1: \documentclass[aps,prd,twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}

2: 

3: %\documentclass[aps,prd,preprint,groupedaddress]{revtex4}

4: %\documentclass[aps,prd,twocolumn,groupedaddress]{revtex4}

5: 

6: \usepackage{amsmath,amssymb}

7: %\usepackage{showkeys}

8: 

9: \usepackage{color}

10: \usepackage[dvips]{graphicx}

11: %\usepackage{epstopdf}

12: \usepackage{subfigure}

13: \renewcommand{\baselinestretch}{1.1}

14: 

15: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

16: %%% Symbol definitions %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

17: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

18: 

19: \newcommand{\lsim}{\mathrel{\mathop{\kern 0pt \rlap

20:   {\raise.2ex\hbox{$<$}}}

21:   \lower.9ex\hbox{\kern-.190em $\sim$}}}

22: \newcommand{\gsim}{\mathrel{\mathop{\kern 0pt \rlap

23:   {\raise.2ex\hbox{$>$}}}

24:   \lower.9ex\hbox{\kern-.190em $\sim$}}}

25: 

26: \newcommand{\sigmav}{\langle \sigma_{\rm ann} v \rangle}

27: \newcommand{\sigmavzero}{\langle \sigma_{\rm ann} v \rangle_0}

28: 

29: \newcommand{\vmin}{v_{\rm min}}

30: \newcommand{\vesc}{v_{\rm esc}}

31: \newcommand{\ivmin}{{\cal I}(\vmin)}

32: \newcommand{\mc}{\m_{\chi}}

33: \newcommand{\pbar}{\bar{p}}

34: 

35: \newcommand{\beq}{\begin{equation}}

36: \newcommand{\eeq}{\end{equation}}

37: \newcommand{\bea}{\begin{eqnarray}}

38: \newcommand{\ena}{\end{eqnarray}}

39: \newcommand{\etal}{{\it et al.}}

40: \newcommand{\ie}{{\it i.e.}}

41: 

42: \newcommand{\mbf}{\mathbf}

43: \newcommand{\into}{\rightarrow}

44: \newcommand{\lrarrow}{\leftrightarrow}        

45: \newcommand{\be}{\begin{equation}}

46: \newcommand{\ee}{\end{equation}}

47: \newcommand{\ba}{\begin{array}}

48: \newcommand{\ea}{\end{array}}

49: \newcommand{\nn}{\nonumber}

50: \newcommand{\eea}{\end{eqnarray}}

51: \newcommand{\sigmavz}{\langle \sigma v \rangle_0}

52: %

53: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

54: \newcommand{\app}[3]{Astropart.\ Phys.\ {\bf #1}, #3 (#2)}

55: \newcommand{\hepex}[1]{{hep-ex/#1}}

56: \newcommand{\hepph}[1]{{hep-ph/#1}}

57: \newcommand{\astroph}[1]{{astro-ph/#1}}

58: \newcommand{\prep}[3]{Phys.\ Rep.\ {\bf #1}, #3 (#2)}

59: \newcommand{\plb}[3]{Phys.\ Lett.\ B\ {\bf #1}, #3 (#2)}

60: \newcommand{\npb}[3]{Nucl.\ Phys.\ B\ {\bf #1}, #3 (#2)}

61: \newcommand{\npps}[3]{Nucl.\ Phys.\ Proc.\ Suppl.\ {\bf #1}, #3 (#2)}

62: \newcommand{\cpc}[3]{Comm.\ Phys.\ Comm.\ {\bf #1}, #3 (#2)}

63: \renewcommand{\apj}[3]{Astrophys.\ J.\ {\bf #1}, #3 (#2)}

64: \newcommand{\aeta}[3]{Astron.\  Astrophys.\ {\bf #1}, #3 (#2)}

65: \newcommand{\pr}[3]{Phys.\ Rev.\ {\bf #1}, #3 (#2)}

66: \renewcommand{\prl}[3]{Phys.\ Rev.\ Lett. {\bf #1}, #3 (#2)}

67: \renewcommand{\prd}[3]{Phys.\ Rev.\ D\ {\bf #1}, #3 (#2)}

68: \renewcommand{\rmp}[3]{Rev.\ Mod.\ Phys.\ {\bf #1}, #3 (#2)}

69: \newcommand{\rnc}[3]{Riv.\ Nuovo\ Cim.\ {\bf #1}, #3 (#2)}

70: \newcommand{\zfpc}[3]{Z.\ Phys.\ C\ {\bf #1}, #3 (#2)}

71: 

72: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

73: %%%%%%%%%%%%%% Beginning of text %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

74: 

75: 

76: \begin{document}

77: 

78: \preprint{DFTT26/2006}

79: 

80: 

81: \title{Additional bounds on the pre Big--Bang--Nucleosynthesis Expansion by means of

82: $\gamma$-rays from the Galactic Center}

83: 

84: % address or url should go in the {}'s for \email and \homepage.

85: % Please use the appropriate macro for each each type of information

86: 

87: % \affiliation command applies to all authors since the last

88: % \affiliation command. The \affiliation command should follow the

89: % other information

90: % \affiliation can be followed by \email, \homepage, \thanks as well.

91: 

92: %

93: \author{F. Donato}

94:  \affiliation{Universit\`a degli Studi di Torino and \\

95: Istituto Nazionale di Fisica Nucleare, Sezione di Torino

96: \\ via P. Giuria 1, I--10125 Torino, Italy \\ {\tt (donato@to.infn.it)}}

97: %

98: \author{N. Fornengo}

99: \affiliation{Universit\`a degli Studi di Torino and \\

100: Istituto Nazionale di Fisica Nucleare, Sezione di Torino

101: \\ via P. Giuria 1, I--10125 Torino, Italy \\ {\tt (fornengo@to.infn.it)}}

102: %

103: \author{M. Schelke}

104: \affiliation{Istituto Nazionale di Fisica Nucleare, Sezione di Torino

105: \\ via P. Giuria 1, I--10125 Torino, Italy \\ {\tt (schelke@to.infn.it)}}

106: 

107: 

108: 

109: %

110: 

111: %

112: \date{\today}

113: 

114: \begin{abstract} The possibility to use $\gamma$--ray data from the Galactic Center

115: (GC) to constrain the cosmological evolution of the Universe in a phase prior to

116: primordial nucleosyntesis, namely around the time of cold dark matter (CDM)

117: decoupling, is analyzed. The basic idea is that in a modified cosmological scenario,

118: where the Hubble expansion rate is enhanced with respect to the standard case, the

119: CDM decoupling is anticipated and the relic abundance of  a given dark matter (DM)

120: candidate enhanced. This implies that the present amount of CDM in the Universe may

121: be explained by a  Weakly Interacting Massive Particle (WIMP) which possesses

122: annihilation cross section (much) larger than in standard cosmology. This enhanced

123: annihilation implies larger fluxes of indirect detection signals of CDM. We show that

124: the HESS measurements can set  bounds for WIMPs heavier than a few hundreds of GeV,

125: depending on the actual DM halo profile. These results are complementary to those

126: obtained in a previous analysis based on cosmic antiprotons.  For a Moore DM 

127: profile, $\gamma$--ray data limit the maximal Hubble rate enhancement to be below a

128: factor of 100. Moreover, a WIMP heavier than 1 TeV is not compatible with a

129: cosmological scenario with enhanced expansion rate prior to Big Bang Nucleosynthesis

130: (BBN). Less steep DM  profiles provide less stringent bounds, depending of the

131: cosmological scenario.

132: \end{abstract}

133: 

134: \pacs{95.35.+d,95.36.+x,98.80.-k,04.50.+h,96.50.S-,98.70.Sa,98.80.Cq}

135: % 95.35.+d DM 

136: % 95.36.+x Dark energy

137: % 98.80.-k Cosmology

138: % 04.50.+h Gravity in more than four dimensions, KK theory, unified

139: %          field theories, alternatives theories of gravity

140: % 96.50.S- Cosmic rays in the solar system

141: % 98.70.Sa Cosmic rays

142: % 98.80.Cq Particle theory and field theory models in the early Universe

143: 

144: \maketitle

145: 

146: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

147: \section{Introduction}

148: \label{sec:intro}

149: 

150: In a recent paper \cite{pbpaper} we discussed the possibility to derive limits to the

151: expansion rate of the Universe around the time when cold dark matter (CDM) decouples

152: from the thermal bath, by using the fact that today these dark matter (DM) particles form

153: the halo of our Galaxy and they may annihilate producing antiprotons. 

154: The measured antiproton flux has been shown to be compatible with the expected background

155: originated by standard cosmic--ray spallation: this fact leads to the possibility to

156: use antiprotons as a powerful tool for constraining DM properties (see

157: {\em e.g.} Refs. \cite{pbar,pbarlight}). By assuming that DM is in the

158: form of a generic Weakly Interacting Massive Particle (WIMP) candidate, 

159: we obtained relatively stringent bound on the Universe

160: dynamics in a period prior to the primordial nucleosynthesis phase (preBBN), which

161: is  not directly constrained by other observations. We showed that, despite

162: the large uncertainties in the knowledge of the  galactic propagation of antiprotons

163: \cite{pbar}, bounds on the Hubble rate enhancement ranging  from a factor of a few

164: to a factor of 30 are present for DM masses lighter than 100 GeV, while for a mass of

165: 500 GeV the bound falls  in the range 50--500. These bounds loosen for heavier 

166: DM particles. 

167: 

168: We remind here that the possibility to set a bound to the enhancement of the Hubble expansion  rate

169: in the early Universe by means of WIMP indirect detection signals relies on the fact that a larger

170: Hubble rate induces an anticipated DM decoupling and an ensuing larger relic abundance, for a

171: fixed annihilation cross section. This may happen, for instance, in scalar--tensor cosmology

172: \cite{scalartensorrelic}, in  quintessence models with a kination phase \cite{kination}, and for

173: anisotropic expansion and other models of modified expansion \cite{Barrow,kamionkowskiturner}. 

174: 

175: When the decoupling is anticipated, a DM relic abundance able to explain the current observational

176: data is obtained for larger DM annihilation cross sections, as compared to the standard

177: cosmological case. This means that larger indirect detection signals (which are proportional to the

178: DM  annihilation cross section) are predicted: the larger the enhancement of the Hubble rate, the

179: larger the indirect detection signal \cite{scalartensorrelic}. Comparison with experimental data

180: reflects in limits on the DM possible Hubble rate enhancement. The actual results depend also on the

181: specific cosmological model:  in Ref. \cite{pbpaper} we considered a general parametrization of the

182: Hubble rate temperature dependence, and for definiteness we studied different classes of models: a

183: Randall--Sundrum brane cosmology scenario \cite{Randall}, a kination scenario \cite{kination}, a

184: scalar--tensor cosmology scenario \cite{scalartensorrelic} and a simple case where the Hubble rate is

185: just boosted by a constant factor. The same modellizations will be considered  in the present

186: study.

187: 

188: In this paper we extend our analysis by including also the possibility to use

189: the data on $\gamma$--rays coming from the GC.  DM particles may

190: annihilate also to photons and the current experimental observations on the

191: $\gamma$--ray flux, especially the ones coming from the GC where a large

192: concentration of DM is expected, may be useful for our scope. With

193: $\gamma$--rays we have to face the large uncertainty coming from the DM

194: halo profile toward the GC, which is largely unknown. 

195: Nevertheless we will show that for typical DM halo

196: profiles, bounds to the preBBN Hubble rate may be set. 

197: 

198: The experimental data we will  use are those from the GC obtained by the HESS

199: \cite{hess2004,hess2006} and EGRET \cite{egret} telescopes. Since the HESS data refer to large

200: energies (above few hundreds of GeV),   we will be able to derive bounds for large-mass DM (above few

201: hundreds of GeV): this nicely complements our previous results on antiprotons \cite{pbpaper}, which

202: were quite stringent for low mass WIMPs, but loose for heavy ones. We will also show that EGRET

203: data in the GeV region are able to set limits in

204: the intermediate WIMP mass range (at least for the most steep DM profiles). 

205: 

206: Our analysis will be

207: independent of the specific DM candidate: the only assumption is that the WIMP we are considering is

208: responsible for the observed DM amount \cite{wmap}: 

209: %

210: \be

211: 0.092\le\Omega_{\rm CDM } h^2 \le 0.124

212: \label{oh2 constraint}

213: \ee

214: %

215: where $\Omega_{\rm CDM }$ is the density parameter of CDM, and its annihilation cross

216: section is dominantly temperature--independent (or $s$--wave). Deviations from this situation will

217: lead to changes similar to those discussed in Ref. \cite{pbpaper} for the antiproton case, to which

218: we refer for additional discussion.

219: 

220: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%5

221: \section{Models with increased pre-BBN expansion}

222: %

223: \begin{figure}[t] \centering

224: \vspace{-20pt}

225: \includegraphics[width=1.0\columnwidth]{fig/fig1.eps}

226: \vspace{-20pt}

227:   \caption{Different models for the Hubble rate enhancement function $A(T)$, 

228:   defined in Eq.~(\ref{A def}). The Figure is taken from 

229:   Ref.~\cite{pbpaper}. Notice that the evolution of the Universe runs from right to

230:   left. The solid (red) line has a slope parameter $\nu = -1$, the

231:   dashed (green) line $\nu = 0 $, the dash-dotted (blue) $\nu = 1$ and the

232:   dotted (purple) line has $\nu = 2$. All curves refer to 

233:   $\eta = 10^2$, $T_\textrm{re} = 10^{-2}$ GeV and 

234:   $T_\textrm{f} = 17.3$ GeV (this freeze--out temperature refers, {\rm e.g.}, 

235:   to a particle with mass of 500 GeV and annihilation cross section

236:   $\langle\sigma_{\textrm{ann}}v\rangle =

237:   10^{-7}\,\textrm{GeV}^{-2}$).}

238: \label{fig:enhancem_func}

239: \end{figure} 

240: 

241: As mentioned in Sect. \ref{sec:intro}, it is a common feature of some  

242: cosmological models to predict that the expansion rate $H(T)$

243: in the early Universe is larger than the Hubble expansion rate $H_\textrm{GR}(T)$ 

244: in standard cosmology. Quite generically, 

245: we may introduce a function $A(T)$ to quantify the enhancement of the Hubble rate:

246: %

247: \bea

248: \label{h.Eq.a.h.gr}

249: H(T)&=&A(T)H_{\textrm{{GR}}}(T)\qquad\qquad\textrm{at early times}\\

250: H(T)&=&H_{\textrm{{GR}}}(T)\qquad\qquad\qquad\, \textrm{at later~times}. 

251: \eea

252: %

253: In Ref.~\cite{pbpaper} we introduced a parametrization of the enhancement function 

254: $A(T)$ which was shown to be applicable for important models like some scalar-tensor 

255: gravity models, some models with a kination phase and also some specific brane-world 

256: model. In this paper, we are going to consider the same parametrization, {\em{i.e.}}:

257: %

258: \be

259: A(T)=1+\eta\left(\frac{T}{T_\textrm{f}}\right)^\nu

260:   \tanh\left(

261:  \frac{T-T_{\textrm{re}}}{T_{\textrm{re}}}\right)

262: \label{A def}

263: \ee

264: %

265: for temperatures $T>T_\textrm{re}$ and $A(T) = 1$ for $T \leq T_\textrm{re}$. The 

266: enhancement function is shown in Fig.~\ref{fig:enhancem_func} for some 

267: specific parameter choices. The hyperbolic tangent serves 

268: to assure that $A(T)$ goes continuously to 

269: ``1", and $H\into H_\textrm{GR}$, before some ``\emph{re-entering}" temperature, 

270: $T_\textrm{re}$. We must require $T_\textrm{re}\gsim 1 \, \textrm{MeV}$ to make 

271: sure not to be in conflict with the predictions of BBN and 

272: the formation of the Cosmic Microwave Background (CMB). For $T \gg T_\textrm{re}$ 

273: (and $\eta \gg 1$) we have approximately that:

274: \be

275:  A(T) \sim \eta \left(\frac{T}{T_\textrm{f}}\right)^\nu.

276: \ee

277: Thus, $T_\textrm{f}$ is the normalization temperature at which $A(T_\textrm{f}) 

278: = \eta$. As in Ref.~\cite{pbpaper} we take $T_\textrm{f}$ to be the temperature at 

279: which the WIMP DM candidate freezes out in standard cosmology. This means that 

280: $\eta$, as defined in our parametrization, is the enhancement of the Hubble rate at 

281: the time of the WIMP freeze--out. We will derive our results as bounds on $\eta$ for different cosmological models, characterized by the temperature--evolutionary parameter $\nu$: $\nu = 2$ refers to the Hubble rate evolution in a

282: Randall--Sundrum type II brane cosmology scenario of Ref. \cite{Randall};  

283: $\nu = 1$ is the typical kination evolution, discussed {\em e.g.} in Ref. \cite{kination});

284: $\nu = -1$ is representative of the behavior found in scalar--tensor cosmology in

285: Ref. \cite{scalartensorrelic}. The trivial case $\nu=0$ refers to an overall boost

286: of the Hubble rate.

287: 

288: For more details on the modified cosmological scenarios, the calculation of

289: the relic abundance in these models, including some analytical results

290: and discussion, we refer the reader to Ref. \cite{pbpaper}.

291: 

292: 

293: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

294: \section{The $\gamma$--ray signal from the GC}

295: \label{sec:gamma}

296: %

297: \begin{table*}[t]

298: \begin{tabular}{|c|c|c|c|c|} \hline

299: Isothermal &  NFW &  log--slope  & $\alpha=1.2$ & Moore  \\ 

300: \hline 18.9 & 6892 & 10229 & 98743 & $7.7\cdot 10^6$ \\ \hline

301: \end{tabular}

302: \caption{Values for $I_{\Delta\psi}$ in Eq.(\ref{eq:geometry}) (in units of

303: GeV$^2$ cm$^{-6}$ kpc). See text for details. }

304: \label{table:geometry}

305: \end{table*}

306: %

307: 

308: 

309: The most recent observations of the GC have been performed by the HESS

310: Collaboration in  the hundreds of GeV - few tens of TeV energy range.  In Refs.

311: \cite{hess2004,hess2004_web,hess2006} they have reported on the spectrum of very high energy

312: $\gamma$-rays from a point-like source in the GC,  with an unprecedented spatial

313: resolution, going down to a solid angle of  about $10^{-5}$ sr.  The HESS spectrum of the

314: central $\gamma$-ray source exhibits a clear power-law shape, with a spectral index  of

315: 2.2\cite{hess2004}-2.25\cite{hess2006}. Diffuse $\gamma$-ray emission extended along the

316: galactic plane has been reported by the same Collaboration in Ref. \cite{hess_diffused}. 

317: The measured spectrum follows a power-law with spectral index near to 2.30 and has been

318: shown to be compatible with a source of locally accelerated protons interacting with 

319: giant molecular clouds which are extended both in longitude and in latitude. 

320: This diffused component  contributes to the central source only for a small

321: fraction (10-15\%) \cite{hess_diffused,hess2006}.  

322: \\

323: The GC hosts more than one potential  sources of $\gamma$-rays, whose nature  is still not

324: clear. The most motivated astrophysical sources rely on particle  acceleration near the

325: supermassive black hole Sgr A$^*$ located at the center of our Galaxy, or in the region of

326: the supernova remnant Sgr A.  The $\gamma$ radiation is produced from accelerated  charged

327: particles (mostly protons) interacting with the ambient matter or radiation. Another

328: intriguing possibility resides in the $\gamma$-ray emission  resulting from the

329: annihilation of DM particles.  Cosmological simulations of hierarchical

330: structure formation predict  a significant density cusp in the central parts of the

331: galaxies.  In that region annihilation of DM  particles would be strongly enhanced,  with

332: extraordinary high expected  fluxes for the annihilation products, such as 

333: $\gamma$-rays.  The angular region explored by HESS is of the same order of the one of the

334: probable black hole, or of the DM  cusp.

335: 

336: We consider here a generic WIMP which composes the galactic DM. 

337: The flux  $\Phi_\gamma(E_\gamma, \psi)$ of $\gamma$-rays of energy  

338: $E_\gamma$ originated from the WIMP pair annihilation and

339: coming from the angular direction $\psi$ is given by:

340: %

341: \begin{equation}

342: \Phi_\gamma(E_\gamma, \psi) = \frac{1}{4\pi} \frac{\langle\sigma

343:  v\rangle_0}{m_\chi^2} \frac{dN_\gamma}{d E_{\gamma}}

344: \frac{1}{2}I(\psi)

345: \label{eq:flux_gamma}

346: \end{equation}

347: %

348: where $\langle\sigma  v\rangle_0$ is the present annihilation cross section times the relative

349: velocity averaged over the galactic velocity distribution function.

350: $dN_\gamma/d E_{\gamma}$ is the energy spectrum of $\gamma$-rays

351: originated from a single WIMP pair annihilation and has been

352: calculated by means of a Monte Carlo simulations with the PYTHIA

353: package \cite{pythia} as described in Ref. \cite{indirect_light}.

354: For definiteness, as we have done also for the antiproton analysis

355: of Ref. \cite{pbpaper}, we are assuming the $\gamma$--ray energy spectrum originated

356: by a $\bar{b}b$ quark pair. A different annihilation final state will

357: not change substantially our results, much less than the astrophysical

358: uncertainties.

359: 

360: The quantity $I(\psi)$ is the contribution of the squared DM 

361: density distribution along the line of sight (l.o.s.):

362: %

363: \begin{equation}

364: I(\psi) = \int_{\rm l.o.s} \rho^2(r(\lambda,\psi))~ d\lambda(\psi).

365: \label{eq:los}

366: \end{equation}

367: %

368: Here $\psi$ is the angle between the l.o.s. and the line

369: pointing toward the GC ($\cos\psi = \cos l \cos b$, $l$ and 

370: $b$ being the galactic longitude and latitude, respectively). 

371: If Eq.(\ref{eq:los}) is used for comparison with experimental data, 

372: it must be averaged over the telescope observing angle $\Delta \psi$:

373: %

374: \beq I_{\Delta\psi} =

375: \frac{1}{\Delta\psi}\int_{\Delta\psi} I(\psi)~d\psi.

376: \label{eq:geometry}

377: \eeq

378: 

379: The geometric factor $I(\psi)$ depends quadratically 

380: on the DM  density profile and is very sensitive to 

381: its features, especially in the GC region where predictions 

382: for $\rho(\vec r)$ differ mostly. 

383: The most common spherically--averaged density profiles can be 

384: parametrized as:

385: \begin{equation}

386: \rho(r) = \rho_l \left(\frac{R_\odot}{r}\right)^\gamma

387: \left[\frac{1+(R_\odot/a)^\alpha}{1+(r/a)^\alpha}\right]^{(\beta

388: -\gamma)/\alpha},  

389: \label{eq:density}

390: \end{equation}

391: %

392: where $r=|\vec r|$, $R_\odot=8$ kpc  is the distance of the

393: Solar System from the GC along the galactic plane, $a$ is a scale

394: length and $\rho_l$ is the total local (solar neighborhood) DM density. 

395: In particular, the isothermal, cored density profile

396: is obtained with $(\alpha, \beta, \gamma)$ = (2,2,0), the Navarro, Frenk

397: and White (NFW) profile \cite{nfw} with $(\alpha,

398: \beta,\gamma)$ = (1,3,1) and the Moore et al.  profile \cite{moore} with

399: $(\alpha,\beta, \gamma)$ = (1.5,3,1.5). 

400: We also consider a non--singular DM 

401: density distribution function derived from extensive numerical 

402: simulations, whose asymptotic regime is well fitted by a 

403: logarithmic slope \cite{navarro}:

404: %

405: \begin{equation}

406:  \rho(r) = \rho_{-2} \; \exp\left\{ -\frac{2}{\alpha}

407:  \left[\left(\frac{r}{r_{-2}}\right)^\alpha -1\right] \right\},

408:  \label{eq:alpha}

409: \end{equation}

410: % 

411: where $r_{-2}$ is the radius where the logarithmic slope is $\delta = -2$, and $\rho_{-2}

412: \equiv \rho(r_{-2})$.  The DM  density predicted by these profiles at the GC (for very

413: small $r$) varies so strongly that  also the predicted DM  signals  may  differ by

414: several orders of magnitude. 

415: The profile as steep as 1.5 is disfavored by the most

416: recent numerical simulations, which seem to indicate a power law index  not

417: exceeding 1.2 \cite{diemand04,diemand04_0}. We notice, however, that 

418: the experiments considered in this paper have a spatial resolution which 

419: is much narrower that the typical resolution size of numerical simulations. 

420: The simulated DM  densities in the GCs are thus mere extrapolations. 

421: Moreover, we must be aware of the fact that the results from many rotational curves for

422: galaxies of different morphological types are hardly explained by central

423: DM  cusps. Instead, they are more easily fitted by cored DM  distributions,

424: flattened towards the central region of the galaxy. 

425: 

426: In Table \ref{table:geometry} we show the values of the geometrical factor

427: $I_{\Delta\psi}$ for the HESS telescope aperture  ($\Delta\psi=10^{-5}$ sr) and for

428: various DM  density profiles.  The first column refers to an isothermal density

429: distribution with a core  $a=3.5$ kpc and the second to a NFW with a scale length $a=25$

430: kpc. The third is for the log-slope of Eq. (\ref{eq:alpha}) with the parameters of the 

431: distribution G1 in Ref. \cite{navarro}:  $\alpha=0.142$, $r_{-2}=26.4$ kpc and

432: $\rho_2=0.035$ GeV cm$^{-3}$.  The fourth column is the result for a profile obtained

433: with  $(\alpha,\beta, \gamma)$ = (1.2,3,1) with $a$=25 kpc \cite{diemand04,diemand04_0}

434: and  the last column refers to  a Moore et al. profile with scale  length $a=30 $ kpc. 

435: %

436:  The value of $\rho_l$ can be determined for each density profile requiring the

437: compatibility with the measurements of rotational curves and the total mass of the Galaxy

438: \cite{bcfs}.  For definiteness, we have fixed  $\rho_l = 0.3$ GeV cm$^{-3}$ for all the

439: density profiles in Table  \ref{table:geometry}. We notice that the parameter $\rho_l$

440: enters  as a mere scaling factor in Eq. \ref{eq:flux_gamma}: the effect of varying

441: $\rho_l$ is therefore easily taken into account. 

442: 

443: 

444: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

445: 

446: 

447: 

448: \section{Constraining the Hubble rate with $\gamma$--rays}

449: \label{sec:constraints}

450: 

451: In order to derive the constraint on the enhancement of the Hubble rate, we first 

452: find the upper bound on $\langle \sigma v\rangle_0$ determined by the $\gamma$--ray observations, for any given halo profile.

453: 

454: Let us first explain our analysis for the HESS observation of high 

455: energy $\gamma$--rays from the GC. Here we have added the expected 

456: WIMP signal to a background that follows a power-law, $kE^{-\Gamma}$. 

457: The normalization and the index of the power-law are taken as free parameters. 

458: For each point in the grid--scan of $(k,\Gamma)$ we find the maximum allowed value 

459: of $\langle \sigma v\rangle_0$, which we statistically define as the value of the cross 

460: section at which the reduced $\chi^2$ equals 3. 

461: The reduced $\chi^2$ is defined as:

462: \begin{equation}

463:   \chi^2_{\textrm{red}} = \frac{1}{N}\sum_{n=1}^D \left(

464:           \frac{\Phi_\chi(n) + \Phi_B(n) -\Phi_\textrm{Obs}(n)}

465: 			{\sigma_\textrm{Obs}(n)}\right)^2

466: \end{equation}

467: %

468: where $N$ is the number of degrees of freedom, $\Phi_\chi(n)$ is the expected WIMP  flux,

469: as calculated using Eq.~(\ref{eq:flux_gamma}), at the energy of the $n$--th data-point 

470: ($\Phi_\chi(n)$ obviously vanishes when $E_n \gsim m_\chi$). The background flux is 

471: assumed to be $\Phi_B(n) = kE_n^{-\Gamma}$, while the observed flux is 

472: $\Phi_\textrm{Obs}(n)$. The 1--$\sigma$ error of the observation is denoted 

473: $\sigma_\textrm{Obs}(n)$. For the analysis, we use the 17

474: data points from the HESS 2003  observation  of the GC \cite{hess2004,hess2004_web}. They

475: cover the energy  range 280 GeV -- 8.83 TeV. These data can therefore be used to

476: constrain WIMP heavier than about 300 GeV (the annihilation process in the Galaxy

477: occurs almost at rest: therefore there is a kinematic cut off at the WIMP mass for the

478: $\gamma$--ray energy). 

479: 

480: \begin{figure}[t] \centering

481: \vspace{-20pt}

482: \includegraphics[width=1.0\columnwidth]{fig/allsigmalim.eps}

483: \vspace{-20pt}

484: \caption{Bounds on the WIMP annihilation cross section  $\langle \sigma v\rangle_0$ as a

485: function of the WIMP mass. The curved lines at large WIMP mass show the {\em upper}

486: bounds derived from the HESS  observation of $\gamma$--rays from the GC. The derivation

487: has been  made for five different DM halo profiles. From top to bottom these are: the 

488: isothermal model, the NFW model, the 'log' slope, a power--law slope with index 1.2, the

489: Moore et al. profile (see text for more details).  The dashed lines show analogous {\em

490: upper} limits  derived using the $\gamma$--ray data from the EGRET detector, for the same

491: set of galactic halo models. The EGRET limits are plotted only in the mass interval which

492: is  relevant for the analysis of this paper. The slanted solid lines show the {\em upper}

493: limits coming from the observations of cosmic antiprotons \cite{pbpaper}.  The central

494: line refers to the best estimate for the antiproton DM signal. The upper and lower lines

495: refer to the astrophysical uncertainties in the galactic propagation parameters

496: \cite{pbpaper,pbar}. Finally, the horizontal dot--dashed lines shows the \emph{lower} 

497: bound on the cross section as derived from the WIMP relic density constraint assuming 

498: the standard cosmological model and a temperature independent WIMP annihilation cross

499: section.}

500: \label{fig:sigmalim_all}

501: \end{figure} 

502: %

503: \begin{figure}[t] \centering

504: \vspace{-20pt}

505: \includegraphics[width=1.0\columnwidth]{fig/spec.eps}

506: \vspace{-20pt}

507: \caption{$\gamma$--ray spectra as a function of energy. The data points refer to HESS \cite{hess2004,hess2004_web}. The solid and dashed (green) curves refer to the maximal allowable

508: contribution to the $\gamma$--ray flux from WIMP annihilation: the cases of 500 GeV and 5 TeV

509: WIMPs are plotted. The solid and dashed (red) straight lines refer to a standard--source

510: power--law contribution to the HESS data, as obtained by our fit (see text for more details).}

511: \label{fig:spec}

512: \end{figure}

513: %

514: \begin{figure}[t] \centering

515: \vspace{-20pt}

516: \includegraphics[width=1.0\columnwidth]{fig/combsiglim.eps}

517: \vspace{-20pt}

518: \caption{Summary of the DM indirect detection limits on the

519: WIMP annihilation cross section (for a temperature independent cross section).

520: The allowed region lies above the horizontal dot--dashed lines (which refer to

521: the relic density constraint) and below the slanted/curved lines, for any

522: given DM density profile. The slanted part of the upper bound is due to

523: cosmic antiprotons, while the curved part to $\gamma$--rays from the GC

524: (from HESS data at large WIMP masses, EGRET at intermediate masses).

525: }

526: \label{fig:sigmalim_combined}

527: \end{figure}

528: 

529: 

530: The $\chi^2$ analysis, as described above, gives us the upper limit of the cross 

531: section for each point in the $(k,\Gamma)$ grid for fixed WIMP mass and halo 

532: profile. Finally we extract the grid-point which gives the biggest value of the 

533: cross section. This upper bound on $\langle \sigma v\rangle_0$ is shown in 

534: Fig.~\ref{fig:sigmalim_all}, as a function of the WIMP 

535: mass for the five different halo profiles discussed above. 

536: As expected, the result depends strongly on the halo profile. 

537: 

538: Before we continue to explain how the upper limit on the cross section was 

539: derived from other observational data, let us show some examples of the 

540: differential photon production which correspond to the upper limit on the 

541: WIMP annihilation cross section. In Fig. \ref{fig:spec} we show the result for 

542: the NFW halo profile and for two different masses. The flux is calculated using 

543: Eq.~(\ref{eq:flux_gamma}) and inserting the upper bound on $\langle \sigma v\rangle_0$ for 

544: the given mass and halo profile. Also shown are the HESS 2003 observations of 

545: the GC as well as the fitted power-law background. Note that as the 

546: parameters of the background are treated as free parameters, they are different 

547: for the different WIMP masses and halo profiles. In the shown examples, the 

548: background parameters associated with the upper bound on the WIMP cross section 

549: are $(k,\Gamma) = (0.177 \cdot 10^{-6}~\mbox{cm$^{-2}$ s$^{-1}$ TeV$^{-1}$ sr$^{-1}$}, 2.00)$ and 

550: $(k,\Gamma) = (0.133 \cdot 10^{-6}~\mbox{cm$^{-2}$ s$^{-1}$ TeV$^{-1}$ sr$^{-1}$}, 1.83)$ for 

551: WIMP masses of 500 GeV and 5 TeV respectively.

552: 

553: Coming back to the limit on the WIMP annihilation cross section,

554: Fig. \ref{fig:sigmalim_all} shows also the upper limit

555: as derived from the EGRET. The EGRET data \cite{egret} span 

556: from energies of around 0.039 GeV to around 14.9 GeV with an 

557:  angular resolution given by the

558: longitude--latitude aperture: $|\Delta l| \leq 5^\circ$, $|\Delta b|

559: \leq 2^\circ$. 

560: The geometric factor for the EGRET experiment has been taken from Ref.

561: \cite{indirect_light}, for the same density distribution functions described in 

562: the previous Section. At these energies there is 

563: a $\gamma$--ray background from nucleonic reactions between cosmic rays and interstellar medium, 

564: electron bremsstrahlung and inverse Compton. We assume it to be the same as 

565: the one described in Ref. \cite{indirect_light}. 

566: We apply the following analysis only for the two highest EGRET energy bins, 

567: i.e.~$E_n \sim 6.2$ GeV and $E_n \sim 14.9$ GeV, which are the most constraining

568: for the masses we are dealing with. For a given WIMP mass and halo 

569: profile we make a scan in the WIMP annihilation cross section to find its upper 

570: limit taken to be the $2\sigma$ bound:

571: \be

572:  \Phi_\chi(n) + \Phi_B(n) -\Phi_\textrm{Obs}(n) \leq 

573: 			2 \, \sigma_\textrm{Obs}(n)

574: \ee

575: The observational error, that we use, only includes the statistical error.

576: We use the limit from the most constraining of the two data points. For the 

577: mass range relevant here, it is always the highest energy bin which provides

578: the bound, 

579: except for a WIMP mass of 100 GeV. In Fig. \ref{fig:sigmalim_all} we

580: show only the WIMP range where the EGRET data could be of importance for the 

581: analysis of this paper. 

582: Finally, Fig. \ref{fig:sigmalim_all} shows, as slanted solid lines, the upper limit on 

583: $\langle \sigma v\rangle_0$ derived in our previous analysis \cite{pbpaper} of the observational data on cosmic antiprotons.

584: The central line refers to the best estimate for the antiproton DM

585: signal, {\em i.e} when the best set of astrophysical parameters are used in

586: the calculation of the diffusion processes for the galactic cosmic rays \cite{pbar}.  

587: The upper and lower lines refer to the uncertainty band arising from astrophysical uncertainties

588: in the galactic propagation parameters \cite{pbpaper,pbar}. The horizontal dot--dashed

589: line denote instead the {\em lower} limit on the annihilation cross section

590: derived from the cosmological bound on the WIMP relic density Eq. (\ref{oh2 constraint}) in standard cosmology.

591: 

592: 

593: The analysis of Fig. \ref{fig:sigmalim_all} shows interesting properties. In addition to

594: the already discussed bound on $\sigmavz$ from antiprotons, which sizeably constrains

595: the maximal allowable annihilation cross section, especially for WIMPs lighter than

596: a few hundreds of GeV, we now also have the bounds coming from $\gamma$--rays from the GC, which instead are relevant for heavy WIMPs. This is a consequence of the fact that HESS data refer to

597: an energy range from a few hundreds of GeV up to few TeV. The figure clearly shows that, in order to have

598: bounds on $\sigmavz$, the signal must be quite sizeable and this is possible only for very

599: steep DM  profiles like the Moore and NFW ones. In the case of the Moore profile, we have

600: a tension between the $\gamma$--rays HESS data and the cosmological limit even for standard

601: cosmology (a situation analogous to the one already observed for the antiproton signal

602: produced by light WIMPs \cite{pbpaper,pbarlight}). Should the DM  profile be the Moore

603: one, very little room would be allowed for $\sigmavz$. This bound would also imply a finite

604: possible range for the WIMP mass: from 50--100 GeV to about 1 TeV (even for standard cosmology),

605: as a combination of cosmological data on the amount of DM, the antiproton component

606: of cosmic--rays and the $\gamma$--ray signal for the GC. Less steep profiles

607: are clearly much less constraining, and the isothermal case is always ineffective, being always less relevant than the antiproton bound. We remind that since antiprotons diffuse in the

608: Galaxy, their flux is only mildly dependent on the DM  profile \cite{pbar}: the antiproton bounds

609: are therefore practically unaffected by the choice of different halo shapes.

610: 

611: A summary of the different bounds derived in our analysis is given in Fig.

612: \ref{fig:sigmalim_combined}, where we combine the upper limits on the WIMP  annihilation cross

613: section from the observations of the cosmic antiprotons and  of the GC $\gamma$--rays, as observed by

614: both EGRET and HESS. For any mass and halo profile we take the bound which is most constraining. The

615: combined upper  limit can be seen in Fig. \ref{fig:sigmalim_combined} for three halo  profiles.

616: 

617: \section{The maximal enhancement of the Hubble rate}

618: 

619: \begin{figure}[t] \centering

620: \vspace{-20pt}

621: \includegraphics[width=1.0\columnwidth]{fig/mvetamax.eps}

622: \vspace{-20pt}\caption{Upper bound on the Hubble rate enhancement--parameter $\eta$.

623: The bound is 

624: shown as a function of the WIMP mass and has been derived by combining the 

625: constraints on the WIMP relic density with the constraints derived from the

626: observations of cosmic antiprotons and GC $\gamma$--rays. 

627: This figure shows the result for $T_\textrm{re} = 10^{-3}$ and $\nu = 2$ 

628: (RSII brane cosmology scenario \cite{Randall}), 

629: where $\nu$ is defined in Eq.~(\ref{A def}) for the Hubble rate enhancement 

630: function. We show the result for three different halo models. The labels on the

631: curves show the origin of the bound.}

632: \label{fig:etamax_nu2}

633: \end{figure} 

634: 

635: \begin{figure}[t] \centering

636: \vspace{-20pt}

637: \includegraphics[width=1.0\columnwidth]{fig/mvetamax_low.eps}

638: \vspace{-20pt}

639: \caption{The same as in Fig. \ref{fig:etamax_nu2}, for different values of the

640: parameter $\nu$: $\nu = 2$ (RSII brane cosmology scenario \cite{Randall}), 

641: $\nu = 1$ (kination scenario \cite{kination}),

642: $\nu = 0$ and $\nu = -1$ (scalar--tensor cosmology scenario \cite{scalartensorrelic}).

643: The NFW and Moore DM density profiles are shown.}

644: \label{fig:etamax_allnu}

645: \end{figure}

646: 

647: An increased Hubble rate in the early Universe would increase the relic density of a given 

648: WIMP as compared to the situation in standard cosmology. The increase of the 

649: relic density is due to an anticipated freeze--out of the WIMP, as the 

650: annihilation rate cannot keep up with the expansion rate as long as in the 

651: standard case. WIMPs which satisfy the density constraint of Eq.~(\ref{oh2 constraint}) 

652: in the modified scenario would thus be underabundant in standard cosmology. 

653: Because of the inverse proportionality between the WIMP relic density and 

654: annihilation cross section, the WIMPs which fulfill the density constraint in 

655: the modified cosmologies have larger cross section than those WIMPs which 

656: fulfill it in the standard case. The possible enhancement of the Hubble rate can 

657: therefore be constrained by applying at the same time the relic density 

658: constraint for the DM and the upper bound on the WIMP annihilation 

659: cross section as derived from the indirect searches for DM.

660: 

661: The argument above built on the crucial assumption that the WIMP annihilation 

662: cross section $\langle \sigma v\rangle$ is temperature independent. If this is 

663: not the case then the constraint from the indirect searches, which bound  

664: the present $\langle \sigma v\rangle_0$, could not be directly combined with the 

665: constraints on the relic density, which depends on the cross section in the early 

666: Universe. The relation between the cross section in the two epochs should be 

667: taken into account, a relation which would often lead to a looser bound on the 

668: enhancement of the Hubble rate. A discussion of this topic was given in Ref. \cite{pbpaper},

669: where it was shown that modifications are usually not very large, unless some

670: specific situations, like {\em e.g.} coannihilation, occur. In this paper we

671: will show our results only for the case where the WIMP annihilation 

672: cross section is temperature independent.

673: 

674: Let us derive the bound on the enhancement of the Hubble rate from 

675: the combination of the constraints on the WIMP relic density and cross section. 

676: To obtain the WIMP relic density we solve the Boltzmann equation implemented

677: with the modified Hubble function, Eq.~(\ref{A def}). Let us 

678: therefore again go through the free parameters of the enhancement function. The 

679: freeze--out temperature, $T_\textrm{f}$, is determined by the WIMP mass and 

680: annihilation cross section and is therefore not a free parameter. The 

681: {\em{re--entering}} temperature, $T_\textrm{re}$, is a free parameter, but we 

682: showed in Ref.~\cite{pbpaper} that the bound on the Hubble expansion is 

683: independent of $T_\textrm{re}$ as long as $T_\textrm{re} \ll T_\textrm{f}$. In this 

684: paper we set $T_\textrm{re} = 10^{-3}$ GeV, which is always much lower than the 

685: freeze--out temperature and which is the lowest value we can assume not

686: to spoil BBN predictions. As we have mentioned earlier, the exponent $\nu$ in the 

687: enhancement function selects the kind of cosmological model. The only true free 

688: parameter is therefore $\eta$, which is normalized as the enhancement of the Hubble 

689: function at the time where the WIMP freezes out in standard cosmology. 

690: 

691: To derive the upper bound on the parameter $\eta$ we use for each WIMP mass 

692: and halo profile the upper bound on the annihilation cross section  

693: displayed in Fig. \ref{fig:sigmalim_combined}. For a given cosmological model, 

694: determined by the value of the $\nu$ parameter, the upper value of $\eta$ is 

695: then found where the solution of the Boltzmann equation 

696: satisfies the upper bound of the density constraint Eq.~(\ref{oh2 constraint}). The upper 

697: bound on $\eta$ as a function of the WIMP mass is shown in 

698: Fig. \ref{fig:etamax_nu2} for $\nu = 2$ and in Fig.~\ref{fig:etamax_allnu} also 

699: for $\nu = -1,0,1$. Knowing $\eta$, we can calculate the enhancement function 

700: $A(T)$ at any time once we choose a cosmological model.

701: 

702: Figs. \ref{fig:etamax_nu2} and \ref{fig:sigmalim_combined} show that $\gamma$--ray data

703: may be quite effective in constraining the preBBN Hubble rate for heavy WIMPs, and nicely

704: complement in this large--mass range the antiproton results. In order to set bounds more

705: stringent than antiprotons, however, a steep density profile is required: in the case of

706: a NFW distribution, the $\gamma$--ray observations of HESS are able to set limits only

707: for WIMPs heavier than 1 TeV.  For this mass range, however, depending on the actual

708: cosmological evolution, the bound can be relevant and much stronger than the antiproton

709: bound: in the case of  kination models, the $\gamma$--rays predicted for a NFW profile

710: limits the maximal Hubble rate enhancement to be less than a factor of 5000; for

711: scalar--tensor cosmologies the maximal enhancement goes down to a factor of 500. On the

712: other hand, for a Moore profile the bounds are quite stringent: the maximal enhancement

713: of the Hubble rate in this case is a factor of 100. In addition, for the Moore profile an

714: enhancement is not possible for all the WIMP mass  exceeding 1 TeV. 

715: 

716: As an example of how our limits can be further used to constrain the basic properties of

717: specific cosmological models, let us consider the implications for the brane

718: Randall--Sundrum II model \cite{Randall}, which in our notations corresponds to 

719: $\nu = 2$ and $\eta = \sqrt{\rho_\textrm{r}(T_\textrm{f})/(2\lambda)}$. 

720: Here $\rho_\textrm{r}$ is the radiation energy density and 

721: $\lambda$ is the tension of the  brane, related to the 5--dimensional Planck mass 

722: $M_5$ by  the relation $\lambda = \frac{3}{4\pi}\frac{M_5^6}{M_\textrm{pl}^2}$.

723: For the Moore profile and WIMP masses $m_\chi = 

724: \mathcal{O}(500\,\textrm{GeV})$ we see from Fig.~\ref{fig:etamax_nu2} that $\eta 

725: < \mathcal{O}(10^2)$. 

726: This implies $\lambda \gsim 2\cdot10^2\,\textrm{GeV}^4$, which corresponds to a lower

727: bound on $M_5$ of $M_5 \gsim 7\cdot10^3$ TeV. This is almost two orders of magnitude

728: better than what was found in Ref. \cite{pbpaper} by using antiproton data for the same

729: value of WIMP mass. It is more stringent than what can be obtained from other

730: cosmological tests: BBN sets the limit $M_5 > 13$ TeV \cite{Scherrer,Durrer}, while 

731: Ref. \cite{Nihei} sets $M_5 \gsim 600$ TeV but for a DM candidate in a specific

732: supersymmetric model. 

733: Microgravity experiments \cite{Durrer} still set the best bound $M_5 > 10^5$ TeV.

734: 

735: 

736: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

737: 

738: \section{Conclusions}

739: \label{sec:conclusion}

740: 

741: In this paper we have discussed the possibility to use $\gamma$--ray data

742: from the GC to constrain the cosmological evolution of the Universe

743: in a phase prior to primordial nucleosynthesis, namely around the time of

744: CDM decoupling from the primeval plasma. We extended the arguments

745: already discussed in a previous paper of ours \cite{pbpaper}, where instead cosmic--ray

746: antiprotons were used. The basic idea is that in a modified cosmological scenario,

747: where the Hubble expansion rate is enhanced with respect to the standard case, the

748: CDM decoupling is anticipated and therefore the relic abundance of

749: a given DM  candidate is enhanced. This implies that the present amount of CDM  in

750: the Universe may be explained by a WIMP which possesses annihilation cross section

751: (much) larger than in standard cosmology. This enhanced annihilation cross section

752: implies larger fluxes of indirect detection signals of CDM, due to the annihilation

753: of relic WIMPs in the halo of our Galaxy.

754: 

755: The stringent bounds on the maximal enhancement of the preBBN cosmic evolution, 

756: determined by the antiproton signal from DM  annihilation and obtained in Ref.

757: \cite{pbpaper}, have been complemented here by $\gamma$--ray searches. We have shown that

758: the HESS measurements, which refer to relatively large $\gamma$--ray energies, are able

759: to set constraints for WIMPs heavier than a few hundreds of GeV, depending on the actual

760: DM halo profile. These results are complementary to those coming from antiprotons, which

761: instead are important for WIMPs lighter than a few hundreds of GeV. In the case of a

762: Moore profile, these bounds are very strong, and imply that a WIMP heavier than about 1

763: TeV is not compatible with a cosmological scenario with enhanced expansion rate prior to

764: BBN. Less steep profiles provide less stringent bounds, always for heavy WIMPs: the NFW

765: halo bounds the maximal Hubble rate enhancement to be below a factor between $5\times

766: 10^2$ and $5\times 10^5$, depending of the cosmological scenario. On the other hand, an

767: isothermal sphere does not provide any relevant limit (better than the antiproton bound)

768: for any mass. $\gamma$--ray data from the EGRET satellite are important for

769: intermediate--mass WIMPs, but only for the very steep Moore profile.

770: 

771: Data from the GLAST satellite--based experiment will add relevant information  for DM 

772: particles in a range of masses which goes from 100 GeV to a few hundreds of GeV,

773: furtherly complementing the analysis we have been able to perform by using antiprotons

774: (relevant for masses below 100-200 GeV) and available  $\gamma$--rays from the GC

775: (relevant for masses above about 500 GeV).

776: 

777: We can therefore conclude that DM indirect detection searches, in addition of being a powerful and

778: important tool for studying the DM  component of the Universe, may also have an important role in

779: constraining the cosmic evolution, with an impact on dark energy models, modified gravity scenarios

780: and theories of extra--dimensions.

781: 

782: 

783: 

784: 

785: 

786: \acknowledgments 

787: %

788: We thank Prof. A. Bottino for stimulating and fruitful discussions.

789: M.S. thanks P. Gondolo for useful and interesting comments. 

790: We acknowledge Research Grants funded jointly by the Italian Ministero

791: dell'Istruzione, dell'Universit\`a e della Ricerca (MIUR), by the

792: University of Torino and by the Istituto Nazionale di Fisica Nucleare

793: (INFN) within the {\sl Astroparticle Physics Project}.  

794: 

795: 

796: 

797: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

798: \begin{thebibliography}{99}

799: 

800: \bibitem{pbpaper}

801:  Schelke M, Catena R,  Fornengo N, Masiero A, Pietroni M, (2006) 

802: Phys. Rev. D {\bf 74} 083505.

803: 

804: \bibitem{pbar}

805:  Donato F, Fornengo N, Maurin D, Salati P and

806: Taillet R, (2004)  Phys. Rev. D {\bf 69} 063501. 

807: 

808: \bibitem{pbarlight} 

809: A. Bottino, F. Donato, N. Fornengo, P. Salati,

810: Phys. Rev. D {\bf 72} (2005) 083518. 

811: 

812: \bibitem{scalartensorrelic} 

813: R.~Catena, N.~Fornengo, A.~Masiero,

814: M.~Pietroni and F.~Rosati, Phys.\ Rev.\ D {\bf 70}, 063519 (2004)

815: 

816: 

817: \bibitem{kination}

818:   P.~Salati, Phys.\ Lett.\ B {\bf 571}, 121 (2003);

819:   F.~Rosati, Phys.~Lett.~B {\bf 570}, 5 (2003);

820:   S.~Profumo and P.~Ullio, JCAP {\bf 0311}, 006 (2003);

821:   C.~Pallis, JCAP {\bf 0510}, 015 (2005).

822: 

823: \bibitem{Barrow} 

824:   J.~D.~Barrow, Nucl.~Phys.~B {\bf 208}, 501 (1982).

825: 

826: \bibitem{kamionkowskiturner}

827:   M.~Kamionkowski and M.~S.~Turner, Phys.~Rev.~D {\bf 42}, 3310 (1990).

828: 

829: \bibitem{Randall}

830:   L.~Randall and R.~Sundrum,

831:   %``An alternative to compactification,''

832:   Phys.\ Rev.\ Lett.\  {\bf 83}, 4690 (1999).

833:   

834: \bibitem{hess2004}

835: F. Aharonian {\it et al.} (HESS Collaboration), Astron.Astrophys. 425 (2004) L13-L17

836: 

837: \bibitem{hess2006}

838: F. Aharonian {\it et al.} (HESS Collaboration), Phys. Rev. Lett. 97 (2006) 221102

839: 

840: \bibitem{egret}

841: S.D. Hunter {\it et al.} (EGRET Collaboration), Astrophys. J. {\bf 481}, 205 (1997).

842: 

843: \bibitem{wmap}

844: Spergel, D N {\it et al.} (WMAP Collaboration), [arXiv:astro-ph/0603449], subm. Astrophys. J. 

845: 

846: \bibitem{hess2004_web} 

847: Data points have been taken from the table on the Collaboration web site:

848: http://www.mpi-hd.mpg.de/hfm/HESS/HESS.html

849: 

850: \bibitem{hess_diffused}

851: F. Aharonian {\it et al.} (HESS Collaboration), Nature 439 (2006) 695. 

852: 

853: \bibitem{pythia}

854: T.~Sj\"ostrand, P.~Eden, C.~Friberg, L.~Lonnblad,

855: G.~Miu, S.~Mrenna and E.~Norrbin,

856: Comput.\ Phys.\ Commun.\  {\bf 135}, 238 (2001).

857: 

858: \bibitem{indirect_light}

859: A. Bottino, F. Donato, N. Fornengo, S. Scopel, 

860: Phys. Rev. {\bf D70} (2004) 015005.

861: 

862: \bibitem{nfw} 

863: J.F. Navarro, C.S. Frenk and S.D.M. White, Astrophys. J. {\bf 462}, 563 (1996).

864: 

865: \bibitem{moore} 

866: Moore B {\it et al.}, Mon. Not. Roy.  Astron. Soc. {\bf 310}, 1147 (1999). 

867: 

868: \bibitem{navarro} 

869: Navarro J F {\it et al.},  (2004)  MNRAS {\bf 349} 1039. 

870: 

871: \bibitem{diemand04}

872: Diemand J, Moore B, Stadel J, (2004) MNRAS {\bf 352}  535.

873: 

874: \bibitem{diemand04_0}

875: Diemand J, Moore B, Stadel J, (2004) MNRAS {\bf 353}  624.

876: 

877: \bibitem{bcfs} 

878: P. Belli, R. Cerulli, N. Fornengo and S. Scopel,

879: Phys. Rev. D {\bf 66}, 043503 (2002).

880: 

881: \bibitem{Scherrer}

882: Bratt, J D, Gault, A C, Scherrer, R J, and Walker, T P, 

883: (2002) Phys. Lett. {\bf 546} 19. 

884: 

885: 

886: \bibitem{Durrer}

887:   R.~Durrer,   %``Braneworlds,''

888:   AIP Conf.\ Proc.\  {\bf 782}, 202 (2005)

889:   [arXiv:hep-th/0507006].

890:   

891: \bibitem{Nihei}

892: Nihei, T, Okada, T, and Seto, O, (2005) Phys. Rev. {\bf D71} 063535

893:   

894: \end{thebibliography}

895: 

896: \end{document}

897: 

898: 

899: