.5 keV

Two RGS units provide high-throughput, high-resolution spectroscopy of point-like and moderately extended sources. Both RGS instruments are operating nominally with the exception of one CCD on each RGS. However, the lost CCDs cover different wavelength bands so the loss of science capability is minimal.

The spectroscopic resolving power ranges from 200 to 800 over the

keV band, and is intermediate between the HETG and LETG spectrometers on Chandra, but with significantly higher effective area (counting rates are typically $2.5-5\times$ higher). With its exceptionally large dispersion angles, the RGS has a unique capability for spectroscopy of moderately extended sources (full spectral resolution for source diameters up to

, and useful resolution for sources up to

in radius), which has been applied with great success.

As with EPIC, the CTI of the RGS CCDs is slowly increasing over time. The increase is at the levels that were budgeted in the design, indicating that the RGS CCDs should remain functional throughout the planned ten-year lifetime of the mission.

Total Energy Band	Å ( keV)
Field of View	(in cross-dispersion)
Timing Resolution	16 ms to 9 s, mode dependent
Spectral Resolution	0.06 Å (-1), 0.04 Å (-2)
Sensitivity	$\sim3\times10^{-14}$ ergs cm $^{-2}$ s $^{-1}$
Effective Area	180 cm at 15 Å (-1, -2 comb.)

The OM is a 30 cm Ritchey-Chretien optical/UV telescope co-aligned with the X-ray telescopes, and can provide simultaneous optical/UV imaging, spectroscopy, and photometry with the X-ray data. This is a capability not matched by any other observatory. The microchannel-plate intensified photon counting CCD detector allows precision timing with a resolution of 0.5 s (with work underway for 0.05 s resolution). The OM has a wide variety of modes, but basically covers a $17'\times17'$ field of view, sensitive in the 1600-6500 Å band, with an angular resolution of

. The sensitivity in the 3000-6500 Å band is as expected but there has been a reduction in the UV sensitivity compared to pre-launch calculations. There has been no change in instrument performance with time.

Total Bandwidth	Å (6 bands)
Spectral Bandwidth	Å (two grisms)
Sensitivity Limit	V = 23.5 m in 1 ks
Field of View	$17'\times17'$
PSF (FWHM)	(filter dependent)
Timing Resolution	0.5 s (will be 0.05 s)
Spectral Resolution	5.0/10.0 Å (two grisms)
Spatial Resolution	/pixel
Brightness Limit	V 7.4 m (filter dependent)

Because of the wealth of XMM-Newton data and results, we have focused on a small number of what we believe to be the more interesting results. We have tried to represent some of the broad science that XMM-Newton has performed, and is capable of, but realize that this has led us to be quite selective. Given that there have been a total of more than 1700 XMM-Newton GO proposals, the breadth and depth of XMM-Newton science is representative of a ``Great Observatory'' like Chandra or Hubble. Some of the breadth of XMM-Newton can be seen in the abstracts for the 2001 November ESTEC meeting ``New Visions of the X-ray Universe in the XMM-Newton and Chandra era'' http://www.estec.esa.nl/conferences/01C12/, the special issue of Astronomy and Astrophysics devoted to XMM-Newton results (Volume 365), and the GOF publication list http://heasarc.gsfc.nasa.gov/docs/xmm/xmmhp_bibliography.html

A large fraction of the XMM-Newton observing time has gone to this field, and the early results are commensurate with the observation time with over 35 published refereed papers. We have focused on four fundamentally new discoveries.

Since the early ASCA discoveries of relativistically broadened Fe K lines, there have remained lingering detailed problems of interpretation. XMM-Newton has made a major contribution in this area. In the case of MCG-6-30-15, Wilms et al. (2002) have shown that the line is so broad that much of the region producing it must lie extremely close to the central object, requiring the black hole to be spinning rapidly. The intensity and shape of the line are inconsistent with an origin in an externally illuminated accretion disk and, the authors suggest that a large fraction of the energy comes from direct extraction from the black hole rather than accretion.

**Figure 1:** The lower panel shows the 2-10 keV data on NGC 4051 taken with the PN (curve) and RXTE (diamonds, normalized to the PN data). The upper panel shows the OM UV data (solid circles) with the reprocessing model (Mason et al. 2002).
$\begin{figure}\centerline{\psfig{file=agn-2.ps,angle=0,width=3.4in}}\end{figure}$

Despite the extensive optical and X-ray data previously obtained on AGN, the connection between these two bands is not well understood. The ``reflection continuum'' present in many Seyfert 1 galaxies strongly suggests that 50% of the total X-ray energy is reprocessed into UV and optical light. However, in the most detailed study to date, Nandra et al. (1999) showed that there was no correlation between X-ray and UV light curves in the RXTE and IUE observation of NGC 7469. In a 1.5 day XMM-Newton observation of NGC 4051 (a low luminosity Seyfert I galaxy), Mason et al. (2002) report that the UV lags the X-ray by $0.14 \pm 0.02$ d (much longer than expected for an accretion disk), and that

% of the total UV flux originates in the reprocessed component (Figure 1). A geometric model of the reprocessing region suggests that it is ringlike, covers less than 20% of the central source, is

light days from the central object, and is inclined to the line of sight by

degrees. The inability to detect this correlation in previous studies was probably due to insufficient sampling and too short a baseline.

**Figure 2:** RGS1 (red) and RGS2 (blue) spectra of NGC 1068. Line labeling indicates the final state ion. All H-like ( $\alpha$ ) and He-like (r, i, and f) principal order lines are labeled (Kinkhabwala et al. 2002).
$\begin{figure}\centerline{\psfig{file=kink-1.ps,angle=90,width=3.4in}}\end{figure}$

The spectacular XMM-Newton RGS spectrum of NGC 1068 (Figure 2) shows strong, narrow radiative recombination continua, implying that most of the soft photons arise in low-temperature plasma (kT $\sim$ few eV). This plasma is photoionized by an inferred nuclear continuum (obscured along our line of sight). Compared to pure recombination, there is excess emission in all resonance lines up to the photoelectric edge, demonstrating the importance of photoexcitation as well. A cone of plasma irradiated by the nuclear continuum provides a remarkably good fit to the H-like and He-like ionic line series, with inferred radial ionic column densities consistent with recent observations of warm absorbers in Seyfert 1 galaxies. These data and the Chandra images imply that the warm absorber in NGC 1068 is a large-scale outflow. To explain the ionic column densities, a broad distribution of the ionization parameter is necessary, spanning log $\xi=0-3$ ergs cm s $^{-1}$ . This suggests either radially stratified ionization zones or the existence of a broad density distribution (spanning a few orders of magnitude) at each radius.

The XMM-Newton RGS detected for the first time (in IRAS 13349+2438) a broad absorption feature around 16-17Å identified as an unresolved transition array of 2p-3d inner-shell absorption by Fe (Sako et al. 2001). The M-shell ions originate in a much cooler medium than the other X-ray absorption features, implying the existence of at least two distinct regions, one of which is tentatively associated with the medium that produces the optical/UV reddening.

Because of their low flux and low surface brightness, spatially resolved spectroscopy of clusters at greater than one third of the virial radius, and groups over their entire extent, was difficult with Beppo-SAX and ASCA. The important scientific problems of the mass distributions, chemical composition, and cooling flows require XMM-Newton's capabilities.

Cooling Flows: One of the most surprising results from XMM-Newton has come by combining RGS spectroscopy of cooling flows with EPIC spectroscopic imaging. While the distribution of projected temperature with radius derived from the EPIC data is in excellent agreement with the deprojection analysis of ROSAT and ASCA data, it is in strong contradiction to standard cooling flow theory. The gas is not multi-phase and shows much less cool material than expected.

The total soft X-ray luminosity is roughly consistent with the predicted deposition rate, but the emission measure distribution is considerably steeper than the standard cooling flow model (Peterson et al. 2002). The expected emission lines of O VII, Fe XVII, and Fe XX from an isobaric cooling flow model are simply not present, but Fe XXII, Fe XXIV, and other low-energy lines from gas at 50% of $T_{\rm max}$ are observed at the predicted level. Similar results are obtained from elliptical galaxies (Xu et al. 2002).

It should take six times longer for gas in the hot phase to cool from 8 to 3 keV than to cool from 3 to 0.1 keV, so it is perplexing why there is no evidence for further cooling after it has cooled by 85% of $T_{\rm max}$ . It is difficult to find a dynamical timescale, closely connected to the cooling time, which could vary by orders of magnitude depending on the local plasma conditions.

Virial masses of clusters: Before the launch of XMM-Newton, there was considerable controversy about the temperature profiles in clusters (Markevitch et al. 1998, De Grandi & Molendi 2001; Ezawa et al. 1997; White 2000). Some analysis indicated a universal temperature gradient while others indicated isothermal gas. Based on cold dark matter and hydrodynamic simulations, theory predicted a factor of two drop in temperature from the center to 50% of the virial radius (Frenk et al. 1999). It is therefore surprising that the XMM-Newton temperature profiles for a set of clusters (e.g., Pratt et al. 2002) are very flat out to $\sim0.7$ R $_{virial}$ . This is likely to revolutionize our understanding of the formation of large scale structure, and is a challenge to the standard cold dark matter models. The data allow a determination of the total gas mass and its relative fraction with radius. Under the assumption that clusters are fair samples of the universe, this places an upper limit on the density of the universe at $\Omega<0.25{({{\rm h}\over65})}^{-{1\over2}}$ . Furthermore, these new results confirm the scaling between temperature and cluster mass found by ASCA, which lies below the theoretical predictions.

**Figure 3:** The azimuthally averaged EPIC results for the temperature, O, Si, and Fe abundances in the NGC 4325 group (Mushotzky et al. 2002).
$\begin{figure}\centerline{\psfig{file=mush-1.ps,angle=-90,width=3.4in}}\end{figure}$

Chemical Abundances: XMM-Newton allows a derivation of the chemical abundances of O, Si, S, and Fe with radius (e.g., Tamura et al. 2001) in a reasonable number of clusters. At large radii the abundances are constant with radius, but in the central $\sim100$ kpc there is often a sharp rise in Fe, but not in O. The determination of the O abundances for a fair sample of clusters (Kaastra et al. 2002) provides the first robust test of the origin of the elements. A predominance of Type II SN products exist in the outer regions, with a strong contribution from Type Ia's in the central cusp. The first measurement of abundances in groups (see Figure 3) shows a result similar to that of rich clusters - a flat abundance profile and small changes in the relative abundances with radius.

High Redshift Systems: XMM-Newton is ideal for serendipitous searches for clusters (see Figure 4). Pointings have already resulted in the detection of clusters at

, and the detailed study of at least two objects at $z\sim1.3$ (Hashimoto et al. 2002). The detection of hot systems at such large redshifts is a strong constraint on all theories of structure formation and an indication of a low density universe.

**Figure 4:** Mosaic in true X-ray colors of a large solid-angle survey. Red: soft sources ( keV), blue: hard sources ( keV). The individual images have a diameter of (Pierre et al. 2002).
$\begin{figure}\centerline{\psfig{file=clus-1.ps,width=3.4in}}\end{figure}$

XMM-Newton has obtained fundamentally new information on X-ray binaries in nearby galaxies, hot gas in elliptical galaxies, starburst galaxies, and the nature of diffuse emission in nearby spiral galaxies and the Milky Way.

Rapidly Star forming Galaxies: Starburst galaxies and their superwinds are important contributors to the energizing and metal enrichment of the intergalactic medium (IGM). M82 is one of the nearest and brightest of these. The RGS data have shown a wealth of strong emission lines in the spectrum, originating from a wide range of species and temperatures. These data allow a robust determination of the chemical composition of the central regions, and support a direct search for the signature of production of the elements. The EPIC observations have shown that the superwind extends out to $\sim14$ kpc above the disk of the galaxy. These observations can constrain the mass and chemical composition of the superwind, allowing a direct test for the ejection of metals into the IGM. Only XMM-Newton has the combination of angular resolution and collecting area to derive the spectrum of the starburst wind far from the galaxy. Early results on the RGS data for NGC 253 show that the gas is radiating via collisions (Pietsch et al. 2001). The strong Fe XVII lines seen in the nuclear region indicates the presence of a large emission measure of hot gas. If this gas is due to the sum of young supernova remnants, then the implied SN rate is $\sim0.2$ yr $^{-1}$ .

The nature of X-ray emission from ultra-luminous IR galaxies was the subject of much debate in the ASCA and ROSAT era, and has strong implications for the origin of the IR and X-ray backgrounds. The XMM-Newton data (Braito et al. 2002) indicate that these objects are most often composites, with all of them showing hard emission from a power law component. However in only 60% of the sources is it clear that this component is an active galaxy nuclei. All of them show soft emission from hot gas produced by star formation. These preliminary results agree with Chandra observations of far-IR SCUBA sources and ISO observations of XMM-Newton hard X-ray sources (Franceschini et al. 2002), which show that $\sim10$ % of the objects are AGN dominated based solely on their X-ray luminosity. Similar results on three other objects have been obtained by Sanders et al. (2002). The confirmation from XMM-Newton observations that the composite nature of IR selected luminous galaxies extends up to the highest IR luminosities has strong implications for the nature of the highest redshift objects, and the creation of the first black holes.

Nearby Normal Spiral Galaxies: The XMM-Newton data for the M31 X-ray binaries have better spectral resolution and similar sensitivity as UHURU observations of binaries in the Milky Way! For the first time phenomena like bursts, dips, flares, atoll, and Z sources can be studied in other galaxies (e.g., Trudolyubov et al. 2002; Barnard et al. 2002).

**Figure 5:** *Left*: RGS spectrum of N103B (van der Heyden et al. 2002). *Right*: RGS spectrum of 1E0102.2-7219 (Rasmussen et al. 2001).
$\begin{figure}\rightline{\psfig{file=figure1.ps,angle=0,width=7.0in}}\end{figure}$

XMM-Newton has measured the spectral parameters for large samples of ultra-luminous X-ray sources in nearby galaxies (Soria & Kong 2002) and, combined with Chandra observations, have developed the first large body of long-term light curves. Some of these objects are transient in nature. The simultaneous OM UV images provide tight upper limits on possible massive star companions.

In several nearby spirals, XMM-Newton has detected diffuse gas in the nuclear regions. This gas is line dominated and, in the case of M31, has a temperature consistent with the virial temperature of the bulge (Shirey et al. 2001). In M81 the gas is multi-phase, with the bulk of the emission measure at kT $\sim0.65$ keV. These gas temperatures indicate that the gas is in equilibrium with the central potential well in these galaxies.

Elliptical galaxies: The ISM in elliptical galaxies is a reservoir of stellar mass loss and supernovae. The gas is hot and in equilibrium with the dark matter potential well. ASCA and ROSAT data could not uniquely determine the temperature or abundances of the gas because of the inherent degeneracies in low resolution data. These limitations have been overcome by XMM-Newton (Xu et al. 2002). The detection of optically thick Fe XVII lines allowed a direct check of the thermal and spatial models. The derived abundances of O, Mg, Ne, N, and Fe showed sub-solar abundances for all the elements, and solar or sub-solar ratios of [O-Mg/Fe]. This is in sharp contrast to what is expected from the gas being enriched by Type Ia supernovae and stellar mass loss, and presents a strong challenge to the theories of the origin and evolution of elliptical galaxies and Type Ia supernova rates. The RGS data provide the most accurate abundance determination to date for both the stars and the gas in elliptical galaxies.

As extended objects with high temperatures and a complex set of abundances and ionization conditions, supernova remnants (SNRs) are prime objects for studies with XMM-Newton. Spatially resolved SNR spectra can identify stratification of ejecta, separate the forward and reverse shock regions, measure spatial variations in temperature and ionization conditions, and search for compact hard emission associated with a pulsar or its nebula. The XMM-Newton RGS provides high-resolution spectroscopy of SNRs with radii as large as a couple arc minutes. This permits the use of line diagnostics for density and ionization state determinations, as well as providing sufficient spectral resolution to measure velocities.

Plasma Diagnostics: Plasma diagnostics are at the heart of SNR evolution studies. They are the keys to temperature, ionization, and elemental abundance measurements, leading the way to a better understanding of the progenitor star, a more accurate measure of all the different elements produced in the SN explosion, as well as their relative mixing and ionization states. All these studies require the XMM-Newton capabilities of large collecting area and high spectral resolution.

The results are spectacular. Figure 5 shows the RGS spectra of two young SNRs, N103B and 1E0102.2-7219. N103B displays a rich spectrum indicative of a multi-phase plasma whereas non-equilibrium ionization effects dominate the 1E0102.2-7219 spectrum. Elemental abundances are tracers of different periods in the history of the SNR's evolution: The Fe L lines that are detected in 1E0102.2-7219 most likely trace the swept-up mass, and the absence of N suggests that what was present in the progenitor was either burned into other elements or blown off before the SN event.

**Figure 6:** EPIC-MOS images of N132D in narrow wavelength bands. Each image is labeled with the principal line-emitting ion in its band (Behar et al. 2001).
$\begin{figure}\centerline{\psfig{file=n132d_im.ps,width=2.75in}}\end{figure}$

The narrow-band EPIC images of N132D (Figure 6) map the elemental and temperature structure in the remnant. With the exception of O, the dominant part of the soft X-ray emission originates from shocked ISM along the southeastern and northwestern edges of the expanding shell. In contrast, strong Fe-K emission is detected near the center, perhaps indicating that the high-Z ejecta are too highly ionized to be observed at longer wavelengths. O $^{\rm 7+}$ is present throughout the remnant, while the brightest spot of O $^{\rm 6+}$ emission is near the northeastern rim. The O $^{\rm 6+}$ emission can be attributed either to low temperatures or to hot, recently shocked material that is in the process of being ionized. On the other hand, the RGS spectrum shows that the intercombination line is very weak whereas the recombination and forbidden lines are strong. This confirms the hot ionizing conditions, and suggests that the O-rich gas is still in the process of being ionized.

XMM-Newton has provided impressive images of SNRs which can be correlated with images at other wavelengths, and used to rule out some production mechanisms. Such a study was done for Cas A (Bleeker et al. 2001) and RCW 86 (Figure 7). In both cases, the lack of correlation between the high energy XMM-Newton and the radio images suggests that the hard X-rays detected in both SNRs are not from synchrotron radiation but rather from non-thermal bremsstrahlung generated by a population of supra-thermal electrons.

**Figure 7:** Mosaic of exposure corrected EPIC PN and MOS images covering the SW, SE and NW of RCW 86. The image is color coded by the X-ray energy: red - soft, green - medium, blue - hard (Vink et al. 2002).
$\begin{figure}\centerline{\psfig{file=rcw86.ps,width=3.0in}}\end{figure}$

Composite and Plerionic SNRs: Particles injected from a central pulsar diffuse through the nebula, the synchrotron cooling of the particle population results in a spectrum that should soften with radius. This variation in the power law index with radius was confirmed by XMM-Newton for the Crab Nebula (Willingale et al. 2001) and other Crab-like SNRs (Warwick et al. 2001 for G21.5-0.9; Bocchino & Bykov 2001 for IC443). For both G21.5-0.9 and IC443, images reveal a faint shell extending beyond the radio plerion. The emission is non-thermal in G21.5-0.9 but could not be uniquely characterized in the case of IC443. These X-ray halos are perplexing because the long synchrotron lifetimes of the radio-emitting particles should typically result in larger radio nebulae.

For the first time, XMM-Newton observations show indications of thermal emission from the outer region of the plerion, 3C58, providing the first look at the effects of its blast wave (Bocchino et al. 2001). This result provides crucial information on very young pulsar-driven systems. Measurements carried out with the RGS (van der Heyden et al. 2001) have also revealed such a thermal shell in SNR 0540-69, housing a bright, fast pulsar. Emission from O, Ne, and Fe associated with the shell of 0540-69 reveals a shock velocity in excess of $2000 {\rm\ km\ s}^{-1}$ . Such emission can give compelling constraints on the age of the remnant, the shock velocity, and the density of the environment.

XMM-Newton EPIC is proving useful in studies of Galactic diffuse hot plasma sources. Guerrero et al. (2002) observed the planetary nebula (PNe) NGC 7009, and have detected extended X-ray emission in its central cavity. The diffuse X-ray emission originates from the shocked fast stellar wind, and the spectra show that the temperature of the hot plasma is $1.8\times10^{6}$ K. The same authors also detect 10 $^{6}$ K emission from the WR wind-blown bubble S308.

Gamma-ray bursts: XMM-Newton's fast response time to ToO's has so far resulted in three rapid observations of Gamma-Ray Burst afterglows. Data acquisition was started within 12 hours of the GRB alert, and the data were made available to the community within a very short period. XMM-Newton is in many ways the ideal observatory for the study of GRB X-ray afterglow emission: it provides a large throughput and bandwidth (including optical/UV!), combined with high spectral resolution with no instrumental tradeoffs required. This combination has already yielded one potentially very interesting result. The afterglow spectrum of GRB011211 observed in the PN camera, for a brief 5000 s period at the start of the ToO observation, appears to show discrete emission from the mid-Z elements (Mg, Si, S, Ar, Ca), but not from Fe (Reeves et al. 2002). This naturally suggests the abundance pattern expected from nucleosynthesis in massive stars, which would be the first direct connection between GRBs and massive stellar evolution.

X-ray Binaries: X-ray binaries (XRBs) and cataclysmic variables (CVs) comprise the largest class of galactic X-ray sources. The deposition of ballistic material onto the surface of a degenerate star results in copious X-ray emission, either from the heated atmosphere of the star, or shocks above its surface. Their proximity, well-defined binary geometry, and range of emission (extending over the entire electromagnetic spectrum) result in a class of objects where the fields of accretion, magnetic activity, degenerate plasma physics, and binary evolution meet.

An unexpected result was the detection of Fe K resonance absorption lines in three high-inclination XRBs, suggesting these may be common characteristics of accreting systems with close to edge-on disks (Parmar et al. 2002). The features are identified with the K $\alpha$ absorption lines of Fe XXV and Fe XXVI, and since they are detected over a wide range of orbital phases, the absorbing material is most likely located in a cylindrical geometry with an axis perpendicular to the accretion disk. The material responsible is constrained to be close to the central source, at a radius 200 times smaller than the outer edge of the accretion disk. Equivalent widths increase during dips, corresponding to a column of $> 10^{17.3}$ Fe atoms cm $^{-2}$ . Until recently the only XRBs known to exhibit narrow X-ray absorption lines were superluminal jet sources, and it had been suggested that these features are related to the jet formation mechanism. It now appears likely that ionized Fe absorption features may be common characteristics of accreting systems with accretion disks.

A broad, skewed Fe K emission line has been detected in the transient source X1650

50 (Miller et al. 2002), which suggests that this system hosts a black hole. The line shape indicates a steep disk emissivity profile that is hard to explain in terms of a standard accretion disk model. Similar to the XMM-Newton data for the Seyfert galaxy MCG 6-30-15, these results can be explained by the extraction and dissipation of rotational energy from a black hole via magnetic connection to the inner accretion disk. If this process is at work in both sources, a fundamental prediction of general relativity will be confirmed across a factor of

in black hole mass.

Cataclysmic Variables: Despite their ubiquitous nature, CVs have low luminosities ( $<10^{33}$ ergss $^{-1}$ ), and so detailed X-ray experiments have only been possible on a small, high luminosity sub-sample of the class. With its high-resolution, high-throughput cameras, its ability to observe bright sources in full timing and spectral modes, and its simultaneous optical and UV capabilities, XMM-Newton is revolutionizing CV astronomy.

A long-standing prediction suggests that post-shock flows are stratified in temperature and density. The RGS data place demanding tests on this model; Figure 8 compares the best stratified model fit with single temperature zone fits. The remaining small residuals in the lower panel are the result of irradiation of the accretion column. Two well-constrained products of the fit are the mass of the white dwarf and the accretion rate. Both are fundamental CV properties that have proved notoriously difficult to measure in the past. A sample of these quantities is necessary to understand the interdependence of CV phenomena, determine the origins of the systems and predict their future evolution. For example, by comparing magnetic field strengths for CVs and single white dwarfs it is apparent that accretion suppresses fields. Masses and accretion rates will help us understand this process. Extending this analysis, Ramsay et al. (2001) have shown that thermal emission from the white dwarf surface is not a standard feature of the CV spectrum. They argue that only the bright, high mass transfer systems radiate thermal spectra in the X-ray band.

**Figure 8:** The RGS spectrum of the magnetic CV EX Hya with stratified model fit. Residuals are shown in the bottom panel. Residuals from 1- and 3-temperature MEKAL model fits are shown in the middle panels (Cropper et al. 2002).
$\begin{figure}\rightline{\psfig{file=cropper.ps,width=4.0in}}\noindent \end{figure}$

Optical and UV emission collected by the OM independently probes the cyclotron emission and the colder atmospheres of the pre-shock accretion flow and companion star. Wheatley & West (2002) have shown that the X-ray and cyclotron emission regions occupy different volumes, the cyclotron region being larger. After detecting a unique new type of X-ray flare from UZ For in its quiescent state, Still & Mukai (2001) and Pandel & Cordova (2002) concluded this was an accretion, rather than coronal, event. Both results would have been impossible without simultaneous OM data.

ASCA observations indicated that Ne was overabundant in many coronal sources, but line blends did not allow conclusive measurements. By taking advantage of large effective area and dispersive spectroscopy, XMM-Newton has settled the argument over the so-called ``Inverse FIP'' (First Ionization Potential) effect (Brinkman et al. 2001). RGS data on the RS CVn system HR 1099 showed an increase of coronal abundances with their respective FIP (elements with a higher FIP are more abundant), the exact opposite of what is observed in the Sun. This result has been confirmed in the rapid rotator AB Dor (Guedel et al. 2001), both showing anomalously strong Ne IX lines. Audard et al. (2001) find an inverse FIP affect in HR 1099 during quiescence but a direct FIP effect during flares. These flares are thought to be directly associated with the process of element fractionation in stellar atmospheres. Since elements with different FIPs will be ionized by differing amounts and low FIP material has a higher probability of being accelerated along magnetic loops and diffused into the chromosphere, there is a higher abundance of high FIP particles in the lower atmosphere that will be lifted into the corona by flares.

An XMM-Newton RGS observation of the Wolf-Rayet (WR) star WR110, which has no known companion, shows an unexpected and unexplained hard X-ray component (equivalent to a 3 keV thermal plasma). The analysis (Skinner et al. 2002) shows that half of the flux from the star arises from the hard component. Simultaneous VLA observations probed the atmosphere and exclude the possibility that this hard X-ray emission arises in wind shocks, leaving unresolved the mystery of its origin.

Observations of the symbiotic star, Z And, obtained shortly after outburst, show a complex spectrum with photon energies out to at least 5 keV (Sokoloski et al. 2002). Using simultaneous FUSE spectra to fix the absorption column yields a spectrum consisting of an 80 eV blackbody, a 0.7 keV thermal plasma, and a hard tail possibly arising in shocked winds. A follow-up observation obtained five months later will reveal the evolution of the X-ray spectrum during the optical decline.

The diffuse X-ray background is the sum of emission from several different regions of hot plasma: the local ISM, Galactic halo, Local Group intercluster medium (ICM), and the hot IGM. It also includes the previously unresolved extragalactic background now known to arise mostly from AGN (e.g., Hasinger et al. 1998).

Studies of this diffuse emission require large effective areas and large solid angles to provide sufficient counts, low detector background, and good angular resolution to allow removal of contaminating point sources. Not only does XMM-Newton EPIC satisfy all these requirements, but its spectral response is good down to $\sim0.2$ keV, covering the energies where the Galactic halo and the hot IGM are most easily observed.

A major outstanding issue is the origin of the diffuse background at high Galactic latitudes in the $3\over4$ keV band. Only about 50% of the observed flux can be attributed to recently resolved cosmological sources. Another 10-20% of the observed flux can be attributed to Galactic emission from thermal plasmas (and a small amount to unresolved stellar contributions). Determining the origin of the remaining fraction has both cosmological and Galactic implications.

Recently, major constraints on the properties of these hot plasmas were determined by absorption studies. Significant O VII and O VIII resonance line absorption at zero redshift were found in deep RGS exposures of the AGN 3C 273, Mkn 421, and PKS 2155-304 (Figure 9). These features, consistent with the detection of O VII and O VIII in the Chandra LETGS spectrum of PKS2155-304 (Nicastro et al. 2002), are definitely at zero redshift. The lines have turbulent Doppler velocities of a few hundred km s $^{-1}$ or less. Measured column densities in O VII and O VIII are of order a few times $10^{16}$ ions cm $^{-2}$ .

**Figure 9:** Spectra of Capella, 3C273, Mkn 421, and PKS2155-304. Absorption is seen at 18.97 Å (O VIII Ly $\alpha$ ) and 21.6 Å (O VII resonance line) in the QSO spectra. Absorption at 23.5 Å is due to neutral O in the Galaxy. The Capella spectrum demonstrates the accuracy of the wavelength scale (Rasmussen et al. 2002)
$\begin{figure}\centerline{\psfig{file=xmm.spectra.ps,angle=-90,width=3.0in}}\end{figure}$

The density and size of the hot plasma can be established by combining absorption measurement with constraints on emission from the same medium. Since the medium cannot overproduce the $1\over4$ keV background, most of the absorption cannot be due to hot plasma in the ISM of our own Galaxy; thus we are seeing absorption by hot plasma located in an extended halo of the Galaxy or in the intercluster gas of the Local Group. The electron temperature of this medium, determined from the ratio of column densities in O VII and O VIII, is 200-300 eV, consistent with expectations based on the dynamics of the Local Group galaxies. This opens up a whole new field of study, with measurements constraining the total baryon census at redshift zero, as well as the evolution of the diffuse gas in collapsed structures of low to moderate overdensity.

**Figure 10:** Left Panel: The distribution of 2-9 keV X-ray emission in the Radio Arc Region. The image covers a region $0.5^\circ \times 0.5^\circ$ . The bright extended source associated with Sgr A East is visible on the right-hand side of the image. Right Panel: The corresponding hardness (H-M/H+M) ratio distribution (Hard: 5-9 keV, Medium: 2-5 keV, Warwick 2002).
$\begin{figure}\centerline{\psfig{file=rwarwick-E1_fig5.ps,angle=-90,width=3.4in}}\end{figure}$

Figure 10 shows the EPIC image and hardness ratio of the 0.4-9 keV emission from the central $\sim 300$ pc of our Galaxy (Warwick 2002). There is complex structure with large-scale extended X-ray emission associated with the non-thermal radio source Sgr A East, asymmetrically distributed around the region. The diffuse X-ray component present in the Radio Arc Region exhibits considerable spectral variations on scales down to

, consistent with a complex mix of thermal and non-thermal contributions. While Chandra has resolved extensive emission in this region into the contributions of point sources (Wang et al. 2002), the origin and nature of the diffuse X-ray emission seen in the Galactic Center Region and in the Galactic Plane in the form of the Galactic X-ray Ridge remain topics of active debate.

The ability of XMM-Newton to rapidly survey large fields to faint levels with accurate positions and a reasonable number of X-ray spectra is a breakthrough in this field. Most exposures reach as faint as the deepest ROSAT surveys in the soft band and ten times fainter than ASCA and Beppo-SAX in the hard band. XMM-Newton detects normal galaxies to z $\sim0.05$ , Seyfert galaxies and groups to z $\sim 1$ , and quasars and rich clusters to arbitrarily high redshift. XMM-Newton data provide a unique database for studying the correlation function of X-ray sources and their clustering properties. XMM-Newton has already undertaken a total of $\sim70$ observations at high Galactic latitude with exposure times greater than 50 ks. On average, each such field contains about 50 serendipitous sources of which $\sim8$ are bright enough for X-ray spectroscopy, and reaches a flux threshold of $2\times10^{-16}$ ergs cm $^{-2}$ s $^{-1}$ in the 0.5-2 keV band (Lumb et al. 2001).

This serendipitous database will provide the largest sample of distant clusters, and allow the measurement of the cosmic evolution of the temperature and luminosity functions of clusters and groups. The XMM-Newton database will be vastly superior to the previous databases used for this purpose. It will push these measurements to higher redshifts, where the sensitivity to the cosmological parameters is greatly enhanced (Henry 2000).

A comparison between the 1 Ms Chandra observation of the Hubble Deep Field North, and a 155 ks XMM-Newton observation of the same field shows that XMM-Newton does extremely well in a much shorter exposure time (Figure 11). There is some source confusion present in the XMM-Newton image, limiting the effective flux threshold in the fields to $F(x)_{(0.5-2keV)} >2\times10^{-16}$ ergs cm $^{-2}$ s $^{-1}$ , but the confusion limit in the 5-10 keV band has not yet been reached. The two Chandra deep surveys show remarkable differences due to cosmic variance and it is clear that the much larger XMM-Newton solid angle and similar sensitivity will allow it to make a major contribution in this area.

**Figure 11:** *XMM-Newton* observation of the Hubble Deep Field North. Red: 0.5-2.0 keV, Green: 2.0-4.5 keV, Blue: 4.5-10.0 keV (Griffiths et al. 2002).
$\begin{figure}\centerline{\psfig{file=grif-3.ps,angle=0,width=2.5in}}\end{figure}$

The XMM-Newton SSC XID program
http://xmmssc-www.star.le.ac.uk/ has released multi-color CCD images for many XMM-Newton fields with details of the identifications of sources in the fields, links to optical finding charts, optical photometry of potential counterparts, and reduced optical spectra. This public release of optical follow-up data is unique to XMM-Newton and will continue as part of the SSC function.

The 2-10 keV background has been almost fully resolved by Chandra into point sources. The good response of XMM-Newton at high energies has been used to show that Type 2 Seyferts and unidentified sources are responsible for the bulk of the extragalactic background (Griffiths et al. 2002). The spectra will provide a strong test of the unified models of the X-ray background. What we still need to understand is whether the rates of evolution of sources with different absorption columns are the same or whether they differ. With its large database of X-ray spectra of serendipitous sources, XMM-Newton will make a major contribution in this area.

12ptBraito, V., et al. 2002, Proc. Symposium ``New Visions of the X-ray Universe in the XMM-Newton and Chandra Era'', ed. F. Jansen, in press

12ptKaastra, J. S., et al. 2002, Proc. Symposium ``New Visions of the X-ray Universe in the XMM-Newton and Chandra Era'', ed. F. Jansen, in press

AAPT	American Association of Physics Teachers
AAS	American Astronomical Society
AGN	Active Galactic Nuclei
AO	Announcement of Opportunity
ASCA	Advanced Satellite for Cosmology and Astrophysics
Cal/PV	Calibration and Performance Verification
CCD	Charge Coupled Device
CDM	Cold Dark Matter
Co-I	Co-Investigator
CTI	Charge Transfer Inefficiency
CU	Columbia University
CV	Cataclysmic Variable
DPU	Digital Processing Unit
EPIC	European Photon Imaging Camera
E/PO	Education and Public Outreach
ESA	European Space Agency
FIP	First Ionization Potential
FTE	Full Time Equivalent
FUSE	Far Ultraviolet Spectroscopic Explorer
FWHM	Full Width at Half Maximum
FY	Fiscal Year
GLAST	Gamma-ray Large Area Space Telescope
GO	Guest Observer
GOF	NASA/GSFC Guest Observer Facility
GRB	Gamma-Ray Burst
GSFC	Goddard Space Flight Center
GT	Guaranteed Time
HEASARC	High Energy Astrophysics Science Archive Research Center
HETG	High Energy Transmission Grating
HPD	Half Power Diameter
HTR	High Time Resolution
ICM	Intercluster Medium
IGM	Intergalactic Medium
INTEGRAL	International Gamma-Ray Astrophysics Laboratory
IR	Infra Red
IRAS	Infra Red Astronomical Satellite
ISM	Interstellar Medium
ISO	Infrared Space Observatory
LANL	Los Alamos National Laboratory
LETG	Low Energy Transmission Grating
MOS	MOS style EPIC CCD detector
NASA	National Aeronautics and Space Administration
NSTA	National Science Teachers Association
OGIP	Office of Guest Investigator Programs
OM	Optical Monitor
PI	Principal Investigator
PPS	Pipeline Processing Subsystem
PSF	Point Spread Function
PN	PN style EPIC CCD detector
PNe	Planetary Nebula
PPS	Pipeline Processing System
PV	Performance Validation
RGA	Reflection Grating Array
RGS	Reflection Grating Spectrometer
RMS	Root Mean Square
ROSAT	Röntgen Satellite
RPS	Remote Proposal System
RXTE	Rossi X-ray Time Explorer

SAS	Science Analysis System
SEU	Structure and Evolution of the Universe
SEUEF	Structure and Evolution of the Universe Education Forum
SN	Supernova
SNR	Supernova Remnant
SOC	Science Operations Center
SRON	Space Research Organization Netherlands
SSC	Survey Science Centre
SSU	Sonoma State University
SWT	Science Working Team
ToO	Target of Opportunity
UCSB	University of California, Santa Barbara
URL	Universal Resource Locator
UV	Ultra Violet
WR	Wolf-Rayet
XID	X-ray source IDenfication program
XMM	X-ray Multi-Mirror Mission
XRB	X-ray Binary

About this document ...

This document was generated using the LaTeX2HTML translator Version 99.2beta8 (1.46)

The command line arguments were:
latex2html -t 'U.S. XMM-Newton GOF Senior Review Proposal' -image_type gif -split 3 -show_section_numbers senior-www.tex