1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass{emulateapj}
3: % \documentclass[manuscript]{aastex}
4: %% preprint2 produces a double-column, single-spaced document:
5: % \documentclass[preprint2]{aastex}
6:
7: \newcommand{\myemail}{enomoto@icrr.u-tokyo.ac.jp}
8: \newcommand{\order[1]}{$~\times 10^{#1}$}
9:
10: \slugcomment{To appear in ApJ December 20, 2007, v671n2 issue.}
11:
12: \shorttitle{Search for Gamma-rays from SN 1987A}
13: \shortauthors{Enomoto et al.}
14:
15: \begin{document}
16:
17: \title{CANGAROO-III Search for Gamma Rays from SN 1987A
18: and the Surrounding Field}
19:
20: \author{
21: R.~Enomoto\altaffilmark{1}
22: G.~V.~Bicknell\altaffilmark{2}
23: R.~W.~Clay\altaffilmark{3}
24: P.~G.~Edwards\altaffilmark{4}
25: S.~Gunji\altaffilmark{5}
26: S.~Hara\altaffilmark{6}
27: T.~Hattori\altaffilmark{7}
28: S.~Hayashi\altaffilmark{8}
29: Y.~Higashi\altaffilmark{9}
30: Y.~Hirai\altaffilmark{10}
31: K.~Inoue\altaffilmark{5}
32: S.~Kabuki\altaffilmark{9}
33: F.~Kajino\altaffilmark{8}
34: H.~Katagiri\altaffilmark{11}
35: A.~Kawachi\altaffilmark{7}
36: T.~Kifune\altaffilmark{1}
37: R.~Kiuchi\altaffilmark{1}
38: H.~Kubo\altaffilmark{9}
39: J.~Kushida\altaffilmark{7}
40: T.~Mizukami\altaffilmark{9}
41: R.~Mizuniwa\altaffilmark{7}
42: M.~Mori\altaffilmark{1}
43: H.~Muraishi\altaffilmark{12}
44: T.~Naito\altaffilmark{13}
45: T.~Nakamori\altaffilmark{9}
46: S.~Nakano\altaffilmark{9}
47: D.~Nishida\altaffilmark{9}
48: K.~Nishijima\altaffilmark{7}
49: M.~Ohishi\altaffilmark{1}
50: Y.~Sakamoto\altaffilmark{7}
51: A.~Seki\altaffilmark{7}
52: V.~Stamatescu\altaffilmark{3}
53: T.~Suzuki\altaffilmark{10}
54: D.~L.~Swaby\altaffilmark{3}
55: T.~Tanimori\altaffilmark{9}
56: G.~Thornton\altaffilmark{3}
57: F.~Tokanai\altaffilmark{5}
58: K.~Tsuchiya\altaffilmark{9}
59: S.~Watanabe\altaffilmark{9}
60: Y.~Yamada\altaffilmark{8}
61: E.~Yamazaki\altaffilmark{7}
62: S.~Yanagita\altaffilmark{10}
63: T.~Yoshida\altaffilmark{10}
64: T.~Yoshikoshi\altaffilmark{1}
65: Y.~Yukawa\altaffilmark{1}
66: }
67:
68:
69: \altaffiltext{1}{ Institute for Cosmic Ray Research, University of Tokyo, Kashiwa, Chiba 277-8582, Japan}
70: \altaffiltext{2}{ Research School of Astronomy and Astrophysics, Australian National University, ACT 2611, Australia}
71: \altaffiltext{3}{ School of Chemistry and Physics, University of Adelaide, SA 5005, Australia}
72: \altaffiltext{4}{ Paul Wild Observatory, CSIRO Australia Telescope National Facility, CSIRO,
73: Narrabri, NSW 2390, Australia}
74: \altaffiltext{5}{ Department of Physics, Yamagata University, Yamagata, Yamagata 990-8560, Japan}
75: \altaffiltext{6}{ Ibaraki Prefectural University of Health Sciences, Ami, Ibaraki 300-0394, Japan}
76: \altaffiltext{7}{ Department of Physics, Tokai University, Hiratsuka, Kanagawa 259-1292, Japan}
77: \altaffiltext{8}{ Department of Physics, Konan University, Kobe, Hyogo 658-8501, Japan}
78: \altaffiltext{9}{ Department of Physics, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan}
79: \altaffiltext{10}{ Faculty of Science, Ibaraki University, Mito, Ibaraki 310-8512, Japan}
80: \altaffiltext{11}{ Department of Physical Science, Hiroshima University, Higashi-Hiroshima, Hiroshima 739-8526, Japan}
81: \altaffiltext{12}{ School of Allied Health Sciences, Kitasato University, Sagamihara, Kanagawa 228-8555, Japan}
82: \altaffiltext{13}{ Faculty of Management Information, Yamanashi Gakuin University, Kofu, Yamanashi 400-8575, Japan}
83:
84: \begin{abstract}
85: Optical images of SN 1987A show a triple ring structure.
86: The inner (dust) ring has recently increased in brightness and
87: in the number of hot spots suggesting that the supernova shock wave
88: has collided with the dense pre-existing circumstellar medium,
89: a scenario supported by radio and X-ray observations.
90: Such a shocked environment is widely expected to result in
91: the acceleration of charged particles, and the accompanying
92: emission of very high energy gamma-rays.
93: Here, we report the results of observations made in 2004 and 2006
94: which yield upper limits on the TeV gamma-ray flux, which
95: are compared with a theoretical prediction.
96: %derived from whose model
97: In addition, we set upper
98: limits on the TeV flux for four high energy objects which are
99: located within the
100: same field of view of the observation:
101: the super-bubble 30~Dor~C, the Crab-like pulsar PSR~B0540$-$69,
102: the X-ray binary LMC X-1, and the supernova remnant N157B.
103: \end{abstract}
104:
105:
106: \keywords{gamma rays: search --- supernova: individual (SN 1987A) }
107:
108: \section{Introduction}
109:
110: SN1987A was the closest observed supernova
111: in 380 years, and the evolution of the remnant of this supernova
112: has been studied in great detail over the past twenty years.
113: The radio intensity is growing and its rate of increase is
114: increasing, with the spectral index being observed to flatten
115: \citep{atnf}.
116: Imaging at radio wavelengths has enabled the expansion of the inner ring
117: to be traced, and is revealing increasing structure in the inner ring.
118: The X-ray fluxes observed with XMM and Chandra also
119: continue to rise almost exponentially \citep{xmm,chandra}.
120: Around $\sim$4000 days after the supernova the X-ray
121: light curve increased dramatically, attributed to the
122: arrival of the supernova blast wave at the equatorial ring of
123: circumstellar material. There is some evidence that the X-ray flux
124: is mainly thermal, indicating the blast-wave interaction
125: with dense matter \citep{chandra}.
126:
127: \citet{berezhko} compared radio
128: and X-ray data with their calculations and concluded that there
129: is a high (amplified) magnetic field inside the supernova remnant.
130: The predicted TeV gamma-ray flux, resulting mostly from the
131: decay of neutral pions produced in interactions of the
132: accelerated cosmic rays, is almost one order below the level
133: obtained in previous searches by
134: Imaging Atmospheric Cherenkov Telescopes (IACTs) such as
135: CANGAROO-II in 2001 \citep{enomoto2003}
136: and H.E.S.S.\ in 2003 \citep{rowell}.
137: The wide field of view searched by H.E.S.S.\ did, however,
138: yield marginal excesses in a region north-east of SN 1987A and at LMC X-1.
139:
140: CANGAROO-III is one of two stereoscopic
141: IACTs located in the southern hemisphere. With a three-fold
142: coincidence, the sensitivity is significantly improved from the previous
143: single telescope (CANGAROO-II).
144: We also increased the observation period by a factor of two
145: in the 2006 observations compared to the 2004 and 2001 seasons.
146: We here report the results of these observations.
147:
148: The previous H.E.S.S.\ analysis considered not only SN 1987A but also
149: four other high energy objects within the same field of view \citep{rowell}.
150: We follow that idea and present upper limits on the integral
151: gamma-ray fluxes from
152: the super-bubble 30~Dor~C,
153: the Crab-like pulsar PSR B0540$-$69,
154: the X-ray binary LMC X-1,
155: and the supernova remnant N157B.
156: Super-bubbles are expected to be efficient accelerators
157: of cosmic rays \citep{parizot} and
158: a shell feature near the star 30~Dor~C, within 5 arcmin of SN 1987A, shows
159: an indication of non-thermal hard X-ray emission \citep{bamba}.
160: The energy required to produce the bubble is estimated to be
161: 7$\times$10$^{51}$\,erg, which is
162: higher than generally expected for a single supernova remnant \citep{ueno}.
163: The other three sources belong to categories of
164: high-energy astrophysical objects that may
165: accelerate particles to TeV energies, but,
166: considering their large distance of 50\,kpc, they
167: are less likely to be detected
168: by current IACTs than their Galactic counterparts.
169: However, as they are well within the field of view
170: of the CANGAROO-III telescopes, and given the marginal H.E.S.S.\ excess
171: toward LMC X-1, it is straightforward to search
172: for any evidence of gamma-ray emission from them.
173:
174: \section{CANGAROO-III Stereoscopic System}
175:
176: The CANGAROO-III stereoscopic system consists of four imaging atmospheric
177: Cherenkov telescopes located near Woomera, South Australia (31$^\circ$S,
178: 137$^\circ$E).
179: Each telescope has a 10\,m diameter segmented reflector
180: consisting of 114 spherical mirrors
181: each of 80\,cm diameter \citep{kawachi},
182: providing a total light collecting area of 57.3\,m$^2$.
183: The spherical segments are mounted on a parabolic
184: frame with a focal length of 8\,m.
185: The first telescope, T1, which was the CANGAROO-II telescope
186: \citep{enomoto_nature},
187: was not used for these observations due to its smaller field of view
188: and higher energy threshold.
189: The second, third, and fourth telescopes (T2, T3, and T4) were used for the
190: observations described here.
191: The camera systems for T2, T3, and T4 are identical and are described
192: in \citet{kabuki}.
193: The telescopes are located at the
194: east (T1), west (T2), south (T3) and north (T4)
195: corners of a diamond
196: with sides of $\sim$100\,m \citep{enomoto_app}.
197:
198: \section{Observations}
199:
200: The observations were carried out
201: in the periods from 2004 Nov 11 to 14 (MJD 53320--53323)
202: and from 2006 Dec 12 to 27 (MJD 54081--54096)
203: using ``wobble mode"
204: in which the pointing position of each telescope was
205: shifted in declination between $\pm$0.5 degree
206: every 20 minutes \citep{wobble}
207: from the target:
208: (RA, dec [J2000]) = (83.866139$^\circ$, $-$69.269577$^\circ$).
209: We took no OFF source runs.
210: The most sensitive region of the field of view
211: is within one degree of the average pointing position.
212: LMC~X-1 is located 0.6 degrees south-east from SN~1987A.
213: Considering our angular resolution of 0.24 degree, there is
214: some overlap between
215: SN 1987A, 30~Dor~C, PSR B0540$-$69, and N157B.
216:
217: In the 2004 observation,
218: data with GPS time stamps were recorded for T2, T3 and T4 individually when
219: more than four photomultiplier (PMT) signals
220: exceeded 7.6 photoelectrons (p.e.).
221: In the offline analysis stage we combined the data when
222: the three telescope's GPS times coincided.
223: In the 2006 observation, a hardware coincidence was used to select
224: any two triggered telescopes \citep{nishijima}.
225: The images in all three telescopes were required to have clusters
226: of at least five adjacent pixels exceeding a 5\,p.e.\ threshold
227: (offline three-fold coincidence).
228: The event rate was reduced to 5$\sim$8\,Hz by this criterion
229: depending on the elevation angle and year.
230: Based on time dependence of these rates we can remove data
231: taken in cloudy conditions.
232: The effective observation times
233: for 2004 and 2006 were
234: 632 and 1316 min, respectively.
235: The corresponding mean zenith angles were 42.1$^\circ$
236: and 40.0$^\circ$.
237:
238: The light collecting efficiencies, including the reflectivity
239: of the segmented mirrors, the light guides, and the quantum efficiencies
240: of the photomultiplier tubes were monitored by a muon-ring analysis
241: \citep{enomoto_vela} with the individual trigger data in the
242: same periods.
243: The light yield per unit arc-length is approximately proportional
244: to the light collecting efficiencies.
245: Deterioration is mostly due to dirt and dust settling on the
246: mirrors.
247:
248: \section{Analysis}
249:
250: Here, we briefly describe the analysis procedures, which
251: are identical with those described in \citet{cena}.
252: More details can be found in \citet{enomoto_vela} and
253: \citet{enomoto_0852}.
254:
255: First, the image moments \citep{hillas}
256: were calculated for the three telescopes.
257: The incident gamma-ray direction was determined by minimizing
258: the sum of squared widths ($\chi^2_0$: weighted by the photon yield)
259: of the three images seen from the assumed position (fitting parameter)
260: with a constraint on the distances from the intersection point to each
261: image center.
262:
263: In order to derive the gamma-ray likeliness,
264: we used
265: the Fisher Discriminant (hereafter $FD$) \citep{fisher,enomoto_vela}.
266: Input parameters were
267: $$\vec{P}=(W2,W3,W4,L2,L3,L4),$$
268: where $W2,W3,W4,L2,L3,L4$ are energy-corrected $widths$ and $lengths$ for the
269: T2, T3, and T4
270: and assume that a linear combination of
271: $$FD=\vec{\alpha}\cdot\vec{P},$$
272: provides the
273: best separation between signal and background, then the set of
274: linear coefficients ($\vec{\alpha}$) should be uniquely determined as
275: $$\vec{\alpha}=\frac{\vec{\mu}_{sig}-\vec{\mu}_{BG}}{E_{sig}+E_{BG}},$$
276: where $\vec{\mu}$ is a vector of the mean value of $\vec{P}$ for each
277: sample and $E$ is their correlation matrix.
278: We previously used
279: it in Vela Pulsar analysis \citep{enomoto_vela}
280: to separate ``sharp'' (gamma-ray--like) images from ``smeared'' (background)
281: ones.
282: The values of
283: $\vec{\mu}_{sig}$, $\vec{\mu}_{BG}$,
284: $E_{sig}$, and $E_{BG}$ can be calculated from the
285: Monte-Carlo and observational data (OFF-source runs), respectively.
286:
287: We rejected events with any hits in the outermost layer of a camera (``edge
288: cut"). These rejected events suffer from finite deformations, especially in
289: the $length$ distribution, which results in deformations of the $FD$.
290:
291: We then derived $FD$ distributions position by position.
292: Comparing those in the signal region and control background
293: region, we can determine the gamma-ray--like events.
294: Here, we use Monte-Carlo simulations to determine the $FD$
295: distribution of gamma-ray signals.
296: Note that in the gamma-ray simulations we used a spectrum
297: proportional to $E^{-2.1}$.
298: The fit of the $FD$ distribution of source position
299: with the above emulated
300: signal and control background functions were carried out,
301: to derive the number of gamma-ray--like events.
302: This is a one-parameter fitting with the constraint that
303: sum of signal and background events corresponds to the total number of events.
304: These coefficients can be derived exactly analytically.
305:
306: \section{Results}
307:
308: The threshold of this analysis is considered to be $\sim$1\,TeV
309: which is higher than the typical CANGAROO-III threshold due to
310: the larger zenith angle of the observations, of around 40 degrees.
311: In order to derive morphology, we segmented the field of view into
312: 0.2\,$\times$\,0.2 degree$^2$ square bins. The $FD$ distributions
313: for corresponding bins are made and fitted. The control-background
314: region is defined as
315: the second closest layer,
316: %one-next layers
317: which is more than 0.3
318: degree from the center of target region,
319: i.e., larger than the 0.24 degree point spread function (PSF).
320: The statistics of the
321: control-background is, therefore, sixteen times larger than that
322: of signal bin.
323: The results are shown in Figs.~\ref{fig1}.
324: \begin{figure*}[htbp]
325: \includegraphics[width=300pt]{f1a.eps}
326: \hskip -3cm\includegraphics[width=300pt]{f1b.eps}
327: \vskip -1cm
328: \caption{Excess count maps. The left panel is obtained from the 2004 data and
329: the right from the 2006 data. The average pointing center is SN 1987A itself.
330: The positions of four other high energy objects are also
331: indicated by the black
332: crosses with labels.
333: The PSF is shown by the black dashed circle.
334: The white contours are derived from
335: 4.85\,GHz PMN survey radio data \citep{pmns}
336: obtained from Skyview (NASA) \citep{skyview}.}
337: \label{fig1}
338: \end{figure*}
339: The left panel is obtained from the 2004 data and the right from 2006.
340: The average telescope pointing position is indicated by the
341: cross centered and labeled as ``SN 1987A", with the other
342: four high energy objects within
343: this field of view indicated by crosses.
344: Our sensitivity is considered to be limited up to
345: one degree in radius from the center. The
346: PSF is considered to be a 0.24 degree radius circle.
347: %Therefore, some of interesting targets overlaps.
348: The significance distributions (excess divided by the statistical error)
349: are approximately normal (Gaussian) distributions.
350: The best fit Gaussians have mean values
351: of $-$0.31$\pm$0.08 (2004) and $-$0.18$\pm$0.09 (2004),
352: and standard deviations of 1.21$\pm$0.06 (2004) and 1.31$\pm$0.06 (2006),
353: %Both have marginal deviations. We, however, concluded these are
354: within systematic uncertainties.
355:
356: At first we investigate the region of SN 1987A. Note that
357: this region at some energies is contaminated by 30~Dor~C,
358: a highly energetic source.
359: To indicate the excess distribution from its center, we show
360: so-called $\theta^2$ distributions in Figs. \ref{fig2}.
361: \begin{figure}[htbp]
362: \plotone{f2.eps}
363: \caption{Distributions of $\theta^2$ [degree$^2$]. The left panel
364: is obtained from the 2004 data and the right from the 2006 data.
365: The vertical scale is normalized to counts/minute per bin
366: [0.2 degree$^2$]. The
367: background-subtracted signals (obtained by the fitting
368: procedure described in the text) are plotted.}
369: \label{fig2}
370: \end{figure}
371: Here, the control-background sample was selected in the region
372: $\theta^2$=(0.1--0.2) [degree$^2$].
373: %The left panel is obtained from 2004 data and the right 2006.
374: The background-subtracted signals are shown.
375: Both 2004 and 2006 data have statistically
376: insignificant excesses near $\theta^2=0$.
377: The $\chi^2/$DOFs (degree of freedom) for null assumptions are 18.6/15
378: and 13.2/15 for 2004 and 2006, respectively.
379: %
380: These statistically insignificant excesses
381: appear as the small excess count peaks around the centers of
382: the plots in Fig.~\ref{fig1}.
383: %
384: Note that our PSF corresponds to $\theta^2 <$0.06 degree$^2$.
385: We, therefore, proceed assuming that there is no signal present.
386: In order to show the consistency between 2004 and 2006, the
387: vertical unit is unified to count rate.
388:
389: In order to determine whether or not there is a
390: gamma-ray excess around SN 1987A,
391: we made the
392: $FD$ distributions within the PSF. The control-background was made
393: from the region $\theta^2$=(0.1--0.2). The fitting results are
394: shown in Fig. \ref{fig3}.
395: \begin{figure}[htbp]
396: \plotone{f3.eps}
397: \caption{$FD$ distributions. The left panel is obtained from
398: the 2004 data and the right the 2006 data. The vertical
399: scale is counts/minute per bin [arbitrary unit]. The black
400: crosses are obtained from the on-source region (centered
401: around SN 1987A within the PSF), the green histograms the best fit
402: of the control backgrounds, the blue crosses the
403: background-subtracted signals, and the red histograms
404: the best fit signal distributions.}
405: \label{fig3}
406: \end{figure}
407: %The left panel is obtained from the 2004 data and the right 2006.
408: %The black
409: %crosses are obtained from the on-source region (centered
410: %around SN 1987A within the PSF), the green histograms the best fit
411: %of the control backgrounds, the blue crosses the
412: %background-subtracted signals, and the red histograms
413: %the best fit signal distributions.
414: The background level differs between the two years.
415: This is due to sky conditions, mirror reflectivity, time-dependence of
416: the blur spot size (which improves after maintenance periods), etc.
417: These are considered to be the major systematics for estimating the
418: observational gamma-ray flux in this work.
419: For the light correction efficiency, we may have 10\%
420: uncertainty at maximum.
421:
422: Again there is no statistically significant excess.
423: We carried out a similar analysis, year by year, energy
424: threshold by energy threshold.
425: The obtained 2$\sigma$ upper limits (ULs) are listed in Table~\ref{table1}.
426: \begin{table}[htbp]
427: \caption{The 2$\sigma$ upper limits to the integral fluxes
428: from SN1987A
429: at two epochs with three different energy thresholds.}
430: \label{table1}
431: \begin{tabular}{cccc}
432: \hline\hline
433: Year & Excess Upper Limit & Energy Threshold & Flux Upper Limit \\
434: & Events & GeV & ${\rm cm}^{~2}{\rm s}^{-1}$\\
435: \hline
436: 2004 & \,35\,~ & 1200 & 1.7 $\times$ 10$^{-11}$ \\
437: 2004 & \,21\,~ & 2870 & 5.7 $\times$ 10$^{-13}$ \\
438: 2004 & \,~4.6 & 7180 & 1.0 $\times$ 10$^{-13}$ \\
439: 2006 & 116\,~ & 1080 & 2.7 $\times$ 10$^{-12}$ \\
440: 2006 & \,19\,~ & 2550 & 2.5 $\times$ 10$^{-13}$ \\
441: 2006 & \,~5.3 & 6350 & 6.1 $\times$ 10$^{-14}$ \\
442: \hline\hline
443: \end{tabular}
444: \end{table}
445: Here we used a $E^{-2.1}$ spectrum for the gamma-ray
446: simulation.
447: The ULs range from 6--17\% crab. The worse upper limits in the lower
448: energy range originated from the statistically insignificant
449: excess around the center of the field of view. At higher energies,
450: we do not see any excess.
451:
452: The four other high energy sources in the surrounding region,
453: 30~Dor~C, PSR B0540$-$69, LMC X-1, and N157B, can be
454: analyzed in the same way. They are all within a one degree circle
455: from the average pointing position (the sensitive region).
456: The summary is listed in Table~\ref{table2}.
457: \begin{table}[htbp]
458: \caption{The 2$\sigma$ upper limits to the integral fluxes at energy
459: greater than 1120~GeV for five high energy objects.
460: The 2004 and 2006 data were averaged.}
461: \label{table2}
462: \begin{tabular}{cccc}
463: \hline\hline
464: Target & Excess Upper Limit & Flux Upper Limit \\
465: & Events & ${\rm cm}^{-2}{\rm s}^{-1}$\\
466: \hline
467: SN 1987A & 126 & 1.8 $\times$ 10$^{-12}$ \\
468: 30 Dor C & 165 & 2.5 $\times$ 10$^{-12}$ \\
469: PSR B0540$-$69 & ~66 & 1.1 $\times$ 10$^{-12}$ \\
470: LMC X-1 & ~74 & 1.1 $\times$ 10$^{-12}$ \\
471: N157B & 167 & 2.4 $\times$ 10$^{-12}$ \\
472: \hline\hline
473: \end{tabular}
474: \end{table}
475: These are two year averages.
476: Note that, considering our PSF of 0.24 degree, SN 1987A and
477: 30~Dor~C are highly confused, with other pairs being less confused.
478: Also some of the control-background regions are confused.
479: The acceptance is a slowly and smoothly decreasing function and is reduced
480: to 65\% at a distance of one degree from the center.
481:
482: \section{Discussion}
483:
484: Our morphology indicates a negligibly
485: small excess near the center of field of
486: view both in the 2004 and 2006 data. This is statistically $<$2$\sigma$
487: each year.
488: We note that a similar feature can also be seen in the H.E.S.S.\ data shown
489: in Figs.~2 and 3 of \citet{rowell}. A deep survey,
490: i.e., longer observation, around this region is awaited.
491: The region around LMC X-1 is consistent with zero in our data.
492:
493: The upper limits of SN 1987A in Table~\ref{table1}
494: are plotted in Fig.~ \ref{fig4}.
495: \begin{figure}
496: \plotone{f4.eps}
497: \caption{Spectral energy distribution.
498: The blue points and line are obtained from the 2004 data
499: and the black from the 2006 data. The red points are the CANGAROO-II
500: upper limits from 2001 data \citep{enomoto2003}. The green points are
501: the H.E.S.S.\ data from 2003 \citep{rowell}.
502: The lines are obtained from
503: Fig.\ 4 of \cite{berezhko}. The solid and dashed curves are
504: the predictions of the gamma-ray flux 8249 and 7300
505: days after the supernova, respectively.}
506: \label{fig4}
507: \end{figure}
508: The results presented here (in blue and black) are compared
509: with CANGAROO-II (red)
510: \citep{enomoto2003} and H.E.S.S.\ (green)
511: \citep{rowell} results. The upper limits in this work are
512: improved by factors of up to 10 on the 2001 result.
513: That is consistent with the
514: longer observation time and better signal-to-noise ratio resulting from
515: the three-fold coincidence.
516: However, the theoretical prediction even 7300 days after the supernova
517: (the dashed curve) \citep{berezhko} is still a factor of three
518: below the results of the observations.
519: Note that
520: the predictions \citep{berezhko} were made
521: in the frame of spherical symmetric approach. The existence of
522: strongly asymmetric interstellar matter structure,
523: which is a dense inner ring, makes
524: their prediction of gamma-ray flux uncertain. According to their rough
525: estimation, this uncertainty is not very large (about a factor of 2) due
526: to the stronger supernova shock deceleration in the denser medium.
527: The solid curve is their prediction for 8249 days after the supernova, still
528: lower than the upper limits obtained in this work.
529: In order to constrain this model, we need further improvements
530: on the hardware and/or analysis methods.
531: A deeper observation by H.E.S.S., such as T$>$30 h, around $\sim$2010
532: might give conclusive results.
533: Also proposed future large-scale IACT arrays such as CTA \citep{cta}
534: would be very promising.
535:
536: The limit obtained for 30~Dor~C in Table~\ref{table2} is about
537: 15\% crab level at $\sim$1\,TeV.
538: Considering the distance of 50\,kpc to LMC, this upper limit
539: corresponds to a huge cosmic-ray density comparing with that of
540: the Crab nebula (distance of 2\,kpc).
541: Even if there is supernova complex there, the expected
542: TeV gamma-ray flux is still below our sensitivity limit.
543: Increasing the cosmic-ray target density such
544: as either inter-stellar-matter
545: or photon density within plausible ranges, the expected flux
546: does not exceed this high upper limit.
547: According to the standard estimation on the hadronic gamma-ray production
548: such as \citet{drury,naito} with an assumption of the target density of
549: 1p/cm$^{-3}$, a total kinetic explosion energy of $7\times10^{51}$\,ergs
550: gives a 1 milli-crab flux of gamma-rays at the solar system.
551: Again future larger-scale projects are required to probe such flux levels.
552:
553: For the other three high energy objects, we set upper limits
554: far below those of previous observations.
555: We, however, could not find
556: any physically important constraints on
557: their activity at high energies again
558: due to the distance of 50\,kpc.
559:
560: \section{Conclusion}
561:
562: We have observed SN 1987A at two epochs: MJD 53320--53323
563: (2004 Nov) and MJD 54081--54096 (2006 Dec),
564: approximately 6500 and 7200 days after the supernova, respectively.
565: The effective observation times are 10.5 and 20\,h, respectively.
566: No statistically significant gamma-ray signals were obtained
567: and we set 2$\sigma$ upper limits on the integral fluxes of 6--17\%
568: crab at $\sim$1\,TeV.
569: These are slightly lower than the previous CANGAROO-II and H.E.S.S.\
570: upper limits.
571: Although theory predicts a
572: %an order of magnitude?
573: factor of $\geq$3 lower flux 7300 days after supernova,
574: it also predicts that the future flux level might exceed the
575: sensitivity limit of the existing and future arrays.
576: Continued monitoring of this object at TeV energies
577: is still therefore meaningful in constraining theoretical models.
578: It is also important to improve the sensitivity
579: of the observation methods, as proposed by the CTA project \citep{cta}.
580:
581: In addition, we set 2$\sigma$ upper limits for
582: four more high energy objects inside our
583: field of view:
584: 30~Dor~C (super bubble), PSR B0540$-$69 (Crab-like pulsar),
585: LMC X-1 (X-ray binary), and N157B (supernova remnant).
586:
587: \acknowledgments
588:
589: We thank L.T.\ Ksenofontov for discussion on the estimated gamma-ray
590: flux from SN 1987A.
591: This work was supported by a Grant-in-Aid for Scientific Research by
592: the Japan Ministry of Education, Culture, Sports, Science and Technology,
593: the Australian Research Council, JSPS Research Fellowships,
594: and the Inter-University Research Program
595: of the Institute for Cosmic Ray Research.
596: We thank the Defense Support Center Woomera and BAE Systems.
597:
598: \begin{thebibliography}{}
599:
600: \bibitem[Bamba et al.(2004)]{bamba}
601: Bamba, A., Ueno, M., Nakajima, H., \& Koyama, K., 2004,
602: ApJ, 602, 257
603: \bibitem[Berezhko \& Ksenofontov(2006)]{berezhko}
604: Berezhko, E.G., \& Ksenofontov, L.T., 2006, ApJ, 650, L59
605: \bibitem[Burrows et al.(1995)]{1995ApJ...452..680B}
606: Burrows, C.~J., et al.\ 1995, \apj, 452, 680
607: \bibitem[Daum et al.(1997)]{wobble}
608: Daum, A., et al. 1997, Astropart.\ Phys., 8, 1
609: \bibitem[Drury, Aharonian, \& V\"olk(1994)]{drury}
610: Drury, L.O'C., Aharonian, F.A., \& V\"olk, H.J., 1994, A\&A, 287, 959
611: \bibitem[Enomoto et al.(2002a)]{enomoto_nature}
612: Enomoto, R., et al. 2002a, Nature, 416, 823
613: \bibitem[Enomoto et al.(2002b)]{enomoto_app}
614: Enomoto, R., et al. 2002b, Astropart.\ Phys., 16, 235
615: \bibitem[Enomoto et al.(2003)]{enomoto2003}
616: Enomoto, R., et al., 2003, ApJ, 591, L25
617: \bibitem[Enomoto et al.(2006a)]{enomoto_vela}
618: Enomoto, R., et al.
619: 2006a, ApJ, 638, 397
620: \bibitem[Enomoto et al.(2006b)]{enomoto_0852}
621: Enomoto, R., et al.
622: 2006b, ApJ, 652, 1268
623: \bibitem[Fisher(1936)]{fisher}
624: Fisher, R.~A. 1936, Annals of Eugenics, 7, 179
625: \bibitem[Habert et al.(2006)]{xmm}
626: Habert, F., Geppert, U., Aschenbach, B., \& Hasinger, G., 2006,
627: A\&A, 260,811.
628: \bibitem[Hermann (2007)]{cta}
629: Hermann, G., 2007,
630: Cherenkov Telescope Array (CTA), http://www.mpi-hd.mpg.de/htm/CTA/
631: \bibitem[Hillas(1985)]{hillas}
632: Hillas, A.~M. Proc.\ 19th Int.\ Cosmic Ray Conf.\ (La Jolla) 3, 445
633: \bibitem[Kabuki et al.(2003)]{kabuki}
634: Kabuki, S., et al. 2003, Nucl.\ Instrum.\ Meth., A500, 318
635: \bibitem[Kabuki et al.(2007)]{cena}
636: Kabuki, S., et al. 2007, ApJ, v667, in press
637: (arXiv:0706.0367)
638: \bibitem[Kawachi et al.(2001)]{kawachi}
639: Kawachi, A., et al. 2001, Astropart.\ Phys., 14, 261
640: %\bibitem[Klepach et al.(2000)]{klepach}
641: % Klepach, E.G., Ptuskin, V.S., \& Zirakashvili, V.N., 2000,
642: % Astropart.\ Phys., 13, 161
643: \bibitem[NASA (2007)]{skyview}
644: NASA, 2007,
645: SkyView, http://skyview.gsfc.nasa.gov/
646: \bibitem[Morris \& Podsiadlowski(2007)]{mor07}
647: Morris, T., \& Podsiadlowski, P., 2007, Science, 315, 1103
648: \bibitem[Naito \& Takahara(1994)]{naito}
649: Naito, T. \& Takahara, F., 1994, J.\ Phys.\ G., 20, 477
650: \bibitem[Nishijima et al.(2005)]{nishijima}
651: Nishijima, K., et al. 2005, Proc.\ 29th Int.\ Cosmic Ray Conf.\ (Pune),
652: OG2.7, 101
653: \bibitem[Parizot et al.(2004)]{parizot}
654: Parizot, E., Marcowith, A., van der Swaluw, E., \&
655: Tatischeff, V., 2004, A\&A, 424, 747
656: \bibitem[Park et al.(2007)]{chandra}
657: Park, S., Burrows, D.N., Garmire, G.P., McCray, R., Racusin, J.L.,
658: \& Zhekov, S.A.,
659: 2007,
660: To appear in the proceedings of "Supernova 1987A: 20 Years after
661: Supernovae and Gamma-Ray Bursters", held in Aspen CO. USA, Feb 19-23,
662: arXiv:0704.0209
663: \bibitem[Rowell et al.(2005)]{rowell}
664: Rowell, G., Hinton, J., Benbow, W., \& H.E.S.S. Collaboration,
665: 2005, in
666: AIP Conf. Proc. 745, High Energy Gamma-Ray Astronomy,
667: ed. F.A. Aharonian, H.J. V\"olk, \& D. Horns (Melville: AIP), 299
668: \bibitem[Staveley-Smith et al.(2007)]{atnf}
669: Staveley-Smith, L., Ball, L., Gaensler, B., Kestreven, M.,
670: Manchester, D., \& Tzioumis, T., 2007,
671: ATNF News, Issue No. 61, ISSN 1323-6326,
672: http://www.atnf.csiro.au/news/newsletter/feb07/
673: \bibitem[Ueno, Bamba, \& Koyama(2003)]{ueno}
674: Ueno, M., Bamba, A., \& Koyama, K., 2003, in ``The Universe Viewed in
675: Gamma-Rays'',
676: Proc.\ Univ.\ of Tokyo Symposium,
677: ed. R. Enomoto, M. Mori, \& S. Yanagita
678: (Universal Academy Press: Tokyo), 311
679: \bibitem[Wright et al.(1994)]{pmns} Wright, A.~E., Griffith,
680: M.~R., Burke, B.~F., \& Ekers, R.~D.\ 1994, \apjs, 91, 111
681: \end{thebibliography}
682:
683: \end{document}
684: