1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass[manuscript]{aastex}
3: %\documentclass[preprint2]{aastex}
4:
5: \newcommand{\vdag}{(v)^\dagger}
6: \newcommand{\myemail}{enomoto@icrr.u-tokyo.ac.jp}
7:
8: \slugcomment{To appear in ApJ v596n1, Oct. 10, 2003.}
9:
10: \shorttitle{Constraints on CDM in NGC 253}
11: \shortauthors{Enomoto, Yoshida, Yanagita, Itoh}
12:
13: \begin{document}
14:
15: \title{Constraints on Cold Dark Matter in the Gamma-ray Halo of NGC 253}
16:
17: \author{
18: R.~Enomoto\altaffilmark{1},
19: T.~Yoshida\altaffilmark{2},
20: S.~Yanagita\altaffilmark{2},
21: and C.~Itoh\altaffilmark{3}
22: }
23:
24: \altaffiltext{1}{Institute for Cosmic Ray Research, University of Tokyo,
25: Chiba 277-8582, Japan}
26: \altaffiltext{2}{Faculty of Science, Ibaraki University, Ibaraki 310-8512, Japan}
27: \altaffiltext{3}{Ibaraki Prefectural University of Health Sciences, Ibaraki 300-0394, Japan}
28:
29: \email{\myemail}
30:
31: \begin{abstract}
32: A gamma-ray halo in a nearby starburst galaxy NGC 253 was found by
33: the CANGAROO-II Imaging Atmospheric Cherenkov Telescope (IACT).
34: By fitting the energy spectrum with expected curves from
35: Cold Dark Matter (CDM) annihilations,
36: we constrain the CDM-annihilation rate
37: in the halo of NGC 253.
38: Upper limits for the CDM density were obtained in the wide mass
39: range between 0.5 and 50 TeV.
40: Although these limits are higher than the
41: expected values, it is complementary important
42: to the other experimental techniques,
43: especially considering the energy coverage.
44: We also investigate the next astronomical targets to improve
45: these limits.
46: \end{abstract}
47:
48: \keywords{galaxy: individual (NGC 253)---
49: gamma rays: theory --- dark matter}
50:
51: \section{Introduction}
52:
53: Recently, the gamma-ray halo in the nearby starburst galaxy NGC 253
54: was detected by CANGAROO-II \citep{itoh1,itoh3}.
55: Although it seems to be successful to interpret it as a high-energy
56: cosmic-ray halo \citep{itoh2},
57: we assumed that this radiation is due
58: to cold dark matter (CDM) annihilation,
59: and obtained the upper limits of the CDM
60: density in a wide mass range around TeV,
61:
62: The motivation of this study is the morphology obtained in a
63: TeV gamma-ray observation \citep{itoh1,itoh3}. It marginally differed from a
64: disk shape. Considering the existence of this halo \citep{ostriker},
65: it would be worth obtaining the constraint of CDM
66: by using the observed TeV emissions.
67:
68: In this paper we assumed two processes:
69: inclusive gamma-ray production via
70: the annihilation of weakly interacting
71: massive particles to
72: quark anti-quark pairs \citep{rudaz},
73: and
74: monochromatic gamma-rays
75: from the annihilation to two gamma state \citep{bergstrom}.
76: The former final states
77: produce highly multiple
78: gamma-rays and gave better upper limits than the latter method.
79: The final-state gamma-rays would show
80: an exponential energy spectrum which differs
81: from the usually known cosmic-ray spectrum,
82: i.e., a power law.
83:
84: \section{Property of NGC 253}
85:
86: The nearby starburst galaxy NGC 253 is located inside the Sculptor group,
87: and can be clearly seen from the southern hemisphere. The distance was
88: estimated to be 2.5 Mpc \citep{vau}.
89: It was classified as SABc (Hubble classification), and is one of
90: the closest examples of ``our Galaxy-like object".
91: It is ``edge-on", i.e., suitable to distinguish its halo from the disk.
92: The optical \citep{beck} and radio halos \citep{hummel,carilli}
93: were previously observed in this galaxy, the sizes of which ($\sim$
94: 10 kpc) approximately
95: agreed with a TeV-gamma-ray observation (13 - 26 kpc) \citep{itoh3}.
96:
97: HI studies on the Sculptor group galaxies were carried out \citep{puche}.
98: Calculating the rotation curves, they concluded that many galaxies in this
99: group have massive halos.
100: Especially, that of NGC 253 was estimated to have a density of
101: 0.015 M$_{\odot}$pc$^{-3}$ with a radius of 26.9 kpc, which is also
102: within the size estimation of the TeV-gamma-ray halo \citep{itoh3}.
103:
104: \section{Energy spectrum and mass of CDM}
105:
106: The spectral energy distribution (SED) of GeV-TeV gamma
107: rays is plotted in Fig. \ref{fig1}.
108: The points with error bars were obtained by CANGAROO-II (Table
109: 6 of \citet{itoh3}).
110: The arrows are upper limits obtained by EGRET \citep{egret1,egret2}.
111: According to a theoretical estimation motivated by the cosmic-ray
112: radiation \citep{itoh2}, neither a simple power-law spectrum
113: ($\propto E^{-\gamma}$) of
114: cosmic-rays (curve-A) nor that with a high energy cutoff
115: ($\propto E^{-2}e^{E/E_{max}}$) (curve-B:
116: due to
117: $\pi^0\rightarrow \gamma \gamma$ decays or bremsstrahlung)
118: could simultaneously explain both data.
119: These curves can be well fitted to the TeV data, however,
120: they are inconsistent with the GeV upper limits.
121: The only choice
122: to satisfy both data was inverse Compton scattering oriented by very hard
123: incident electrons
124: ($\propto E^{-1.5}e^{\sqrt{E}/b}$) (curve-C),
125: which may require another mechanism
126: of re-acceleration in the galactic halo.
127:
128: Fig. \ref{fig2} shows
129: a semi-log plot of the differential
130: flux of TeV gamma-rays.
131: The line-E is the best-fitted exponential function
132: ($\propto e^{-aE}$), and shows an agreement.
133: The extrapolation to GeV region is well below the EGRET upper limits.
134:
135: The exponential function has a physical scale (in this case energy scale)
136: and its contribution in the GeV region is negligibly small in SED.
137: The well-known physical process to obtain an energy scale is a fragmentation
138: function ($\frac{1}{\sigma _h} \cdot \frac{d\sigma}{dx}$,
139: where $x$ is a Feynman $x$)
140: for such an inclusive particle spectrum as
141: $e^+e^-\to q\bar{q}\to\gamma X$. It is typically to be fitted with
142: the sum of the exponential functions. The fragmentation function of
143: LEP data ($e^+e^-$ collider experiment at the center of mass energy of
144: $\sim$90 GeV) \citep{lep} was well-fitted with
145: the sum of three exponential functions:
146: \begin{eqnarray*}
147: \frac{1}{\sigma _h} \cdot \frac{d\sigma}{dx}
148: &=& e^{5.5605-34.482x} + e^{3.1777-10.551x}\\
149: & &+ e^{7.2391-123.29x}.
150: \end{eqnarray*}
151: Introducing the energy scale $M_\chi$,
152: the relationship between $x$ and the energy of gamma-ray becomes
153: $x=E/M_\chi$. The annihilation rate ($F$ [cm$^{-2}$s$^{-1}$])
154: and the energy scale
155: were obtained by fitting the TeV-gamma-ray's spectrum with
156: $\frac{F}{\sigma _h} \cdot \frac{d\sigma}{M_\chi \cdot dx}$ to be
157: \begin{eqnarray*}
158: F&=&(1.8\pm 1.1) \times 10^{-11}\ [{\rm cm}^{-2}{\rm s}^{-1}],\\
159: M_\chi &=& (3.0\pm 0.6)\ [{\rm TeV}],
160: \end{eqnarray*}
161: where we used the TeV gamma-ray fluxes from Table 6 in Ref. \citep{itoh3}.
162: Note that these two parameters are highly anti-correlated, which
163: will be reflected in further analysis,
164: i.e., described later.
165: Here, $F$ is the observed annihilation rate
166: per unit area and time at Earth.
167: The result is shown by line-D in Fig. \ref{fig1}.
168: The $\chi^2$ obtained in this fitting was 1.0/D.O.F=4.
169: The EGRET upper limits are also cleared.
170:
171: For the reaction of CDM to $\gamma \gamma$ (i.e., $\chi \chi \rightarrow
172: \gamma \gamma$), the monochromatic gamma-rays were suggested to be searched
173: \citep{bergstrom,bouquet,jungman}. The energy resolution of TeV gamma-rays
174: is approximately 35\% (Table 5 of \citet{itoh3}).
175: The curve-F in Fig. \ref{fig2} is an example of a Gaussian
176: ($\propto e^{-\frac{1}{2}\left(\frac{E-M_\chi}{\sigma_E}\right)^2}$)
177: with that resolution and a center value of 0.7 TeV.
178:
179: \section{Upper limit for the number density of CDM}
180:
181: The annihilation in the volume of the halo should
182: be detected at a rate of
183: $$F=<\sigma v>B_{q\bar{q}}n^2[V/(4\pi d^2)],$$
184: where $\sigma$ is the annihilation cross section,
185: $v$ the relative velocity of CDMs,
186: $B_{q\bar{q}}$ the branching fraction of $\chi\chi\to q\bar{q}$,
187: $n$ the number density of CDM, $V$ the total volume of the halo,
188: and $d$ the distance from Earth.
189:
190: Here, we consider the last volume-distance factor. The dark halo size
191: of NGC 253
192: obtained by an HI measurement is 26.9 kpc \citep{puche}, which
193: corresponds to a solid angle of $\Delta \theta = 0.62^o$.
194: Thus the volume factor becomes $\frac{d}{3}\cdot(\Delta \theta)^3$,
195: proportional to the distance. When we see the same-angular-sized diffuse image,
196: it suggests that distant objects have advantages. For example,
197: comparing the Galactic Center (distance of 8.5 kpc)
198: and NGC 253, this factor becomes 300.
199: Although the Galactic Center may have a CDM concentration factor of, say,
200: 1000 \citep{navarro}, it is highly model dependent.
201: On the other hand, the total volume average of the squared density
202: is less model dependent (only a factor changes
203: under the assumption of $r^{-n}$, n=0,1,..).
204:
205: The annihilation cross section is another source of model dependences.
206: For example, whether CDM is Dirac or Majorana fermion. Also,
207: it depends on the details of particle physics,
208: i.e., the details of SUSY breaking \citep{jungman2}.
209: A much larger dependence is expected to the $\chi\chi\to\gamma\gamma$
210: process. In order to avoid it, we carried out the following.
211: According to the Lee-Weinberg equation of the CDM density evolution
212: \citep{lee,jungman2}, the annihilation cross section is
213: directly related
214: to the cosmological abundance of $\Omega_{CDM}$,
215: $$\Omega_{CDM}=7\times 10^{-27}\ [{\rm cm}^{3}{\rm s}^{-1}]/<\sigma v>,$$
216: which is mass independent.
217: Recently, WMAP determined $\Omega_{CDM}=0.23$ \citep{wmap}.
218: With this, we normalized $<\sigma vB_{q\bar{q}}>$ to
219: an order of $10^{-26}\ [{\rm cm}^{3}{\rm s}^{-1}]$.
220:
221: Now that all of unknown factors have been filled,
222: by fitting the TeV gamma-rays spectrum with the described function,
223: we can derive the best-fitted density for CDM,
224: \begin{eqnarray*}
225: n=(2.4 \pm 0.6) \times 10^{-2}\sqrt{\frac{
226: 10^{-26} {\rm cm}^{3}{\rm s}^{-1}}{<\sigma v B_{q\bar{q}}>}}[{\rm cm}^{-3}],
227: \end{eqnarray*}
228: where $n$ is still highly correlated with the $M_\chi$ value.
229: Changing $n$ to the energy density of the CDM,
230: \begin{eqnarray*}
231: \rho_{CDM}=(70.8 \pm 7.4) \sqrt{\frac{
232: 10^{-26} {\rm cm}^{3}{\rm s}^{-1}}{<\sigma v B_{q\bar{q}}>}}
233: [{\rm GeV}{\rm cm}^{-3}]
234: \end{eqnarray*}
235: is obtained, where the correlation between $M_\chi$ and $\rho_{CDM}$
236: is shown in Fig. \ref{fig3}.
237: Note that $n$ and $\rho_{CDM}$ are the root-mean-squared
238: volume average of those densities.
239: The lines are 1- and 2-$\sigma$ contours.
240: The reasons why the errors became smaller compared to the value of $F$
241: are because $\rho_{CDM}$ is proportional to $\sqrt{F}$ and
242: there is a strong anti-correlation between $F$ and $M_\chi$.
243:
244: To summarize, the final fitting data and functions are as follows:
245: \begin{eqnarray*}
246: \left[\frac{dF}{dE}\right]&=&\frac{F}{M_\chi}\cdot \left[\frac{1}{\sigma _h}
247: \cdot \frac{d\sigma}{d(E/M_\chi)}\right]\\
248: &=& \frac{<\sigma vB_{q\bar{q}}>n^2[V/(4\pi d^2)]}{M_\chi}\\
249: & & \hskip 3cm\cdot \left[\frac{1}{\sigma _h}
250: \cdot \frac{d\sigma}{d(E/M_\chi)}\right] \\
251: &=& \frac{<\sigma vB_{q\bar{q}}>\rho_{CDM}^2[V/(4\pi d^2)]}{M_\chi^3}\\
252: & & \hskip 3cm\cdot \left[\frac{1}{\sigma _h}
253: \cdot \frac{d\sigma}{d(E/M_\chi)}\right],
254: \end{eqnarray*}
255: where $[\frac{1}{\sigma _h} \cdot \frac{d\sigma}{d(E/M_\chi)}]$
256: should be replaced
257: with the linear combination of three exponential functions
258: described so far and $[\frac{dF}{dE}]$
259: are Table 6 of \cite{itoh3} and Fig. 1 of \cite{egret2}, respectively.
260:
261: Although, these values are the best-fitted ones,
262: there are not enough reasons to insist that this is evidence for CDM
263: other than that
264: the energy spectrum was fitted well with the exponential
265: function.
266: This value should be considered to be an upper limit.
267: We, therefore, carried out a scan for various $M_\chi$ assumptions,
268: and obtained upper limits versus $M_\chi$.
269: The results are shown in Fig. \ref{fig4}.
270: Considering the dynamic range of the fragmentation function measurements,
271: the searched range was selected to be from 0.5 to 50 TeV.
272: In the figure, the 2$\sigma$-upper limits are shown.
273: The improvement bellow 0.65 GeV was due to the EGRET upper limits.
274:
275: In addition, we carried out a search for monochromatic gamma-rays
276: in the energy region between 0.5 and 3 TeV.
277: The TeV gamma-ray energy spectrum was fitted with Gaussians
278: under various peak-energy assumptions.
279: A uniform energy resolution of 35\% was assumed.
280: A typical line is shown
281: in Fig. 2 (line-B). The 2$\sigma$-upper limits for
282: the mass densities of CDM
283: , $\rho_{CDM}\sqrt{\frac{10^{-29}{\rm cm}^3{\rm s}^{-1}}
284: {<\sigma_{\gamma\gamma} v>}}$,
285: were obtained ( Fig. \ref{fig5} ).
286: Here we used a smaller normalization factor for $<\sigma_{\gamma\gamma} v>$
287: as was expected from the particle theory \citep{bergstrom}.
288: These upper limits
289: are higher than these in Fig. \ref{fig4}.
290:
291: \section{Discussion}
292:
293: The halo density estimated by the HI studies
294: is 0.015 M$_{\odot}$pc$^{-3}$ (0.57 $[{\rm GeV}{\rm cm}^{-3}]$) \citep{puche},
295: which is two orders lower than our upper limits.
296: The gamma-ray flux from NGC 253 should be explained
297: by the standard cosmic-ray theory.
298: Due to the starburst phenomena, we could not throw away a cosmic-ray
299: interpretation \citep{volk}.
300: If the cosmic-ray emission is accurately determined
301: from the study of multi-wavelength spectrum, this
302: upper limit will be greatly reduced to the error-bar level.
303:
304: A scan of the nearby galaxies that are not starburst will be promising
305: in the search for CDM.
306: Especially, the Sculptor group is an interesting
307: target, which is also suggested by an HI measurement \citep{puche}.
308:
309: The search for gamma-rays from massive galaxies is considered to
310: reduce the upper limit for the galactic density of CDM.
311: Ten-times-heavier astronomical objects that are nearby,
312: would give a sensible result.
313: For example, M 87 is considered to be more than ten-times
314: heavier than our Galaxy \citep{baltz}.
315: The distance is several-times farther than NGC 253.
316: The volume-distance factor, $\frac{d}{3}\cdot({\Delta \theta})^3$, is,
317: therefore, an order
318: lower than our case (reported ${\Delta \theta}<0.127^o$ \citep{m87}).
319: On the other hand, the recent
320: observation by HEGRA reported about a ten-times fainter gamma-ray
321: intensity \citep{m87}.
322: These values should result in the same order of CDM density,
323: while M87's density is considered to be larger than that of NGC 253.
324: Publication of the differential
325: flux of the TeV gamma-ray from M87 is awaited.
326:
327: Although M 31 is also massive, the visible size is larger than
328: the field of view of IACTs, which requires special treatments for
329: background subtractions \citep{m31}.
330:
331: Considering a figure of merit ($FOM$) for the CDM search,
332: we had better consider total mass of galaxy, volume of halo, distance,
333: and visible size simultaneously.
334: The CDM density may be proportional to the total mass
335: ($M_G$) divided by the volume.
336: Thus, the expected gamma-ray's flux should be proportional to
337: the following: $FOM=M_G^2d^{-5}(\Delta \theta )^{-3}$.
338: Selecting those nearby galaxies that have a visible size of between
339: $3\times 10^{-3}$ and $10^{-2}$ radian
340: (favorable size for the IACT measurements), NGC 5128 (Cen-A) and
341: NGC 5236 (M 83) were calculated to have a bigger $FOM$ than that
342: of NGC 253.
343: Especially for Cen-A, hundred-times larger flux is expected.
344:
345: We also applied the same discussion to $\omega$-Centauri
346: \citep{guy}.
347: The dark matter origin of globular clusters was
348: proposed by \cite{peebles}.
349: The distance is close and lower cosmic-ray level is expected.
350: The $FOM$ was calculated to be 10000 times higher than that of NGC 253.
351: An upper
352: limit which is the same order with the baryonic density
353: could be obtained.
354:
355: Also, a high-sensitivity search in the Galactic Center is awaited
356: \citep{tsuchiya}.
357: However, to remove the model dependence and to estimate
358: the cosmic-ray there are keys for this case.
359:
360: Compared to the accelerator experiment, only IACT measurements are
361: sensitive to CDM with a mass heavier than TeV.
362: They are complementary important to each other.
363:
364: \section{Conclusion}
365:
366: A constraint on the cold dark matter (CDM) was obtained
367: using the data of
368: the gamma-ray halo around the nearby starburst galaxy NGC 253.
369: According to this study,
370: upper limits for the CDM density were obtained in the mass
371: range between 0.5 and 50 TeV.
372: Although these limit is higher than the expected value, this is one of first
373: trials from the IACT observational side.
374: The IACTs have been proven to have abilities to detect it.
375: The presently existing IACTs are competitive devices compared with high-energy
376: particle accelerators.
377: The nearby galaxies such as NGC 5128 (Cen-A) and NGC 5236 (M 83)
378: and / or globular cluster $\omega$-Centauri
379: will be next interesting targets.
380: Observational efforts for probable candidates should
381: be systematically continued.
382:
383: \acknowledgments
384:
385: We thank Prof. J. Hisano and Prof. M. Fukugita of ICRR for
386: various discussions.
387:
388: \begin{thebibliography}{00}
389:
390: \bibitem[Ackerstaff et al.(1998)]{lep}
391: Ackerstaff, K., et al. 1998, Eur. Phys. J. C5, 411
392:
393: \bibitem[Aharonian et al. (2003a)]{m87}
394: Aharonian, F., et al. 2003a, submitted to A\&A, astro-ph/0302155
395:
396: \bibitem[Aharonian et al.(2003b)]{m31} Aharonian, F.A., et al., 2003b,
397: submitted to A\&A, astro-ph/0202347
398:
399: \bibitem[Baltz et al.(2000)]{baltz} Baltz, E.A., et al., 2000,
400: Phys. Rev. D 61, 023514
401:
402: \bibitem[Beck et al.(1982)]{beck}
403: Beck, R., Hutschenreiter, G., \& R. Wielebinscki, R. 1982,
404: A\&A, 106, 112-114
405:
406: \bibitem[Bennett et al.(2003)]{wmap}
407: Bennett, C.L., et al. 2003, astro-ph/0302207
408:
409: \bibitem[Bergstr\" om \& Snellman(1988)]{bergstrom}
410: Bergstr\" om, L. \& Snellman, H., 1988, Phys. Rev. D37, 3737
411:
412: \bibitem[Blom et al.(1999)]{egret2}
413: Blom, J.J., Paglione, T.A.D., \& Carraminana, A. 1999,
414: ApJ, 516, 744
415:
416: \bibitem[Bouquet et al.(1989)]{bouquet}
417: Bouquet, A., Salati, P., \& Silk, J. 1989, Phys. Rev. D40, 3168
418:
419: \bibitem[Carilli et al.(1992)]{carilli}
420: Carilli, C.L., Holdaway, M.A., \& Ho, P.T.P. 1992,
421: C.G. de Pree, ApJ, 399, L59
422:
423: \bibitem[de Vaucouleurs(1978)]{vau}
424: de Vaucouleurs, G. 1978, ApJ, 224, 710
425:
426: \bibitem[Guy et al.(2002)]{guy} Guy, J., for the H.E.S.S. collaboration,
427: 2002, proceedings of 4th Int. WS. on the identification of Dark
428: Matter, Sept. 2002, York, England
429:
430: \bibitem[Hummel et al(1984)]{hummel}
431: Hummel, E., Smith, P., \& van der Hulst, J.M. 1984, A\&A,
432: 137, 138
433:
434: \bibitem[Itoh et al.(2002)]{itoh1}
435: Itoh, C., Enomoto, R., Yanagita, S., Yoshida, T., et al. 2002,
436: A\&A, 296, L1-4
437:
438: \bibitem[Itoh et al.(2003a)]{itoh3}
439: Itoh, C., Enomoto, R., Yanagita, S., Yoshida, T., et al. 2003a,
440: A\&A, 402, 443-455
441:
442: \bibitem[Itoh et al.(2003b)]{itoh2}
443: Itoh, C., Enomoto, R., Yanagita, S., Yoshida, S., \& Tsuru, T.G. 2003b,
444: ApJ, 584, L65-68
445:
446: \bibitem[Jungman \& Kamionkowski(1995)]{jungman}
447: Jungman, G. \& Kamionkowski, M. 1995, Phys. Rev. D51, 3121
448:
449: \bibitem[Jungman et al.(1996)]{jungman2}
450: Jungman, G., Kamionkowski, M., \& Griest, K. 1996, Phys. Rep., 267, 195
451:
452: \bibitem[Lee \& Weinberg(1977)]{lee}
453: Lee, B.W. \& Weinberg, S. 1977, Phys. Rev. Lett., 39, 165-168
454:
455: \bibitem[Navarro et al.(1996)]{navarro}
456: Navarro, J.F., Frenk, C.S., \& White, S.D.M. 1996, ApJ. 462, 563
457:
458: \bibitem[Ostriker \& Peebles(1973)]{ostriker}
459: Ostriker, J.P. \& Peebles, P.J.E., 1973, ApJ, 186, 467-480
460:
461: \bibitem[Peebles(1984)]{peebles}
462: Peebles, P.J.E., 1984, ApJ, 277, 470-477
463:
464: \bibitem[Puche \& Carignan(1991)]{puche}
465: Puche, D. \& Carignan, C. 1991, ApJ, 378, 487-495.
466:
467: \bibitem[Rudaz \& Stecker(1988)]{rudaz}
468: Rudaz, S. \& Stecker, F.W. 1988, ApJ, 325, 16-25
469:
470: \bibitem[Streekmar et al.(1994)]{egret1}
471: Streekmar, P., et al. 1994, ApJ, 426, 105
472:
473: \bibitem[Tsuchiya et al.(2003)]{tsuchiya}
474: Tsuchiya, K., Enomoto, R., et al., 2003, Proc. on "The Universe
475: Viewd in Gamma-rays", Univ. of Tokyo Symposium, Universal Academy
476: Press, Tokyo, Japan, pp229-234
477:
478: \bibitem[V\" olk et al.(1996)]{volk}
479: V\" olk, H.J., Aharonian, F.A., \& Breitschwerdt, D. 1996,
480: Space Sci. Rev. 75, 279
481:
482: \end{thebibliography}
483:
484: %\end{document}
485:
486: %\clearpage
487:
488: \begin{figure}[htb]
489: \plotone{f1.eps}
490: \figcaption{
491: Spectral energy distribution. The high-energy data were obtained
492: by CANGAROO-II and the low-energy upper limits by EGRET.
493: The lines are the results of various fitting functions describe in the
494: text.
495: \label{fig1}}
496: \end{figure}
497:
498: \begin{figure}[htb]
499: \plotone{f2.eps}
500: \figcaption{
501: Differential flux of gamma-rays from NGC 253 in the semi-log scale.
502: The data were obtained
503: from CANGAROO-II. Line-E is the best-fitted exponential curve and
504: line-F is an example of a Gaussian with an energy resolution of 35\%
505: and a center value of 0.7 TeV.
506: \label{fig2}
507: }
508: \end{figure}
509:
510: \begin{figure}[htb]
511: \plotone{f3.eps}
512: \figcaption{
513: Correlation between $M_\chi$ and $\rho_{CDM}$. 1- and 2-$\sigma$
514: contours are shown with the cross corresponding to the best-fitted value.
515: \label{fig3}
516: }
517: \end{figure}
518:
519: \begin{figure}[htb]
520: \plotone{f4.eps}
521: \figcaption{
522: 2$\sigma$-Upper limits of $\rho_{CDM}$ versus $M_\chi$
523: for various $M_\chi$ assumptions.
524: \label{fig4}
525: }
526: \end{figure}
527:
528: \begin{figure}[htb]
529: \plotone{f5.eps}
530: \figcaption{
531: Upper limits of $\rho_{CDM}$ versus $M_\chi$. Here,
532: a monochromatic gamma-ray search was carried out for
533: the reaction $\chi\chi\to\gamma\gamma$.
534: \label{fig5}
535: }
536: \end{figure}
537:
538: \end{document}
539:
540: