TeV Gamma-Ray Observations of the Galactic Center
Extracted source text ยท arXiv astro-ph/0403422
\documentclass[preprint]{aastex}
\begin{document}
\newcommand{\Cerenkov}{\v Cerenkov }
\newcommand{\Check}{{\Large \emph{\textbf{CHECK THIS}}}}
\newcommand{\Flux}{$1.6 \times 10^{-8} \pm 0.5\times10^{-8}
(\mathrm{stat}) \pm 0.3 \times 10^{-8}
(\mathrm{sys})~\mathrm{photons~m^{-2}~s^{-1}}$}
\title{ TeV Gamma-Ray Observations of the Galactic Center }
\author{ K. Kosack,\altaffilmark{1}
H. M. Badran, \altaffilmark{2}
I. H. Bond,\altaffilmark{3}
P. J. Boyle,\altaffilmark{4}
S. M. Bradbury,\altaffilmark{3}
J. H. Buckley,\altaffilmark{1}
D. A. Carter-Lewis,\altaffilmark{6}
M. Catanese,\altaffilmark{5}
O. Celik,\altaffilmark{7}
V. Connaughton,\altaffilmark{21}
W. Cui,\altaffilmark{8}
M. Daniel,\altaffilmark{6}
M. D'Vali,\altaffilmark{3}
I. de la Calle Perez,\altaffilmark{3}
C. Duke,\altaffilmark{9}
A. Falcone,\altaffilmark{8}
D. J. Fegan,\altaffilmark{10}
S. J. Fegan,\altaffilmark{5}
J. P. Finley,\altaffilmark{8}
L. F. Fortson,\altaffilmark{19}\altaffilmark{20}
J. A. Gaidos,\altaffilmark{8}
S. Gammell,\altaffilmark{10}
K. Gibbs,\altaffilmark{5}
G. H. Gillanders,\altaffilmark{11}
J. Grube,\altaffilmark{3}
J. Hall,\altaffilmark{12}
T. A. Hall,\altaffilmark{13}
D. Hanna,\altaffilmark{14}
A. M. Hillas,\altaffilmark{3}
J. Holder,\altaffilmark{3}
D. Horan,\altaffilmark{5}
A. Jarvis,\altaffilmark{7}
M. Jordan,\altaffilmark{1}
G. E. Kenny,\altaffilmark{11}
M. Kertzman,\altaffilmark{15}
D. Kieda,\altaffilmark{12}
J. Kildea,\altaffilmark{14}
J. Knapp,\altaffilmark{3}
H. Krawczynski,\altaffilmark{1}
F. Krennrich,\altaffilmark{6}
M. J. Lang,\altaffilmark{11}
S. Le Bohec,\altaffilmark{6}
R. W. Lessard,\altaffilmark{8}
E. Linton,\altaffilmark{4}
J. Lloyd-Evans,\altaffilmark{3}
A. Milovanovic,\altaffilmark{3}
J. McEnery, \altaffilmark{17}
P. Moriarty,\altaffilmark{16}
D. Muller,\altaffilmark{4}
T. Nagai,\altaffilmark{12}
S. Nolan,\altaffilmark{8}
R. A. Ong,\altaffilmark{7}
R. Pallassini,\altaffilmark{3}
D. Petry,\altaffilmark{17}
B. Power-Mooney,\altaffilmark{10}
J. Quinn,\altaffilmark{10}
M. Quinn,\altaffilmark{16}
K. Ragan,\altaffilmark{14}
P. Rebillot,\altaffilmark{1}
P. T. Reynolds,\altaffilmark{18}
H. J. Rose,\altaffilmark{3}
M. Schroedter,\altaffilmark{5}
G. H. Sembroski,\altaffilmark{8}
S. P. Swordy,\altaffilmark{4}
A. Syson,\altaffilmark{3}
V. V. Vassiliev,\altaffilmark{7}
S. P. Wakely,\altaffilmark{4}
G. Walker,\altaffilmark{12}
T. C. Weekes,\altaffilmark{5}
J. Zweerink,\altaffilmark{7}}
\altaffiltext{1}{Department of Physics, Washington University, St. Louis, MO 63130, USA}
\altaffiltext{2}{Physics Department, Tanta University, Tanta, Egypt}
\altaffiltext{3}{Department of Physics, University of Leeds, Leeds, LS2 9JT, Yorkshire, England, UK}
\altaffiltext{4}{Enrico Fermi Institute, University of Chicago, Chicago, IL 60637, USA}
\altaffiltext{5}{Fred Lawrence Whipple Observatory, Harvard-Smithsonian CfA, P.O. Box 97, Amado, AZ 85645-0097}
\altaffiltext{6}{Department of Physics and Astronomy, Iowa State University, Ames, IA 50011-3160, USA}
\altaffiltext{7}{Department of Physics, University of California, Los Angeles, CA 90095-1562, USA}
\altaffiltext{8}{Department of Physics, Purdue University, West Lafayette, IN 47907, USA}
\altaffiltext{9}{Department of Physics, Grinnell College, Grinnell, IA 50112-1690, USA}
\altaffiltext{10}{Experimental Physics Department, National University of Ireland, Belfield, Dublin 4, Ireland}
\altaffiltext{11}{Department of Physics, National University of Ireland, Galway, Ireland}
\altaffiltext{12}{High Energy Astrophysics Institute, University of Utah, Salt Lake City, UT 84112, USA}
\altaffiltext{13}{Department of Physics and Astronomy, University of Arkansas at Little Rock, Little Rock, AR 72204-1099, USA}
\altaffiltext{14}{Physics Department, McGill University, Montre$\acute{a}$l, QC\,H3A\,2T8, Canada}
\altaffiltext{15}{Department of Physics and Astronomy, DePauw University, Greencastle, IN 46135-0037, USA}
\altaffiltext{16}{School of Science, Galway-Mayo Institute of Technology, Galway, Ireland}
\altaffiltext{17}{University of Maryland, Baltimore County and NASA/GSFC, USA}
\altaffiltext{18}{Department of Applied Physics and Instrumentation, Cork Institute of Technology, Cork, Ireland}
\altaffiltext{19}{Department of Astronomy and Astrophysics,
University of Chicago, Chicago, IL, USA}
\altaffiltext{20}{Astronomy Department, Adler Planetarium and
Astronomy Museum, Chicago, Il, USA.}
\altaffiltext{21}{Gamma-Ray Astrophysics Group National Space
Science and Technology Center Huntsville, Alabama}
\email{kosack@hbar.wustl.edu\ \ buckley@wuphys.wustl.edu}
\begin{abstract}
We report a possible detection of TeV gamma-rays from the Galactic
Center by the Whipple 10m gamma-ray telescope. Twenty-six hours
of data were taken over an extended period from 1995 through 2003
resulting in a total significance of 3.7 standard deviations. The
measured excess corresponds to an integral flux of \Flux
above an energy of $2.8\ \mathrm{TeV}$, roughly 40\% of the flux
from the Crab Nebula at this energy. The 95\% confidence region
has an angular extent of about 15 arcmin and includes the position
of Sgr A*. The detection is consistent with a point source and
shows no evidence for variability.
\end{abstract}
\maketitle
%======================================================================
\section{Introduction}
%======================================================================
The central region of our galaxy is now thought to contain a
super-massive black-hole of 2.6$\times 10^6 M_\odot$
\citep{2002AAS...201.6804G,2002Natur.419..694S} coincident with the
unresolved radio source Sgr A* \citep{balick1974}. Chandra
observations reveal X-ray emission from an unresolved point source
as well as an extended structure ($\sim 1.5 \ \mathrm{arcsec}$),
both of which appear to be physically associated with Sgr A* (e.g.,
\citet{2003ApJ...591..891B}). The recent discovery of hour-scale
X-ray \citep{2001Natur.413...45B} and rapid IR flaring
\citep{2004ApJ...601L.159G} point to an active nucleus, albeit with
very low bolometric luminosity compared with the luminosity inferred
from the Bondi accretion rate or with that which is typical of more
powerful AGNs. More recently, \emph{INTEGRAL} (\emph{the
International Gamma-Ray Astrophysics Laboratory}) has detected
time-variable 20-100 keV emission from within 0.$^{'}$9 of Sgr
A*\citep{2004ApJ...601L.163B}. Polarization measurements show the
signature of synchrotron radiation in a Keplarian accretion disk
\citep{2002ApJ...566L..77L}. Taken together, these multi-wavelength
data are not easily described by a one-component model, and the
current theoretical framework combines thermal emission from a
radiatively inefficient Keplarian accretion flow with synchrotron
inverse-Compton emission produced either by electrons accelerated in
the disk or further out in a hypothetical jet-like outflow
(e.g. \citet{2002ApJ...566L..77L,2002A&A...383..854Y,2003astro.ph..4125Y}).
From the present measurements, the maximum energy of the non-thermal
electron distribution in the jet models is ambiguous and theories
alternately explain the high-energy emission as inverse-Compton or
the high-energy extension of the synchrotron spectrum; gamma-ray
measurements may eventually break this degeneracy.
The EGRET experiment detected a strong unidentified source of GeV
gamma-rays marginally consistent with the position of the Galactic
Center \citep{1999ApJS..123...79H}. Both the Whipple and Cangaroo
groups have presented preliminary evidence for TeV emission at the
position of Sgr A* as well \citep{buckley97, cangaroo2003,
kosack2003}. \citet{dingus2002} re-analyzed the higher energy
gamma-ray data from EGRET and found that the most likely position of
the EGRET source may be offset from Sgr A*. However, systematic
uncertainties in the gamma-ray background models and limited angular
resolution make the analysis of the source in the Galactic Center
region difficult. Observations of the Galactic Center are
complicated since Sgr A* is surrounded by a dense cluster of stars
and stellar remnants (including low-mass X-ray binaries and black
hole candidates), molecular clouds, and a large structure that may
be the remnant of a powerful supernova remnant, Sgr A East
\citep{2003ApJ...596.1035F}. Source confusion is particularly
difficult for high-energy gamma-ray observations given the limited
angular resolution of present experiments.
% The Galactic Center region is known to have dense molecular clouds
% and relatively hard diffuse gamma-ray emission
% \citep{1997ApJ...481..205H}, making it difficult to separate
% unambiguously the Galactic Center from the diffuse
% background. Moreover, source confusion from any of a number of point
% sources in the Galactic Center region cannot be ruled out.
High energy gamma-ray observations of the Galactic Center are also
the subject of particular theoretical interest given the possibility
of detecting halo dark matter in our galaxy
(e.g. \citet{1998APh.....9..137B}). Sgr A*, at the dynamical center
of our galaxy, may well be surrounded by a cusp or spike in the dark
matter halo distribution (e.g. \citet{1991ApJ...378..496D,
1996ApJ...462..563N,1999PhRvL..83.1719G, 2003astro.ph.11594M}).
Annihilation of these hypothetical weakly interacting massive
particles could also contribute to the luminosity in the vicinity of
Sgr A* in the radio through gamma-ray waveband. Annihilation of
dark matter would be enhanced by a factor proportional to the
density squared, and might result in an observable gamma-ray line
(from direct annihilation to gamma rays) as well as continuum
emission (from secondary products of annihilation to quarks and
fermions) \citep{1987ApJ...313L..47S, 1989pafe.conf..255B,
1989PhRvD..39.3549R,
1989PhRvD..40.2549G,1989ApJ...343..169S,1995PhRvD..51.3121J}. The
presence of a massive black hole could further steepen the density
profile of the dark matter halo, producing very high radio and
gamma-ray fluxes that exceed the observational upper bounds
\citep{1999PhRvL..83.1719G}. The details of the halo model on scales
$<100\mathrm{pc}$ and the formation history of the central black
hole are critical to predicting the gamma-ray flux, but are,
unfortunately, still poorly understood.
Given the limited angular resolution of GeV and TeV instruments, a
number of different sources could contribute to a signal near
the Galactic Center. The key to distinguishing between all of the
possible emission scenarios is to measure the position, angular
extent, variability and spectrum of the gamma-ray signal. Here we
present first results from an analysis of Whipple telescope data. In
\S\ref{sec:method} we describe the observational method and data
analysis procedure used to observe the Galactic Center at TeV
energies. In \S\ref{sec:analysis} and \S\ref{sec:results}, we
discuss a possible weak detection and consider its impact on various
gamma-ray production scenarios in \S\ref{sec:discussion}.
%======================================================================
\section{Method}
\label{sec:method}
%======================================================================
\begin{table}
\begin{center}
\begin{tabular}[t]{c c c c}
\hline\hline
Season(s) & F.O.V ($^\circ$)& Pixels & Pixel Diameter($^\circ$) \\
\hline
1995-1996 & 3.1 & 109 & 0.26 \\
1996-1997 & 3.1 & 151 & 0.26 \\
1999-2003 & 2.4 & 379 & 0.12\\
\hline\hline
\end{tabular}
\end{center}
\caption{\label{tab:cameras} Camera specifications during seasons
where Galactic Center data were taken.}
\end{table}
Imaging Atmospheric \Cerenkov Telescopes (IACTs), like the Whipple
Gamma-ray Observatory, detect high energy photons by imaging the
flashes of \Cerenkov light emitted by secondary particles in
gamma-ray induced air showers. The Whipple Telescope's 10m mirror
focuses the faint UV/blue \Cerenkov flashes into a camera consisting
of 379 photomultiplier tube pixels (for the latest seasons---see
Table \ref{tab:cameras} for previous seasons). Off-line software
analysis characterizes each candidate shower image, separates signal
(gamma-ray-like) from background (cosmic-ray-like) events, and
determines the point of origin and energy of each gamma ray.
Whipple gamma-ray data are traditionally taken as a series of 28
minute exposures, each of which is followed by an off-source run
which is offset 30 minutes in right ascension for background
subtraction. In the case of Sgr A*, data were taken off-source before
the on-source observations due to a bright star field in the region
30 minutes past the Galactic Center's position. The analysis method
used here (see \S\ref{sec:analysis}) was modified to give sensitivity at
large zenith angles and to provide a two-dimensional map of the TeV
emission in a three-degree diameter field-of-view surrounding the
Galactic Center.
%======================================================================
\section{Data Analysis}
\label{sec:analysis}
%======================================================================
The Galactic Center ($\alpha=17^{h}45^{m}40^{s},\
\delta=-29^{\circ}00'28''$, J2000) transits at a very large zenith
angle ($61^\circ$) as seen from the Whipple Observatory
($31.7^\circ$ N latitude) which significantly alters the shower
geometry and threshold energy. To properly account for the effects
of LZA observations, special techniques that go beyond the standard
Whipple analysis were required. Furthermore, the brightness of the
galactic plane near the Galactic Center results in an increase in
energy threshold and, if not compensated for, a systematic bias in
the observed excess. Pedestal events, containing no image, are
injected at random intervals throughout the run for calibration of
Poisson fluctuations in the night sky background. Both on and
off-source data are analyzed in the same manner, and Gaussian
deviates are added to the pixel signals to bring the background
noise up to the same level in both runs. After this procedure, only
pixels with signals well above the noise level are included in
further image processing. This Gaussian \emph{padding} combined with
a high software trigger threshold largely removes systematic
biases arising from brightness differences, but increases the energy
threshold.
Using the techniques based on the moment fitting procedure outlined
in \citet{1993ApJ...404..206R}, we parameterize the roughly
elliptical gamma-ray images by calculating moments of the light
distribution in the camera. Geometric selection criteria based on
these parameters allow for the rejection of background
(e.g. cosmic-ray induced showers). The first moments give the
centroid of the image, the second moments give the WIDTH (minor
axis) and LENGTH (major axis) and orientation angle of the image.
The elongation of the ellipse is used to determine the point of
origin of each shower image using the formula $ \delta = \epsilon
\left( 1-\frac{\mathrm{WIDTH}}{\mathrm{LENGTH}} \right)$, where
$\delta$ is the displacement of the point of origin from the image
centroid and $\epsilon$ is the elongation factor (determined by
simulations). Of the two possible points of origin for each image,
the asymmetry (or skew) of the shower is used to select the correct
one whenever possible. This procedure is similar to that described
in \citet{2001APh....15....1L}.
To determine the pointing error in the telescope, we look at the
pedestal variation for each tube. The presence of visible light
from a star or other source of sky brightness in the field of view
adds to the pedestal variance in the corresponding photomultiplier
tube. Using this effect, we can generate a crude optical image of
the sky by de-rotating the camera to a common orientation (since the
field of view rotates with time for an altitude-azimuth
telescope) and accumulating the pedestal variations of each pixel
into a two-dimensional histogram. Using this technique, an optical
\emph{sky-brightness} map is generated for each observation. By
comparing the bright spots in the Sgr A* image (or special runs
where the telescope is pointed at a nearby \emph{pointing-check}
star) we can obtain an absolute measure of the pointing error. In
addition, we cross-correlate each pair of maps to determine the
relative pointing offset between them. We only keep runs that have
the correct star field and have a relative pointing offset (compared
with the other Sgr A* runs) which is less than the diameter of one
pixel. This lessens the possibility of accidentally including an
observation with large pointing errors or which was mislabeled.
Using this method, we find residual pointing errors of $\pm
0.1^\circ$ and an absolute offset of $0.14^\circ$. These errors are
attributed in part to the fact that observations have been made near
the balance point of the telescope and near the horizontal position
where flexure of the optical support structure is maximal.
Sgr A* transits at roughly $30^\circ$ elevation as observed from
the latitude of the Whipple Observatory. Large zenith angle (LZA)
observations require several modifications to the standard analysis
due to changes in shower geometry. In addition, we have data on Sgr
A* that spans several epochs during which the Whipple camera was
upgraded twice (see Table \ref{tab:cameras}). In order to combine
all of these data, we needed our gamma-ray selection criteria to
scale properly with changes in zenith angle, the camera throughput
factor (i.e. the ratio of photons hitting the mirror to digital
counts in the digitized PMT signals), and the pixel
size. In addition, for subsequent spectral analysis, we designed our
gamma-ray selection procedure to scale with energy to minimize
spectral biases.
To achieve this goal, we started with the standard Whipple Telescope
data analysis procedure \citep{2001APh....15....1L} and developed a new set
of gamma-ray selection criteria that scale with zenith angle and
energy according to a semi-empirical form derived from simulations
and optimization of LZA Crab Nebula data. Our criteria also
incorporate the geometric effects of differing cameras, allowing us
easily to combine all of the data present for the Galactic Center.
The Crab Nebula was used for optimization and calibration
because it is a bright, steady gamma-ray source with a known
spectrum. In addition, an independent set of Crab Nebula
observations taken at LZA were used to verify that the selection
criteria were more efficient than the standard analysis procedure.
To correct for changes in the overall light sensitivity (throughput) of the
camera between epochs, we look at muon events in data taken at each
epoch. Muons, which show up as bright arcs of \Cerenkov light in the
camera, are useful for throughput calibration because the light per
unit arc-length from a muon event is nearly constant regardless
of the impact parameter and angle of the trajectory. A measure of
the throughput of the telescope can be found by making a histogram
of the \emph{signal/arc-length} distribution of a large set of muon
events. To get an absolute calibration, this distribution for real
observations was compared with a set of simulated muon events which
were generated using the Grinnell/ISU (GrISU) simulation package to produce a
large number of H and He showers that in turn produce muons as
secondary particles. For instance, the relative throughput was
found to have changed by a factor of 2.22 between 1995 and 2001. We
scale the software trigger threshold and energy estimator by this
factor so the trigger cuts are consistent across observations. This
method also serves to calibrate the simulations used to determine
the peak energy at LZA (see \S\ref{sec:results}).
As one observes at increasing zenith angles, the distance to
the core of the air-shower increases and thus the angular size of the shower
and parallactic displacement of the image centroid are reduced. To
derive the scaling laws, we first assume that the WIDTH and LENGTH
of gamma-ray air shower images is approximately proportional to
$\cos^\alpha\theta$, where $\theta$ is the
zenith angle and $\alpha$ is a constant. Additionally, air-shower
simulations show that LENGTH and WIDTH also scale as the logarithm
of the energy, which is proportional to the SIZE (total camera
signal) of the event.
Combining these results, and removing the effects of the finite
pixel size of the camera ($\sigma_{\mathrm{pix}}$) and the point
spread function of the telescope ($\sigma_{\mathrm{psf}}$), the
measured LENGTH ($L$), WIDTH ($W$), and the distance to the image
centroid ($D$) can be converted to scaled values $L'$, $W'$ and $D'$
by the following equations:
\begin{eqnarray}
\Sigma \equiv \ln(\mathrm{SIZE}) - 8.0 \nonumber \\
\mathrm{L}' \simeq
\left[ \frac{ \mathrm{L}^2
- \sigma^2_{\mathrm{pix}}
- \sigma^2_{\mathrm{psf}}
}{\cos^{1.5}\theta}
\right]^\frac{1}{2} - 0.023\Sigma \\
\mathrm{W}' \simeq
\left[ \frac{
\mathrm{W}^2
- \sigma^2_{\mathrm{pix}}
- \sigma^2_{\mathrm{psf}}
}{\cos^{1.2}\theta}
\right]^\frac{1}{2} - 0.020\Sigma \\
\mathrm{D}' \simeq
\frac{\mathrm{D}}{\cos\theta}
\end{eqnarray}
The constant factors and cosine powers were derived from simulation
fits and by optimization on Crab Nebula data (taken at a range of
zenith angles) and then compared with simulations for calibration. Data
selection criteria $0.125^\circ < L'< 0.3^\circ$,
$0.05^\circ<W'<0.135^\circ$, $0.28^\circ<D'<2.2^\circ$ are applied
to select candidate gamma-ray events (examples of these cuts are
shown graphically in Figure \ref{fig:zcuts}). Cuts based on these
intrinsic parameters were verified to be independent of zenith angle
and camera design by application of this method to independent Crab
Nebula data taken over the period 1994-2003.
\begin{figure}[ht!]
\includegraphics[width=\columnwidth]{zcuts_vs_len_wid}
\caption{ \label{fig:zcuts} Plots of our selection criteria for
gamma-ray-like events for two of the parameters (LENGTH and
WIDTH) for two different zenith angles ($0^\circ$ and $60^\circ$).
The dotted lines show the upper and lower cuts on the respective
parameters as a function of the SIZE (total signal) parameter. The
dots are from simulations of gamma rays with a range of energies. }
\end{figure}
\section{Results}
\label{sec:results}
\begin{figure}[ht!]
\includegraphics[width=\columnwidth]{pointing-check}
\caption{ \label{fig:offset} Optical image of a star
($\mathrm{Sgr}\ \gamma_2$) used to check the telescope's
pointing at low elevation. This image shows that the telescope
has an offset of 0.14 degrees down and to the right of camera
center position (0,0).}
\end{figure}
\begin{figure}[ht!]
\includegraphics[width=\columnwidth]{sgra-2d}
\caption{ \label{fig:sgra} A gamma-ray image of the region around
Sgr A*. The image is of excess counts with overlaid significance
contours (1 standard deviation per contour). The axes are labeled in
degrees from the assumed camera center. The true center position
of the camera, which is not exactly at (0,0) due to flexing of
the telescope at low elevation, is marked with a cross. The
dashed lines are the RA and Dec contours at this position. Also
shown (as a light contour) is the 99\% confidence region for the
EGRET observations \citep{dingus2002}.}
\end{figure}
\begin{figure}[ht!]
\includegraphics[width=\columnwidth]{lzacrab}
\caption{ \label{fig:lzacrab} A gamma-ray image of the Crab Nebula
taken at large zenith angle ($\approx62^\circ$) using the same analysis
procedure used for Sgr A*. The offset and pointing variations can be
seen in the resulting image.}
\end{figure}
We have combined observations of Sgr A* from 1995 through 2003
resulting in 26 hours of on-source exposure at an average zenith
angle of $61^\circ$. To determine the pointing offset, observations
were taken centered on a nearby bright star ($\mathrm{Sgr}\
\gamma_2$) which is at the same elevation as Sgr A*. Using the sky
brightness map technique outlined earlier, we determined that the
telescope had a pointing offset of $0.14^\circ$, as shown in Figure
\ref{fig:offset}. Figure \ref{fig:sgra} shows the resulting 2-D map
of gamma-ray excess with overlaid significance contours. The true
center of the camera, correcting for the offset, is plotted as a
cross in the image. This image shows a $4.2$ standard deviation
($\sigma$) excess at the corrected center position. To check the
robustness of this result, we re-ran the analysis ten times to
account for variations due to the Gaussian padding. We find the
average significance at the corrected center position is $(3.7 \pm
0.13)\ \sigma$, somewhat below the initial result. For reference, in
Figure \ref{fig:lzacrab} we show the results of the same analysis
procedure applied to 10 hours of observations of the Crab Nebula at
a similar zenith angle range. Note that the significance of
$6.1\sigma$ of the Crab detection at the offset position is
substantially higher than the result of $3.8\sigma$ obtained
applying the standard small zenith angle analysis procedure to these
LZA data. Also, the similar angular extent in the two results
indicates consistency with a point source within a 99\% confidence
region of radius $\approx 15$ arcmin. A more detailed analysis of
the angular extent would require a prior hypothesis about the nature
of the source.
% ---------------------------------------------------
% Flux
% ---------------------------------------------------
% \begin{figure}[ht!]
% \includegraphics[angle=270,width=\columnwidth]{lza_crab_sim_spectrum}
% \caption{ \label{fig:crabspect} Simulated Crab Nebula spectrum at
% large zenith angle ($\theta=61^\circ$). These simulated data were run
% through the same data analysis software as the real Sgr A* data to
% determine when the rate peaks first, which is at
% $\approx 5\ \mathrm{TeV}$.}
% \end{figure}
To determine the peak energy of the detected flux from Sgr A*, we
simulated gamma rays with a Crab Nebula spectrum (with integral
spectral index $\gamma=1.58$)
\citep{1998APh.....9...15M,2000ApJ...539..317A} and a zenith angle
of $61^\circ$, and analyzed the resulting data with a detector
simulation and our analysis software. We determined the peak
detected energy to be $\approx 2.8\ \mathrm{TeV}$. Noise padding
and a higher trigger threshold makes this result slightly higher
than has been previously reported \citep{1999ApJ...511..149K}, and
we estimate a $20\%$ systematic error in this energy threshold. We
then analyzed a set of real LZA Crab Nebula data runs to find the
Crab count rate and compared this to the corresponding rate for Sgr
A*. The integral flux for Sgr A*, normalized to the Crab, is then:
\begin{equation}
F_{SgrA}(>2.8\ \mathrm{TeV}) = N_{0,\mathrm{Crab}} \cdot
\frac{(2.8\ \mathrm{TeV})^{-\gamma}}{\gamma} \cdot
\frac{R_{\mathrm{SgrA*}}}{R_{\mathrm{Crab}}}
\end{equation}
Where $N_{0,Crab}$ is the flux normalization factor for the Crab
Nebula ($3.12\times10^{-7} \mathrm{m^{-2}
s^{-1}}$), $\gamma$
is the integral Crab spectral index, and $R_{\mathrm{SgrA*}}$ and
$R_{\mathrm{Crab}}$ are the corresponding Sgr A* and Crab Nebula
gamma-ray count rates. From the LZA Crab data, we find a gamma-ray rate of
$R_{\mathrm{Crab}}(>2.8\ \mathrm{TeV})=0.501 \pm
0.087~\mathrm{photons\ min^{-1}}$ and from Sgr A* we obtain an
average rate of $R_{\mathrm{SgrA*}}(>2.8\ \mathrm{TeV})=0.205 \pm
0.057\ \mathrm{photons\ min^{-1}}$. Hence, the gamma-ray
flux from the Galactic Center region above 2.8 TeV is \Flux, or
about 0.4 times that of the Crab Nebula (the flux error includes the
uncertainty in the Crab Nebula measurement).
% ---------------------------------------------------
% Variability
% ---------------------------------------------------
\begin{figure}[ht!]
\includegraphics[angle=90,width=\columnwidth]{gc_lc}
\caption{ \label{fig:gcvar}
Flux of Sgr A* as a function of time. Each data point
represents a single 28 minute run. Time gaps in the data have
been removed where indicated. The dashed line is a least squares fit
of a constant function to the data. }
\end{figure}
The count rate from Sgr A* is shown as a function of time in
Figure \ref{fig:gcvar}. To determine the probability for steady
emission, a $\chi^2$ fit of a constant function ($f(t) = A$) was
applied to this data and, for comparison, to a series of data taken
of Mrk 421 (a source which is known to be highly variable) at a
similar zenith angle range as Sgr A*. The total significance of this
Mrk421 data sample was $2.3\sigma$. The Sgr A* data yields a
constant count rate of $6.12 \pm 1.59 \ \gamma\cdot\mathrm{min}^{-1}$ with
a reduced $\chi^2$ of $1.13$ (with 54 degrees of freedom), which
corresponds to a $25\%$ probability that there is no
variability. The result for Mrk~421 yields a constant count rate of $6.86
\pm 6.13 \ \gamma\cdot\mathrm{min}^{-1}$ with a reduced $\chi^2$ of
3.03 (with 6 degrees of freedom) and a $1.2\%$ chance of no
variability.
%======================================================================
\section{Discussion}
\label{sec:discussion}
%======================================================================
The TeV excess observed near the position of the Galactic Center is
unlikely to have occurred by chance and constitutes a probable, as
yet unconfirmed, detection of a new TeV source. Possible
systematics that could contribute to a false detection include the
effects of additional noise from the relatively bright off-source
region. While we have largely corrected for these effects, some
systematic uncertainties remain. We have taken into account trials
factors by formulating an explicit \emph{a priori} hypothesis that
we would only look for emission at the exact position of the
Galactic Center after a pointing correction was applied.
Statistical variations in the analysis method (due to the addition
of simulated noise in padding) have been taken into account by
repeating the analysis ten times, and taking the average
significance, giving a conservative estimate of $3.7\sigma$ for the
detection significance.
The lack of significant variability in our data makes it difficult
to uniquely identify the source with a compact point source such as
Sgr A*, but inspires some confidence in the stability of our
observations at large zenith angle. Note that the analysis procedure
was designed to mitigate against changes in the count rate due to
variations in the instrument. The same ISU simulation package
was used here to analyze Whipple observations of the Crab Nebula
giving a spectrum in good agreement which that measured a small
zenith angle \citep{1999ApJ...511..149K}. In the past, our group
reported a positive excess of $2.4\sigma$ for 1995-1997 observations
\citep{buckley97} and $2.4\sigma$ for 1999-2003 observations
\citep{kosack2003} at the position of Sgr A*. The combined
significance for our refined results is consistent with these
earlier analyses. The large error circles for both EGRET (7.2 arcmin) and
Whipple (15 arcmin) observations make identification with a
particular source difficult, but given the dearth of TeV sources, an
accidental angular coincidence of a new source along the line of
sight is unlikely, and it is probable that the emission comes from a
non-thermal source physically near the Galactic Center.
The high level of emission $\approx 0.4\ \mathrm{Crab}$ at a
distance of roughly four times that of the Crab Nebula, qualifies
this as an unusually luminous galactic source. Previous TeV
observations of relatively nearby galactic sources such as X-ray
binaries and shell-type and plerionic supernovae have produced
numerous upper-limits, or (at best) unconfirmed detections, making
the detection of such an object at 8.5kpc even more unlikely.
\citet{1998A&A...335..161M} came to a similar conclusion about the
GeV emission based on the high luminosity of the EGRET unidentified
source and lack of significant variability. If the Sgr A East
supernova shock were the source of the EGRET gamma rays, it would
have been an unusually intense explosion \citep{1996ApJ...457L..61K}
and a density of $1000\ \mathrm{cm^{-3}}$ and magnetic field of $B
\sim 0.18 \mathrm{mG}$ (well above the canonical values) would be
required \citep{2003ApJ...596.1035F}. While a typical galactic
source such as an SNR, pulsar, or stellar mass black hole is
unlikely, an active nucleus at our Galactic Center is still a viable
possibility and the detection of correlated variability in future
gamma-ray and X-ray observations is required to make the
identification with Sgr A* compelling. If we associate this
emission with either the super-massive black hole Sgr A* or
supernova remnant Sgr A East, the observed emission could come from
self-Compton scattering by electrons with energies up to at least
2.8 TeV or from pion-decay gamma-rays from primary protons of even
higher energy (kinematics require their energy to be at least
several times the maximum gamma-ray energy).
The lack of significant variability and the consistency with the
Galactic Center position allow more exotic possibilities such as the
annihilation of very high mass ($> 2\ \mathrm{TeV}$) dark matter
particles at the Galactic Center. While not particularly
constraining, these results and a more detailed analysis of the
spectrum will be used to derive upper limits for dark matter
annihilation in a subsequent paper.
In summary, based on 26 hours of data taken with the Whipple 10m
gamma-ray observatory, we report a probable detection of a gamma-ray
point source consistent with the position of the Galactic Center.
We also describe a modified analysis procedure that we have
developed to analyze gamma-ray data taken by Imaging Atmospheric
\Cerenkov telescopes operating at large zenith angles. The continued
observation of TeV gamma rays from the Galactic Center has important
theoretical implications for understanding a variety of
astrophysical phenomena in the nuclear region of the Milky Way.
\paragraph{Acknowledgements}
{\small
We would like to thank Fluvio Melia, Paolo Gondolo, Johnathan Katz,
and Ramanath Cowsik for useful discussion.
The VERITAS Collaboration is supported by the U.S. Dept. of Energy,
N.S.F., the Smithsonian Institution, P.P.A.R.C. (U.K.) and
Enterprise-Ireland.}
\begin{thebibliography}{}
\bibitem[Aharonian et al.(2000)]{2000ApJ...539..317A} Aharonian, F.~A., et
al.\ 2000, \apj, 539, 317
\bibitem[Baganoff et al.(2001)]{2001Natur.413...45B} Baganoff,
F.~K.~et al.\ 2001, \nat, 413, 45
\bibitem[Baganoff et al.(2003)]{2003ApJ...591..891B} Baganoff, F.~K.~et
al.\ 2003, \apj, 591, 891
\bibitem[Balick \& Brown (1974)]{balick1974} Balick, B. and Brown,
R. L. 1974, \apj, 194, 265
\bibitem[B{\' e}langer et al.(2004)]{2004ApJ...601L.163B} B{\' e}langer,
G.~et al.\ 2004, \apjl, 601, L163
\bibitem[Bergstrom(1989)]{1989pafe.conf..255B} Bergstrom, L.\ 1989,
Particle Astrophysics: Forefront Experimental Issues, 255
\bibitem[Bergstr{\" o}m, Ullio, \&
Buckley(1998)]{1998APh.....9..137B} Bergstr{\" o}m, L., Ullio, P.,
\& Buckley, J.~H.\ 1998, Astroparticle Physics, 9, 137
\bibitem[Buckley et. al. (1997)]{buckley97} Buckley, J. H. et al.\
1997, Proceedings of 25th ICRC (Durban) 3, 237.
\bibitem[Dubinski \& Carlberg(1991)]{1991ApJ...378..496D} Dubinski, J.~\&
Carlberg, R.~G.\ 1991, \apj, 378, 496
\bibitem[Fatuzzo \& Melia(2003)]{2003ApJ...596.1035F} Fatuzzo,
M.~\&Melia, F.\ 2003, \apj
\bibitem[Ghez et al.(2002)]{2002AAS...201.6804G} Ghez, A.~M.~et al.\
2002, Bulletin of the American Astronomical Society, 34, 1219
\bibitem[Ghez et al.(2004)]{2004ApJ...601L.159G} Ghez, A.~M.~et al.\ 2004,
\apjl, 601, L159
\bibitem[Giudice \& Griest(1989)]{1989PhRvD..40.2549G} Giudice, G.~F.~\&
Griest, K.\ 1989, \prd, 40, 2549
\bibitem[Gondolo \& Silk(1999)]{1999PhRvL..83.1719G} Gondolo, P.~\&
Silk, J.\ 1999, Physical Review Letters, 83, 1719
\bibitem[Hartman et al.(1999)]{1999ApJS..123...79H} Hartman,
R.~C.~et al.\ 1999, \apjs, 123, 79
\bibitem[Hillas (1985)]{hillas} Hillas, A.M.\ 1995, Proceedings of
19th ICRC (La Jolla), vol 3
\bibitem[Hooper \& Dingus (2002)]{dingus2002} Hooper, D. and
B. Dingus,\ 2002, ArXiv Astrophysics e-prints, astro-ph/0212509.
\bibitem[Hunter et al.(1997)]{1997ApJ...481..205H} Hunter, S.~D.~et al.\
1997, \apj, 481, 205
\bibitem[Jungman \& Kamionkowski(1995)]{1995PhRvD..51.3121J} Jungman, G.~\&
Kamionkowski, M.\ 1995, \prd, 51, 3121
\bibitem[Khokhlov \& Melia(1996)]{1996ApJ...457L..61K} Khokhlov,
A.~\& Melia, F.\ 1996, \apjl, 457, L61
\bibitem[Kosack et. al. (2001)]{kosack2001} Kosack, K. et al.\ 2001,
Proceedings of 27th ICRC (Hamburg).
\bibitem[Kosack et. al. (2003)]{kosack2003} Kosack, K. et al.\ 2003,
Proceedings of 28th ICRC (Tsukuba).
\bibitem[Krennrich et al.(1999)]{1999ApJ...511..149K} Krennrich,
F.~etal.\ 1999, \apj, 511, 149
\bibitem[Lessard et al.(2001)]{2001APh....15....1L} Lessard, R.~W.,
Buckley, J.~H.,Connaughton, V., \& Le Bohec, S.\ 2001, Astroparticle
Physics, 15, 1
\bibitem[Liu \& Melia(2002)]{2002ApJ...566L..77L} Liu, S.~\& Melia, F.\
2002, \apjl, 566, L77
\bibitem[Mayer-Hasselwander et al.(1998)]{1998A&A...335..161M}
Mayer-Hasselwander, H.~A.~et al.\ 1998, \aap, 335, 161
\bibitem[Merritt(2003)]{2003astro.ph.11594M} Merritt, D.\ 2003,
ArXiv Astrophysics e-prints, astro-ph/0311594
\bibitem[Mohanty et al.(1998)]{1998APh.....9...15M} Mohanty, G., et al.\
1998, Astroparticle Physics, 9, 15
\bibitem[Navarro, Frenk, \& White(1996)]{1996ApJ...462..563N} Navarro,
J.~F., Frenk, C.~S., \& White, S.~D.~M.\ 1996, \apj, 462, 563
\bibitem[Reynolds et al.(1993)]{1993ApJ...404..206R} Reynolds, P.~T.~et
al.\ 1993, \apj, 404, 206
\bibitem[Rudaz(1989)]{1989PhRvD..39.3549R} Rudaz, S.\ 1989, \prd, 39, 3549
\bibitem[Sch{\" o}del et al.(2002)]{2002Natur.419..694S} Sch{\"
o}del, R.~et al.\ 2002, \nat, 419, 694
\bibitem[Silk \& Bloemen(1987)]{1987ApJ...313L..47S} Silk, J.~\& Bloemen,
H.\ 1987, \apjl, 313, L47
\bibitem[Stecker \& Tylka(1989)]{1989ApJ...343..169S} Stecker, F.~W.~\&
Tylka, A.~J.\ 1989, \apj, 343, 169
\bibitem[Tanimori et al.(1994)]{1994ApJ...429L..61T} Tanimori,
T.~etal.\ 1994, \apjl, 429, L61
\bibitem[Tsuchiya et al. (2003)]{cangaroo2003} Tsuchiya, K. et
al. 2003. Proceedings 28th ICRC (Tsukuba).
\bibitem[Yuan, Markoff, \& Falcke(2002)]{2002A&A...383..854Y} Yuan,
F., Markoff, S., \& Falcke, H.\ 2002, \aap, 383, 854
\bibitem[Yuan, Quataert, \& Narayan(2003)]{2003astro.ph..4125Y}
Yuan, F., Quataert, E., \& Narayan, R.\ 2003, ArXiv Astrophysics
e-prints, astro-ph/0304125
\end{thebibliography}
\end{document}