------------------------------------------------------------------------ From: wqd@astro.umass.edu To: gcnews@aoc.nrao.edu Subject: 2009MNRAS.399.1429 \documentclass[manuscript]{aastex} %\documentclass{emulateapj} \usepackage{epsf,wrapfig} \usepackage{graphics} \begin{document} \title{A Large Scale Survey of X-Ray Filaments in the Galactic Center} \author{S. P. Johnson, H. Dong, and Q. D. Wang} \affil{Department of Astronomy, University of Massachusetts Amherst, Amherst, MA} %\author{M. P. Muno} %\affil{Department of Physics and Astronomy, University of California, Los Angeles, CA 90095} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Abstact %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{abstract} We present a catalog of 17 filamentary X-ray features located within a 68'x34' view centered on the Galactic Center region from images taken by \textit{Chandra}. These features are described by their morphological and spectral properties. Many of the X-ray features have non-thermal spectra that are well fit by an absorbed power-law. Of the 17 features, we find 6 that have not been previously detected, 4 of which are outside the immediate 20'$\times$20' area centered on the GC. 8 of the 17 identified filaments have morphological and spectral properties expected for pulsar wind nebulae with X-ray luminosities of 10$^{32}$-10$^{33}$ ergs s$^{-1}$ in the 2.0-10.0 keV band and photon indexes of $\Gamma$=1-2. In one feature, we suggest the strong neutral Fe K$\alpha$ emission line to be a possible indicator for past activity of Sgr A*. For G359.942-0.03, a particular filament of interest, we propose the model of a ram pressure confined stellar wind bubble from a massive star to account for the morphology, spectral shape and 6.7 keV He-like Fe emission detected. We also present a piecewise spectral analysis on two features of interest, G0.13-0.11 and G359.89-0.08, to further examine their physical interpretations. This analysis favors the PWN scenario for these features. \end{abstract} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Introduction %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \section{Introduction} The Galactic Center (GC) region is a unique environment that is home to numerous energetic processes. Radio, infrared and X-ray observations show structures that can only be distinguished in our own Galaxy due to its proximity. The GC thus provides a laboratory in which to probe the Galactic nuclear environment and the interactions between star formation regions, the interstellar medium (ISM) and the supermassive black hole Sgr A*. These observations can help to understand the nuclear environments of nearby galaxies. Radio observations have detected prominent non-thermal radio filaments which have been intensively studied \citep[e.g.][]{Yusef84}. Many of the non-thermal radio filaments may come from milliGauss magnetic fields illuminated by energetic particles. These fields are seen to be primarily perpendicular to the galactic plane and indicate the general structure of the magnetic field around the GC. The exact origins and implications of the radio non-thermal filaments in relation to the GC are still in debate. \textit{Chandra} observations have detected numerous large scale X-ray structures from thermal to non-thermal filaments \citep[e.g.][]{bamba02,lu07,muno07} primarily in the local 20'$\times$20' area centered on the GC. The commonly accepted scenarios for non-thermal thread-like X-ray features in the GC include pulsar wind nebulae (PWNe) \citep[e.g.][]{wang93,wang06}, with the filament-like structure produced through ram-pressure confinement or strong magnetic fields, and magnetohydrodynamical shock fronts from supernova remnants (SNRs) \citep{Yusef05}. For the SNR case, the elongation of features represent shock fronts in the ISM. For the PWN case, the average non-thermal spectrum is well fit by a power law model with a photon index in the range of $\Gamma$=1.1-2.4 \citep{gotthelf02} and X-ray luminosities typically in the range of $10^{32}$ to $10^{37}$ ergs s$^{-1}$ for the 0.2-10 keV energy band \citep[e.g.][]{gaensler06,kaspi06}. The PWN scenario is motivated primarily by the similarities between the features seen in the GC and X-ray features associated with known pulsars \citep[see also][]{karga08}. Few cases of K$\alpha$ transitions from neutral iron (Fe) stem from the reflection of hard X-rays from external sources such as the supermassive black hole Sgr A* \citep[e.g.][]{koyama89}. By analyzing these X-ray features, one can in principle trace gas dynamics and magnetic fields in the GC region \citep[e.g.][]{wang02} and examine the history of GC activity \citep{koyama89}. Previous broad scale studies of X-ray features focus on the immediate 20'$\times$20' area surrounding Sgr A* \citep[e.g.][]{muno07,lu07}. Here, we present a wide scale study on 17 X-ray thread-like features. The features are detected over a roughly 68'x34' area of the sky, relating to approximately 158 pc by 79 pc, around Sgr A* based on all available observations. Some of these features have been examined in previous literature. However, with improved counting statistics, it becomes possible to perform piecewise analysis of the features to detect trends in the spectra and therefore better constrain the physical model \citep[see also][]{lu07}. In addition to the piecewise analysis of some features, we also present an additional physical model in order to explain a feature with Helium (He)-like Fe K$\alpha$ emission line and with a comet-like morphology and steep spectrum. For consistency within this study, the spectral and morphological characteristics were performed for all features listed. All physical distance measurements assume an 8 kpc distance to the GC. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Observations %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \section{Observations and Detection} The \textit{Chandra} X-ray Observatory has observed the GC with the Advanced CCD Imaging Spectrometer imaging array (ACIS-I) for a total integrated exposure time of $\sim$2 Msec between 1999 and 2007 over 81 observations \citep{muno08}. The array is composed of four individual CCDs that operate together to provide a 17'$\times$17' view with sub-arcsecond resolution at the center of the array. The images were first reprocessed using the standard \textit{Chandra} Interactive Analysis of Observations (CIAO) routines (version 3.4.0) then combined by celestial coordinates to produce a merged events file. Initial identification of X-ray filaments comes from a careful visual examination of the merged event image in the 2-8 keV band. We examine the merged events file under a logarithmic scale with different upper/lower limits and look for filament-like structures. After initial identification, the features were then culled so as to remove features whose apparent sizes and/or morphologies differed from the average linear feature in the GC and those that had spectrum consistent with the background. Figure 1 shows the composite intensity map with identified features labeled and indicated by their surrounding ellipses. Figure 2 shows the inner 20'$\times$20' around Sgr A* with features again labeled and constrained by their surrounding ellipses. The intensity maps based on \textit{Chandra} observations are given in the 0.5-8.0 keV band. Source and background spectra were then extracted using our new method described below. In taking the source spectra, point sources and other such possible contaminators were excluded from the source regions and similar steps have been applied for the background spectra. We used local background subtraction for each filament due to the widespread area and time integration over which the filaments are found. For features that are surrounded by strong diffuse emission, e.g. those within the Sgr A* complex, backgrounds were selected to account for the immediate intensities. Typical spectral extraction methods pose a problem for the features that were covered by multiple observations. While some of the 81 observations have the same celestial coordinates in their pointing, they could have different roll angles. As such, a feature could fall into different positions on the ACIS-I detector over different observations including the gap between the CCDs that compose the ACIS-I array. Additionally, a feature may be completely contained within one observation while partially contained in another. Therefore, the traditional spectral extraction could not be used due to time and spatial variability of the Auxiliary Response File (ARF) and Redistribution Matrix File (RMF) of the instrument. Another way is to extract the spectra of the filaments from individual observations and use "FTOOLS" to merge them into a final spectrum. However, this is just suitable for a single object and can become difficult when dealing with many filaments over many observations. Therefore, we developed a new method to directly extract the spectrum from the merged events file. We first project all the events in the merged events file of a certain filament into the instrument coordinate, since the calculation of the ARF and RMF is based on events' positions on the CCD, not the celestial coordinate. The ARF is not only a function of the events' positions, but also depends on when the observation was taken. The quantum efficiency of the CCDs has been seen to degrade over time in part due to molecular contamination build up on the the CCD itself and/or the filter. However, we notice that above 2 keV, the ARF of ACIS-I decreases less than 5\% during the 7 years observational period (absorption towards the GC is large for energies below 2 keV, making the 2 keV lower limit optimal in terms of spectral analysis). Using this lower limit, the ARF can then be considered to be time-independent. The RMF file is very stable and just the function of position, not the time (we have checked the RMF file used by the CIAO's routine for each observation one by one. All of them use the same calibration file). Therefore, a new RMF file (the multiple of the ARF and RMF file) for each small grid on the CCD (the place of each grid in the CCD is predefined by the calibration file) with events was created from the calibration file and the total number of events falling into each grid was used as the weight for merging all the new RMF files in different grids into the final merged RMF. We have tested this method in the center 7 arc minute region around SGR A* and found that the fitting result from this method is nearly identical to that from the same region in the observation with the longest exposure time, proving that our method is robust and reliable. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % X-ray analysis %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \section{Analysis of X-ray Features} For each of the identified filaments, we classify its basic appearance as well as spectral proprieties. Based on their appearance, we can classify the features as being either 'filamentary' or 'cometary'. The sizes and shapes of the features are approximations based on the observed flux and apparent surface brightness relative to the local background. The name associated with each feature comes from the approximate center or brightest point of the feature. Figures 3 through 7 show the individual features with ellipses indicating regions used in extracting the source and background spectra for each feature. The plot next to each \textit{Chandra} count map give the extracted spectrum for each feature. In each case, the spectrum is well fit using a power law model with foreground absorption in XSPEC (\textit{pha(po)} for short). A Gaussian model is included for those features which exhibit line emission of 6.4 keV neutral Fe or 6.7 keV He-like Fe (\textit{pha(po+ga)}). The spectral parameters for the \textit{pha(po)} or \textit{pha(po+ga)} single fit models as well as the morphological properties for the features are listed in Tables 1 and 2. The following summary of features focuses on those with unique traits or observations. \subsection{G0.223-0.012} G0.223-0.012 is located near the bright X-ray point source CXOGCS J174621.05-284343.2, removed from the image in Figure 3, to the northeast of the feature. The feature's position as well as orientation rejects the possibility that G0.223-0.012 could be the result of a CCD readout streak from CXOGCS J174621.05-284343.2. \subsection{G0.13-0.11} G0.13-0.11 was previously studied by \cite{wang02} as a possible PWN. Unlike other PWN candidates in the GC region, G0.13-0.11's morphological proprieties are fairly unique in that it has a curved morphology similar to that of a co-existing non-thermal radio filament in the radio Arc region \citep[see][Figure 1]{wang02}. The morphology of G0.13-0.11 appears to include two tail-like segments stemming from the apparent point source giving the feature a wing-like appearance. We identified the feature as cometary due to the more prominent eastern segment. \cite{wang02} proposed that the unique morphology of this feature is likely due to interactions of high-energy particles from the pulsar in a strong magnetic field, traced by the radio polarization of the Radio Arc region. To determine if there is some spectral evolution across the feature, as expected for PWNe, we perform a piecewise analysis of the apparent point source and the 'wings' of G0.13-0.11. In separating the point source from the wing, we adopt the 1.5" extraction radius used in \cite{wang02}. A joint fit of the point source and total wing spectra, keeping N$_H$ the same for both features, produces a good fit ($\chi^2$/d.o.f=79.7/86) with N$_H$=6.00$^{+2.34}_{-3.60} \times 10^{22}$ cm$^{-2}$ and $\Gamma$ of 1.04$^{+.65}_{-.93}$ for the point source and 1.10$^{+.29}_{-.75}$ for the wing; these values roughly agree with the analysis in \cite{wang02}. Segmenting the wing and performing the joint fit again provides a joint fit column density of N$_H$=5.23$^{+2.98}_{-2.63} \times 10^{22}$ cm$^{-2}$ and $\Gamma$ of .88$^{+.78}_{-.67}$, .45$^{+.61}_{-.56}$, and 1.19$^{+.73}_{-.62}$ for the point source, wing segment containing the point source and wing segment away from the point source, respectively, with $\chi^2$/d.o.f=90.7/89. While the photon index appears to have steepened away from the point source, the error bars for the photon indexes of the two wing segments have considerable overlap. As such, we can not fully constrain the spectral evolution with the available data. Additional observations can aid in improving the counting statistics and the signal-to-noise ratio to constrain the spectral evolution and nature of G0.13-0.11 with greater certainty. \subsection{G0.017-0.044} Spectral analysis of G0.017-0.044 shows a strong 6.4 keV emission line of neutral Fe K$\alpha$ transition. Given its projection from Sgr A*, approximately 11 pc, and this strong neutral Fe emission, our current hypothesis for this feature is that it is likely caused by radiative illumination from Sgr A* past activity as proposed by \cite{koyama89}. \cite{lu07} studied a clump of diffuse X-ray emission located to the west of G0.03-0.06 (F10 in their study) which has similar line emission and equivalent width as G0.017-0.044. The photon index of G0.017-0.044 is likewise comparable to the photon index of the clump of diffuse emission though with lower signal and thus larger error bars. The similarities between the spectra of this West Clump \citep[as referred to by][]{lu07} and G0.017-0.044 indicate that they are likely produced through the same mechanisms. \subsection{G359.942-0.03} G359.942-0.03 displays morphological properties similar to a PWN but with a noticeable 6.7 keV emission line from He-like Fe. The presence of this line and the steepness of the spectrum indicates that the emission is thermal in nature. We applied the XSPEC model MEKAL with absorption (\textit{pha(mekal)}) for an optically-thin thermal plasma with emission lines. Using the default abundances (XSPEC angr), the model provides a good fit ($\chi^2$/d.o.f=29.4/30) with a temperature of 7.31$_{-2.25}^{+3.37}$ keV ($\sim$8.5$^{+3.9}_{-2.6}\times 10^7$ K) and an absorption column density of N$_H$=18.6$^{+4.22}_{-3.83}\times 10^{22}$ cm$^{-2}$. % (see also Figure 10). The best fit normalization of 4.28$^{+1.16}_{-.84}\times 10^{-5}$ gives a volume emission measure of $\sim$1.11$^{+.31}_{-.22}$ cm$^{-6}$ pc$^3$. In \S~4.2, we examine G359.942-0.03 as a ram pressure confined stellar wind bubble generated by a massive star. \subsection{G359.89-0.08} \cite{lu03} have previously studied G359.89-0.08 in detail from archived \textit{Chandra} data under the SNR and PWN scenarios. The SNR case was determined unlikely in part due to spatial differences between the non-thermal radio emission from the proposed SNR, G359.92-0.09, and the X-ray emission from G359.89-0.08. The high X-ray absorption column density \citep[see][section 6.1.1]{lu03} also questions the SNR scenario. In the previous study, they used two observations pointed at Sgr A* (ID 242 and 1561) with an integrated exposure time of 100.9 ks. By performing a piecewise analysis of G359.89-0.08, using the greater exposure time and counting statistics of this study, we can further constrain the physical model of the X-ray emission. Performing a joint fit, keeping N$_H$ the same for all parts, after segmenting G359.89-0.08 along its length and width, we see no noticeable trend across the width of the feature. That is, the spectral properties of the width segments appear to be mirrored across the major axis of G359.89-0.08 instead of displaying a systematic trend indicative of SNR shock fronts. However, Figure 8 shows a spectral softening across the length of G359.89-0.08 (Figure 9). In contrast, we don't see any sign of the spectral change in the direction of the feature's minor axis, which does not provided supporting evidence for the SNR case. However, due to the relatively large uncertainty present in the joint fits and the lack of a confirmed pulsar signal, the SNR case can not be completely ruled out. \subsection{Joint Analysis of Faint Features} The features G0.223-0.012, G359.55+0.16, G359.43-0.14 and G359.40-0.08 were observed over much lower effective exposure times (in the range of $\approx$100 to 150 ks) than those features near the central region ($\approx$700 ks to 1 Ms). Due to the lower signal-to-noise of these features, the inferred model parameters tend to have rather large 90\% error bars (e.g. N$_H$ upper-lower bound difference of $>10^{23}$ and $\Gamma$ upper-lower bound difference of $>2$). To better constrain the parameter bounds, we perform a joint fit of these 4 'faint' features (faint in terms of low signal-to-noise ratios as their fluxes are still fairly high to be seen in the low exposure time). Performing joint fits with $\Gamma$ and N$_H$ fixed for different models, respectively, the spectral parameters of the 'faint' features are consistent with each other. That is, the N$_H$ of the individual features in the $\Gamma$ joint fit is consistent with the average N$_H$ from the N$_H$ joint fit and similarly for $\Gamma$. The spectral parameters for the N$_H$ joint fit are given at the bottom of Table 2. Due to the low signal-to-noise ratios of the individual features, constraining their spectra and morphology is difficult, even with the joint fits. Nevertheless, these fits show the relative flatness present between the features and the overall photon index consistent with a primarily non-thermal origin. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Discussion %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \section{Nature of the Filamentary X-ray Features} \subsection{Comparison of GC Linear X-ray Threads and Known Pulsars} Outside of the GC, similar linear X-ray features are found directly associated with known pulsars, providing the basis for our PWN interpretation. \cite{karga08} have collected a number of PWNe detected with \textit{Chandra}. Many of the PWNe listed are affected by the relative motion with the ambient medium, e.g. ram pressure confinement. These features closely resemble GC features in their morphological appearance, most notably the Mouse PWN \citep[ID 22][]{karga08}. It should be noted that not all X-ray features associated with pulsars can be easily classified using a simple model, such as the pequilar case of the puslar B2224+65, also responsible for the Guitar Nebula, and its linear X-ray feature which is $\sim$118$^\circ$ offset from the pulsar's proper motion \citep{hui07}. Based on known pulsar/X-ray feature associations, we are able to derive simple models for many of the GC X-ray filaments. While the X-ray counting statistics for many of the features studied are indeed low, the power-law model provides a reasonable fit to the spectra. The spectral parameters of many of the brighter features then support the PWN scenario while the low signal-to-noise ratios of the fainter features do not give enough information to accurately constrain their physical models. Moreover, scattering towards the GC makes the detection of the radio pulsar signal difficult, which prevents a definite confirmation of the PWN scenario at present. For those features which show emission lines (namely G0.017-0.044 and G359.942-0.03), the inclusion of a Gaussian profile with the power-law model reduces the $\chi^2$ compared to the power-law only fit with F-test significance $>$98\%. Statistically, other thermal models, such as Bremsstrahlung, produce satisfactory fits but would have unreasonably high temperatures, i.e. $>$10 keV \citep[see also][]{lu07}. The presence of lines at either 6.4 keV or 6.7 keV are obvious indicators that the features are not PWNe. The 6.4 keV line, in particular, is likely due to fluorescent emission from either reflection of transient bright X-ray sources (e.g. Sgr A*; \citet{koyama89}) or local enhanced low energy cosmic ray electrons \citep{yusef02}. \subsection{G359.942-0.03 as a Ram Pressure Confined Stellar Wind Bubble} We find a point-like near-IR source that may be associated with G359.942-0.03 in the 2MASS catalog \citep[ID 17453582-2900050,][]{skrutskie06} and in a recent HST/NICMOS survey of the GC at 1.90 $\mu$m \citep{wang08}. The 2MASS counterpart has an apparent H-band magnitude of 10.85$\pm0.06$ and an H-K color of 1.67 mag, in agreement with those observed for massive stars in the GC region. Interestingly, the color excess E(H-K)=1.72 (assuming (H-K)$_{intrinsic}\approx$-.05 for early-type stars \citep[see also][]{figer99a,panagia73}) is significantly lower than that inferred from the X-ray absorption (see below). This discrepancy is expected if the star has a strong wind, which can produce a free-free emission enhancement in the H-band. Indeed, this enhancement is known for such stars. For example, the brightest massive star identified in the Arches Cluster \citep{figer02} has an apparent H-band magnitude of 9.5 (2MASS ID 17455043-2849215), comparable to that observed for the counterpart of G359.942-0.03. Another massive star in the Archers Cluster (\# 8 in \cite{figer02}) has been examined in a bit more detail with its line-of-sight velocity ($\sim$54 km s$^{-1}$) and wind speed (1100 km s$^{-1}$). This source has an H-band magnitude of 10.8 (2MASS ID 17454970-2849258). This comparison thus indicates that the counterpart of G359.942-0.03 is indeed a massive star with a strong wind. To see if the presence of the massive star would lead to the detectable Pa$\alpha$ emission in the HST/NICMOS survey by \cite{wang08}, we assume that the surrounding material is optically thick to the ionizing photons and that the star has a luminosity of 10$^{49}$ ionizing photons per second (typical of a O type star). We can then infer an intrinsic Pa$\alpha$ flux of $\sim$4.6 $\times 10^{-13}$ ergs s$^{-1}$ cm$^{-2}$ from the Pa$\alpha$ to ionizing photon ratio, which is a product of the n=4 to total recombination ratio \citep[Eqs. 10.64 \& 10.65][]{rybicki} and the n=4 to n=3 transition ratio \citep[Eq. 4.8][]{osterbrock89}). We further use the column density from the X-ray spectral fit (\S~3.4) to estimate the Pa$\alpha$ extinction, Assuming N$_H\approx 1.61\times 10^{21} A_V$ and that the Pa$\alpha$-emitting region is comparable to the observed X-ray feature of G359.942-0.03 in size, we estimate A$_{Pa\alpha}\approx 16_{-4}^{+3}$. This extinction places the apparent Pa$\alpha$ intensity many orders of magnitude lower than the detection limit of the survey. The Pa$\alpha$ intensity would be lower if it arises from a region larger than that of the X-ray feature, as expected in our ram-pressure confined wind bubble model. We propose this model based on the presence of the massive star as well as the cometary morphology and the apparent thermal spectrum of the X-ray emission (\S~3.4). The motion of the star creates a bow shock ahead of it that sweeps up material from the ISM, similar to the model presented by \cite{vanburen88}. We can infer the model parameters from the properties of the X-ray-emitting gas (\S~3.4). First, the strong stellar wind from the massive star is expected to be heated in a reverse shock to a temperature of $T \sim 6 \times 10^7 {\rm~K} (v_w/2)^2$, where $v_w$ is the stellar wind velocity (in units of $10^3$ km s$^{-1}$). Within the uncertain of the stellar wind, this temperature can be high enough to match the value ($\sim$8.5$^{+3.9}_{-2.6}\times 10^7$ K) inferred from the spectral analysis. Second, from the measured volume emission measure ($\sim$1.11$^{+.31}_{-.22}$ cm$^{-6}$ pc$^3$; \S 3.2), we can further estimate the density of the shocked wind material as $n_e \sim$ 30 cm$^{-3}$, assuming a cylinder volume of the observed width and length (Table 1). Third, further constraints can be obtained from relating the pressures of the shocked wind material and the ram-pressure, $2 n_e k T \sim \mu_H n_0v_*^2$, where $v_*$ is the stellar velocity (km s$^{-1}$), $\mu_H$ is the average molecular weight per hydrogen atom ($\mu_H$=0.6 for approximately solar metallicity) and $n_0$ is the density of the ambient ISM (cm$^{-3}$). The ram-pressure also determines the size of the bow shock. Van Buren \& McCray (1988) showed that the distance to the contact discontinuity of the bow shock is $\sim 1.5l_1$ where $l_1$ is the length scale characterizing the point where the ISM ram pressure equals that of the stellar wind. The length scale $l_1$ is then given by \begin{equation} l_1=(177 {\rm~pc}) \dot{M}^{1/2}v_w^{1/2}v_*^{-1}\mu_H^{-1/2}n_0^{-1/2},??? \end{equation} where $\dot{M}$ is the mass lost rate in units of $10^{-5} M_{\sun}$ yr$^{-1}$. Since the measured width of G359.942-0.03, $\sim$ 0.08 pc, should be $\sim 3l_1$, as the shocked wind gas is primarily responsible for the X-ray emission, we have an estimate of \begin{equation} v_* \sim (27 {\rm~km~s^{-1}}) \dot{M}^{1/2}v_w^{1/2} (n_0/10^4)^{-1/2}???. \begin{equation} >From the above two equations, we infer $\dot{M} = ???$ and $v_*^2 n_0 \sim ???$, consistent with what may be expected from a run-away massive star into a dense cloud. Therefore, this ram-pressure confined wind bubble model naturally explains the thermal spectrum, luminosity, and the morphology of the X-ray feature G359.942-0.03 as well as its apparent association of a massive star. \section{Summary And Final Remarks} To conclude, X-ray features in the GC likely have a heterogeneous origin. Non-thermal features are typically produced through synchrotron emission of PWNe or magnetohydrodynamical shock fronts from SNRs. For the PWN model, one would expect elongation due to either the pulsar movement in the ISM or magnetic field confined flows of the pulsar wind materials. In such a case, one would expect to see spectral softening across the direction of elongation along with typical photon indexes of $\Gamma$=1.1-2.4. For the SNR case, the direction of elongation would be caused by shock fronts such that the photon index would change primarily across the width of the feature. Of the features presented, many exhibit characteristics of PWN, although the SNR case has not been completely ruled out. That is, their non-thermal spectra as well as typical comet-like morphology are consistent with the ram pressure confined PWN model. The lack of any detected pulsar signal from these PWNe prevents a definitive conclusion on their origin. Some of these pulsars may be detected in future pulsar searches using high frequency and high resolution radio observations. In the case of G359.942-0.03, the model of a ram pressure confined stellar wind bubble best explains the detection of the 6.7 keV He-like Fe emission and steep spectral shape. The near-IR images shows an apparent counterpart of G359.942-0.03 consistent with a massive star where the lack of extended emission could be due to a combination of extinction and inefficiency in ionizing photon conversion. In addition to their physical models, the X-ray features provide information into the underlying structure of the GC. Features such as G0.017-0.044 and the "clumps" as studied by \cite{lu07} map out the history of the GC (e.g. the possible radiative illumination of G0.017-0.044 from Sgr A*). The orientations of PWN-like features help to map the magnetic, gas and stellar dynamics of the GC. \cite{lu07} predicted that the orientation of comet like features is due to a strong galactic nuclear wind comparable to typical pulsar velocities oriented away from Sgr A*. Many central features support this claim with the 'tails' pointing away from the Sgr A*. Outer features, including some inner features, such as G359.89-0.08 and G359.942-0.03, do not necessarily follow this claim; however, they can still provide information on the gas dynamics as the vector field of the galactic wind could be influenced by regions of varying density. Additionally, the filament-like features could be regulated by local ordered magnetic field lines as with G0.13-0.11's morphology. For the GC observations, few X-ray filaments were identified in regions outside the central 20'$\times$20' region. These regions have integrated exposure times significantly less than the central region. In addition, few of the identified features within the central 20'$\times$20' region were found in initial observations with exposure times $<$200 ks. Thus, the features identified may represent only a tip of the "iceberg" of X-ray features. Indeed, some features may be aligned such that they appear like point sources with their direction of elongation along the line of sight. If one considers the average star formation rate of the GC ($\sim 10^{-7}$ M$_{\sun}$ yr$^{-1}$ pc$^{-3}$ according to \cite{figer04}) then the proposed PWNe along with those undetected through alignments or low signal are within the expected number of pulsars. With higher sensitivity and spatial resolution in multiple wavelength bands, including X-ray, radio and near-IR, the identities of these X-ray features can be more accurately determined. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Acknowledgements %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \section*{Acknowledgments} The project was funded in part by the North East Alliance and the Nation Science Foundation under grant number NSF HRD 0450339 and by NASA/SAO through grant GO7-8091B. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Bibliography %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \begin{thebibliography}{} \bibitem[Bamba et al. (2002)]{bamba02}Bamba, A., Murakami, H., Senda, A., Takagi, S.-I., Yokogawa, J.,Koyama, K. 2002, astro.ph/0202010v1 \bibitem[Bandiera (2008)]{bandiera08}Bandiera, R 2008 astro.ph/0809.2159v1 \bibitem[Barlow, Smith, \& Willis (1981)]{barlow81}Barlow, M.J., Smith, L.J., Willis, A.J. 1981, \mnras, 196, 101 \bibitem[Bessel \& Brett (1988)]{bessel88}Bessel, M.S. \& Brett, J.M. 1988, PASP, 100, 1134 \bibitem[Blum, Sellgren, \& DePoy (1996)]{blum96}Blum, R.D., Sellgren, K., DePoy, D.L. 1996, \apj, 470, 864 \bibitem[Bohlin et al. (1978)]{bohlin78}Bohlin, R.C., Savage, B.D., Drake, J.F. 1978, ApJ, 224, 132 \bibitem[Castor, J., McCray, R., \& Weaver, R. (1975)]{castor75}Castor, J., McCray, R., \& Weaver, R. 1975, ApJL, 200, L107 \bibitem[Cheng, Tamm, \& Wang (2004)]{cheng04}Cheng, K.S., Tamm, R.E., \& Wang, W. 2004, \apj, 617, 480 \bibitem[Cheng, Tamm, \& Wang (2006)]{cheng06}Cheng, K.S., Tamm, R.E., \& Wang, W. 2006, \apj, 641, 427 \bibitem[Crowther et al. (2006)]{crowther06}Crowther, P.A., Hadfield, L.J., Clark, J.S., Negueruela, I., Vacca, W.D. 2006 \mnras, 372, 1407 \bibitem[Figer et al. (1999a)]{figer99a}Figer, D.F, Kim, S.S., Morris, M., Serabyn, E., Rich, R.M., \& McLean, I.S. 1999a, \apj, 525, 750 \bibitem[Figer et al. (2002)]{figer02}Figer, D.F, Najarro, F., Gilmore, D., Morris, M., Kim, S.S., Serabyn, E., McLean, I.S., Gilbert, A.M., Graham, J.R., Larkin, J.E., Levenson, N.A., \& Teplitz, H.I 2002, \apj, 581, 258 \bibitem[Figer et al. (2004)]{figer04}Figer, D.F., Rich, R.M., Kim, S.S., Morris, M., \& Serabyn, E. 2004 \apj, 601, 319 \bibitem[Gaensler \& Slane (2006)]{gaensler06}Gaensler, B.M., Slane, P.O., 2006 \araa, 44, 17 \bibitem[Gotthelf \& Olbert (2002)]{gotthelf02}Gotthelf, E.V. \& Olbert, C.M. 2002, in ASP Conf. Ser. 271, Neutron Stars in Supernova Remnants, ed. P.O. Slane \& B.M. Gaensler (San Francisco:ASP), 171 \bibitem[Guerrero \& Chu (2008)]{guerrero08}Guerrero, A.M. \& Chu, Y.H. 2008 astro-ph:0802.0503v1 \bibitem[Hobbs et al. (2005)]{hobbs05}Hobbs, G., Lorimer, D. R., Lyne, A. G., \& Kramer, M. 2005, \mnras, 360, 974 \bibitem[Hui \& Becker (2007)]{hui07}Hui, C.Y. \& Becker, W. 2007 A\&A, 467, 1209 \bibitem[Kargaltsev \& Pavlov (2008)]{karga08}Kargaltsev, O. \& Pavlov, G.G. 2008 astro-ph:0801.2602v1 \bibitem[Kaspi, Roberts \& Harding (2006)]{kaspi06}Kaspi, V.M., Roberts, M.S.E., Harding, A.K., 2006, \textit{Compact Stellar X-ray Sources}, ed. Walter Lewin \& Michiel van der Klis, Cambridge Astrophysics Series, No. 39 \bibitem[Koyama et al. (1989)]{koyama89}Koyama, K., Awaki, H., Kunieda, H., Takano, S., Tawara, Y., Yamauchi, S., Hatsukade, I., \& Nagase, F. 1989, \nat, 339, 603 \bibitem[Lu, Wang, \& Lang (2003)]{lu03}Lu, F. J., Wang, Q. D., \& Lang, C. C. 2003, \aj, 126, 319 \bibitem[Lu, Yuan, \& Lou (2007)]{lu07}Lu, F.J., Yuan, T.T., \& Lou, Y.-Q. 2007, \apj, accepted (astro-ph/0710.2668v1) \bibitem[Martins et al. (2008)]{martins08}Martins, F., Hillier, D.J., Paumard, T., Eisenhauer, F., Ott, T., \& Genzel, R. 2008, A\&A, 478, 219 \bibitem[Miyazaki \& Tsuboi (2000)]{miyazaki00}Miyazaki, A. \& Tsuboi, M. 2000, ApJ, 536, 357 \bibitem[Muno et al. (2007)]{muno07}Muno, M.P., Baganoff, F.K., Brandt, W.N., Morris, M.R., \& Starck, J.L. 2007, ApJ, submitted (astro-ph/0707.1907) \bibitem[Muno et al. (2008)]{muno08}Muno, M.P.,Bauer, F.E., Baganoff, F.K., Bandyopadhyay, R.M., Bower, G.C., Brandt, W.N., Broos, P.S., Cotera, A., Eikenberry, S.S., Garmire, G.P., Hyman, S.D., Kassim, N.E., Lang, C.C., Lazio, T.J.W., Law, C., Mauerhan, J.C., Morris, M.R., Nagata, T., Nishiyama, S., Park, S., Ramirez, S.V., Stolovy, S.R., Wijnands, R., Wang, Q.D., Wang, Z., \& Yusef-Zadeh, F. 2008, astro-ph:0809.1105v1 \bibitem[Nishiyama et al. (2006)]{nishiyama06}Nishiyama, S., Nagata, T., Kusakabe, N., Matsunaga, N., Naoi, T., Kato, D., Nagashima, C., Sugitani, K., Tamura, M., Tanabe, T., \& Sato, S. 2006, \apj, 638, 839 \bibitem[Osterbrock (1989)]{osterbrock89}Osterbrock, D.E. 1989, \textit{Astrophysics of Gaseous Nebulae and Active Galactic Nuclei}, University Science Books \bibitem[Panagia (1973)]{panagia73}Panagia, N. 1973 AJ, 78, 929 \bibitem[Rieke et al. (1999)]{rieke99}Rieke, M., Alonso-Herrero, A., Becklin, E., McCarthy, D. W., Rieke, G., Scoville, N., \& Stolovy, S. R. 1999 in ASP Conf. Proc 186 "The Central Parsecs of the Galaxy", 32, ed. Falcke, H., Cotera, A., Huschl, W., Melia, F., \& Rieke, M. \bibitem[Rieke \& Lebofsky (1985)]{rieke85}Rieke, G.H., \& Lebofsky, M.J. 1985, ApJ, 288, 618 \bibitem[Rybicki \& Lightman (1979)]{rybicki}Rybicki, G.B., Lightman, A.P. 1979 \textit{Radiative Processes In Astrophysics}, John Wiley \& Sons \bibitem[Sakano et al. (2003)]{sakano03}Sakano, M., Warwick, R. S., Decourchelle, A., \& Predehl, P. 2003, \mnras, 340, 747 \bibitem[Scoville et al. (2003)]{scoville03}Scoville, N.Z., Stolovy, S.R., Rieke, M., Christopher, M.H., Yusef-Zadeh, F. 2003, \apj, 594, 294-311 \bibitem[Skrutskie et al. (2006)]{skrutskie06}Skrutskie, M.F, Cutri, R.M.,Stiening R., Weinberg, M.D., Schneider, S., Carpenter, J.M., Beichman, C., Capps, R., Chester, T.,Elias, J., Huchra, J., Liebert, J., Lonsdale, C., Monet, D.G., Price, S., Seitzer, P., Jarrett, T.,Kirkpatrick, J.D., Gizis, J., Howard, E., Evans, T., Fowler, J., Fullmer, L., Hurt, R., Light, R., Kopan, E.L., Marsh, K.A., McCallon, H.L., Tam, R., Van Dyk, S. \& Wheelock, S. 2006, \aj, 131, 1163. \bibitem[Tsuboi, Handa, \& Ukita (1999)]{tsuboi99}Tsuboi, M., Handa, T., \& Ukita, N. 1999, ApJS, 120, 1 \bibitem[van Buren \& McCray (1988)]{vanburen88}van Buren, D. \& McCray, R. 1988, \apj, 329, 93 \bibitem[Wang, Li, \& Begelman (1993)]{wang93}Wang, Q. D., Li, Z.-Y., \& Begelman, M. C. 1993, \nat, 364, 127 \bibitem[Wang, Lu, \& Lang (2002)]{wang02}Wang, Q. D., Lu, F. J., \& Lang, C. C. 2002, \apj, 581, 1148 \bibitem[Wang, Lu \& Gotthelf (2006)]{wang06}Wang, Q. D., Lu, F. J., \& Gotthelf, E. V. 2006, \mnras, 367, 937 \bibitem[Wang et al. (2008)]{wang08}Wang, Q. D. et al. 2008 in preparation \bibitem[Wong et al. (2003)]{wong03}Wong, D.S., Cordes, J.M, Chatterjee, S., Zweibel, E.G., Finley, J.P., Romani, R.W., \& Ulmer, M.P. 2003 High Energy Processes and Phenomena in Astrophysics, IAU Symposium, Vol. 214 \bibitem[Weaver et al. (1977)]{weaver77}Weaver, R., McCray, R., Castor, J., Shapiro, P., Moore, R. \apj, 218, 377-395 \bibitem[Wilkin, F.P. (1996)]{wilkin96}Wilkin, F.P. 1996, ApJ 459, 31 \bibitem[Yusef-Zadeh et al. (1984)]{Yusef84}Yusef-Zadeh, F., Morris, M., Chance, D. 1984, \nat, 310, 557 \bibitem[Yusef-Zadeh, Law \& Wardle (2002)]{yusef02}Yusef-Zadeh, F., Law, C., Wardle,M. 2002, \apj, 568, L121 \bibitem[Yusef-Zadeh et al. (2005)]{Yusef05}Yusef-Zadeh, F., Wardle, M., Muno, M., Law, C. \& Pound, M. 2005, Advances in Space Research, 35, 107 \end{thebibliography} \clearpage \begin{figure}[] \figurenum{1} \centerline{\includegraphics[width=1.3\textwidth]{../gc_full.ps}} \figcaption[]{Full view of the 81 compiled observations taken by \textit{Chandra} around the GC. Relevant features are defined by their surrounding ellipses and labeled according to their galactic coordinates. The image is the exposure-corrected intensity map produced by dividing the ACIS-I count map with the exposure map. This and subsequent images are given in logarithmic scaling with North defined as up and East as left. \label{fig:fig1}} \end{figure} \begin{figure}[] \figurenum{2} \centerline{\includegraphics[width=1.2\textwidth]{../sou_images/gc.ps}} \figcaption[]{Magnified view of the intensity map from Figure 1 on the inner 20' by 20' area centered on Sgr A*. Identified features again selected and labeled by surrounding ellipses. \label{fig:fig2}} \end{figure} \begin{figure}[] \figurenum{3} \include{fig3} \figcaption[]{\textit{Left:} The X-ray count maps of individual features in the 0.5-8.0 keV band. The upper and lower limits for each count map are 25 counts/arcsec$^2$ and 1 count/arsec$^2$ for G0.223-00.012, 50 counts/arcsec$^2$ and 5 counts/arcsec$^2$ for G0.13-0.11, 150 counts/arcsec$^2$ and 15 counts/arcsec$^2$ for G0.03-0.06, and 150 counts/arcsec$^2$ and 10 counts/arcsec$^2$ for G0.017-0.044, respectively. Source regions are identified by the solid ellipses and background regions by dashed ellipses. \textit{Right:} The X-ray spectra for the matching features given in the left column in the 2.0-8.0 keV band. All spectra fit with absorbed power-law model (\textit{pha(po)}). Model for G0.017-0.044 fitted with a Gaussian centered at 6.4 keV in addition to the \textit{pha(po)}. \label{fig:fig3}} \end{figure} \begin{figure}[] \figurenum{4} \include{fig4} \figcaption[]{Continued X-ray count maps and spectra similar to Figure 3. Upper and lower limits for each count map are 150 counts/arcsec$^2$ and 10 counts/arcsec$^2$ for G0.007-0.014, 100 counts/arcsec$^2$ and 2 counts/arcsec$^2$ for G359.97-0.038, 50 counts/arcsec$^2$ and 1 count/arcsec$^2$ for G359.97-0.009, 100 counts/arcsec$^2$ and 3 counts/arcsec$^2$ for G359.964-0.052, respectively. \label{fig:fig4}} \end{figure} \begin{figure}[] \figurenum{5} \include{fig5} \figcaption[]{Continued X-ray count maps and spectra. Upper and lower limits for each count map are 50 counts/arcsec$^2$ and 2 counts/arcsec$^2$ for G359.96-0.028, and 300 counts/arcsec$^2$ and 5 counts/arcsec$^2$ for G359.95-0.04, and 50 counts/arcsec$^2$ and 3 counts/arcsec$^2$ for G359.942-0.03 and G359.94-0.05, respectively. G359.942-0.03 and G359.94-0.05 fitted with a 6.7 keV Gaussian in addition to the \textit{pha(po)} model. \label{fig:fig5}} \end{figure} \begin{figure}[] \figurenum{6} \include{fig6} \figcaption[]{Continued X-ray count maps and spectra. Upper and lower limits for each count map are 50 counts/arcsec$^2$ and 3 counts/arcsec$^2$ for G359.933-0.037, 250 counts/arcsec$^2$ and 10 counts/arcsec$^2$ for G359.89-0.08, and 25 counts/arcsec$^2$ and 1 count/arcsec$^2$ for G359.55+0.16 and G359.43-0.14, respectively. \label{fig:fig6}} \end{figure} \begin{figure}[] \figurenum{7} \include{fig7} \figcaption[]{Final X-ray count map and spectra. The count map has upper and lower limits of 25 counts/arcsec$^2$ and 1 count/arcsec$^2$. \label{fig:fig7}} \end{figure} \begin{figure}[] \figurenum{8} \centerline{\includegraphics[width=.5\textwidth, angle=-90]{G359.89-0.08_gammaplot.ps}} \figcaption[]{Plot of the photon index $\Gamma$ across the major axis of G359.89-0.08. Positive distance is measure as going northwest away from the 'head'. The error bars are given at 90\% confidence. \label{fig:fig8}} \end{figure} \begin{figure}[] \figurenum{9} \centerline{\includegraphics[width=.9\textwidth]{G359.89-0.08_c.ps}} \figcaption[]{Contour plot of G359.89-0.08. The image, bin size of 1 pixel, was smoothed using a Gaussian with a FWHM of 3 pixels. The upper and lower limits were placed at 30 counts/arsec$^2$ and 1 count/arcsec$^2$, respectively. The contour levels were placed with a step size of 5 counts/arcsec$^2$ and a starting value of 4 counts/arcsec$^2$. \label{fig:fig9}} \end{figure} \clearpage \begin{deluxetable}{llccc} \tabletypesize{\scriptsize} \tablewidth{0pt} \tablecaption{Morphological Properties of Identified X-ray Features} \tablehead{\colhead{ID}& \colhead{Shape}& \colhead{Size}& \colhead{Orientation}& \colhead{Notes}\\ & & pc & &} \startdata G0.223-0.012& filamentary& .23x3.12& ESE-WNW& Faint\\ G0.13-0.11& cometary& .28x1.29& ESE-WNW& Curved feature\\ G0.03-0.06& cometary& .27x1.16& SE-NW& Curved 'tail'\\ G0.017-0.044& cometary& .08x.50& ESE-WNW& Faint\\ G0.007-0.014& filamentary& .12x.39& ENE-WSW& Faint\\ G359.97-0.038& cometary& .23x.54& SW-NE&\\ G359.97-0.009& cometary& .08x.39& SSE-NNW&\\ G359.964-0.052& filamentary& .08x.50& NNE-SSW&\\ G359.96-0.028& filamentary& .12x.39& SE-NW&\\ G359.95-0.04& cometary& .08x.31& NNE-SSW&\\ G359.942-0.03& cometary& .08x.23& ENE-WSW& Faint\\ G359.94-0.05& filamentary& .08x.39& ESE-NWN&\\ G359.933-0.037& cometary& .08x.27& ENE-NWN& Faint\\ G359.89-0.08& cometary& .31x.93& SE-NW&\\ G359.55+0.16& filamentary& .31x2.18& E-W& Faint\\ G359.43-0.14& cometary& .15x.83& N-S& Faint\\ G359.40-0.08& cometary& .20x1.07& S-N& Faint, Curved feature\\ \enddata \tablecomments{\footnotesize{Sizes are taken from apparent surface brightness as compared to the local background. Orientation is based on North and East defined as up and left in \textit{Chandra} images, respectively. Notation for orientation follows that of the cardinal directions. "Faint" refers to features that have low signal-to-noise or low surface brightness when compared to the respective local background.}} \end{deluxetable} \begin{deluxetable}{lllccccr} \tabletypesize{\scriptsize} \tablewidth{0pt} \tablecaption{X-ray Spectral Properties of Identified Features} \tablehead{\colhead{ID}& \colhead{N$_H$} & \colhead{$\Gamma$} & \colhead{F$_x$} & \colhead{L$_x$} & \colhead{EW} & \colhead{Line Energy} & \colhead{$\chi^2/\nu$}\\ & 10$^{22}$ cm$^{-2}$ & & 10$^{-14}$ ergs/s/cm$^2$& 10$^{32}$ ergs/s& keV & keV} \startdata G0.223-0.012& 5.14$_{-5.14}^{+79.8}$& .59$_{-1.58}^{+9.41}$& 15& 15& \nodata& \nodata& 6.7/8\\ G0.13-0.11& 7.02$_{-1.91}^{+2.20}$& 1.52$_{-.17}^{+.48}$& 34& 40& \nodata& \nodata& 96.0/100\\ G0.03-0.06& 6.31$_{-2.78}^{+1.86}$& 1.09$_{-0.37}^{+0.39}$& 12& 13& \nodata& \nodata& 118.9/119\\ G0.017-0.044& 0$_{-0}^{+22.4}$& -.72$_{-.58}^{+2.99}$& 4& 3& .623$_{-.338}^{+.577}$& 6.48$_{-.09}^{+.13}$& 15.4/16\\ G0.007-0.014& 5.74$_{-5.74}^{+16.8}$& 0.97$_{-1.66}^{+3.38}$& 2& 2& \nodata& \nodata& 24.3/20\\ G359.97-0.038& 11.7$_{-1.9}^{+5.1}$& 1.37$_{-0.26}^{+0.79}$& 11& 15& \nodata& \nodata& 109.7/137\\ G359.97-0.009& 9.61$_{-5.36}^{+8.82}$& 1.21$_{-0.61}^{+0.66}$& 4& 5& \nodata& \nodata& 52.5/59\\ G359.964-0.052& 11.1$_{-1.5}^{+2.4}$& 1.87$_{-0.29}^{+0.49}$& 23& 37& \nodata& \nodata& 148.4/177\\ G359.96-0.028& 7.23$_{-3.17}^{+4.50}$& .90$_{-0.40}^{+0.57}$& 5& 5& \nodata& \nodata& 66.3/68\\ G359.95-0.04& 6.0$_{-1.0}^{+1.95}$& 1.82$_{-.15}^{+0.31}$& 52& 93& \nodata& \nodata& 300.7/304\\ G359.942-0.03& 58.8$_{-32.8}^{+32.8}$& 4.13$_{-2.40}^{+2.99}$& 2& 20& .821$_{-.218}^{+.459}$& 6.69$_{-.09}^{+.08}$& 22.3/28\\ G359.94-0.05& -0$_{-0}^{+9.31}$& -0.42$_{-0.96}^{+1.38}$& 3& 2& \nodata& & 5/12\\ G359.933-0.037& 8.29$_{-4.23}^{+5.81}$& .65$_{-0.46}^{+0.37}$& 5& 5& \nodata& \nodata& 45.3/53\\ G359.89-0.08& 31.4$_{-2.0}^{+5.4}$& 1.26$_{-.23}^{+.59}$& 47& 104& \nodata& \nodata& 246.9/240\\ G359.55+0.16& 5.26$_{-3.49}^{+7.11}$& 1.1$_{-.55}^{+1.89}$& 19& 20& \nodata& \nodata& 20.7/23\\ G359.43-0.14& 0$_{-0}^{+4.46}$& -.42$_{-0.72}^{+1.47}$& 6& 5& \nodata& \nodata& 15.2/9\\ G359.40-0.08& 16.2$_{-10.4}^{+12.6}$& 1.26$_{-1.40}^{+1.66}$& 15& 24& \nodata& \nodata& 22.1/24\\ 'Faint' Features& 5.51$_{-3.63}^{+3.91}$&\multicolumn{5}{c}{\nodata}& 70.3/67\\ G0.223-0.012 &\nodata &.74$_{-1.36}^{1.32}$& 15& 15\\ G359.55+0.16 &\nodata &1.21$_{-1.05}^{+1.26}$& 18& 20\\ G359.43-0.14 &\nodata &1.06$_{-1.28}^{0.99}$& 4& 4\\ G359.40-0.08 &\nodata &-0.05$_{-0.64}^{+0.72}$& 20& 18 \enddata \tablecomments{\footnotesize{Spectral parameters are obtained through the XSPEC absorbed power-law model (\textit{pha(po)}). In order, the parameters are the hydrogen gas column density N$_H$, X-ray photon index $\Gamma$, observed X-ray flux and unabsorbed X-ray luminosity in the 2.0-10.0 keV band F$_X$ and L$_X$, equivalent width and energy centroid of the emission line if present, and $\chi^2$ per degree of freedom. Sizes and luminosities are based on the assumed distance of 8 kpc to the Galactic center. Spectral parameters for "faint" features are based on the joint fit model with N$_H$ as a common parameter. Errors are given to the 90\% confidence interval.}} \end{deluxetable} \end{document} \begin{figure}[] \figurenum{10} \centerline{\includegraphics[width=.5\textwidth,angle=-90]{../sou_spec/G359.942-0.03_mekal.ps}} \figcaption[]{ACIS-I spectrum of G359.942-0.03 fitted with the \textit{pha(mekal)} model}. \end{figure} We also need to correct for the line-of-sight extinction. Using extinction relations from \cite{nishiyama06}, \cite{rieke99} and \cite{scoville03}, we find that the Pa$\alpha$ extinction can be related to the observed H-K color by A$_{Pa\alpha}$=1.89E(H-K) where the E(H-K) color excess is found as E(H-K)=(H-K)$_{observed}$-(H-K)$_{intrinsic}$. The 2MASS counterpart has an observed H-K color of 1.67 mag which gives E(H-K)=1.72 and A$_{Pa\alpha}=3.25$ (assuming (H-K)$_{intrinsic}\approx$-.05 for O stars \citep[see also][]{figer99a,panagia73}). With this extinction correction, the predicted Pa$\alpha$ intensity is then $2 \times 10^{-15}$ ergs s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$, if the Pa$\alpha$ region is comparable to the observed X-ray feature of G359.942-0.03 in size. This intensity is a factor of $\sim$2 below the detection limit of $\sim$4$\times 10^{-15}$ ergs s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$ of the Pa$\alpha$ survey \citep{wang08}. The true size of the Pa$\alpha$ region should be larger, making it even more difficult to detect. include the band info in Fig. 1 caption. All images should probably be produced in the 2-10 keV band. The 0.5-2 keV range only contributes to noise. The digits of the long dimension of the 3rd column of Table 1 seem to be wrong (by a factor of 100?).