1: \documentclass[12pt,preprint]{aastex}

2: 

3: %% manuscript produces a one-column, double-spaced document:

4: 

5: %% \documentclass[manuscript]{aastex}

6: 

7: %% preprint2 produces a double-column, single-spaced document:

8: 

9: %% \documentclass[preprint2]{aastex}

10: 

11: %% Sometimes a paper's abstract is too long to fit on the

12: %% title page in preprint2 mode. When that is the case,

13: %% use the longabstract style option.

14: 

15: %% \documentclass[preprint2,longabstract]{aastex}

16: 

17: 

18: %\shorttitle{Absence of the Fundamental Plane}

19: \shortauthors{Bregman}

20: 

21: %% This is the end of the preamble.  Indicate the beginning of the

22: %% paper itself with \begin{document}.

23: 

24: \begin{document}

25: 

26: %% LaTeX will automatically break titles if they run longer than

27: %% one line. However, you may use \\ to force a line break if

28: %% you desire.

29: 

30: \title{The Absence of the Fundamental Plane of Black Hole Activity}

31: 

32: \author{Joel N. Bregman}

33: 

34: \affil{Department of Astronomy, University of Michigan, Ann Arbor, MI 48109}

35: \email{jbregman@umich.edu}

36: 

37: \begin{abstract}

38: An analysis of the radio and X-ray luminosities, along with black hole masses has led to a relationship, the

39: fundamental plane of black hole activity, where logL$_R$ = 0.60logL$_X$ + 0.78logM$_{BH}$.  We show

40: that this same relationship can be obtained by using upper limit data or by randomizing the radio

41: fluxes.  In those cases, the relationship arises because one is effectively plotting distance vs

42: distance in a flux-limited sample of objects.  To correctly establish a relationship between L$_R$,

43: L$_X$, and M$_{BH}$, one would need to analyze a volume-limited sample of objects.

44: 

45: The distance effect can be removed from the sample, where a relationship

46: between L$_R$/L$_X$ and M$_{BH}$ is found, showing that the L$_R$ rises more quickly than L$_X$ with

47: increasing M$_{BH}$.  However, until a well-define sample is used, it is unclear the degree to which

48: this relationship is influenced by the sample-selection.

49: \end{abstract}

50: 

51: \keywords{black hole physics --- galaxies: nuclei --- X-rays: binaries}

52: 

53: \section{Introduction}

54: 

55: Fundamental relationships between observed astrophysical quantities have been of

56: tremendous value in advancing understanding.  Some examples are the H-R diagram for stars or

57: the fundamental plane for early-type galaxies.  Recently, another such relationship has been

58: advanced, the fundamental plane of black hole activity \citep{merl03, falc04}.  These

59: authors have considered both supermassive black holes in the centers of galaxies as well as

60: stellar mass black holes in the Milky Way.  They sought and found a relationship between the

61: radio luminosity (L$_R$), X-ray luminosity (L$_X$), and black hole mass (M$_{BH}$).  

62: The specific relationship found by Merloni is logL$_R$ = 0.60logL$_X$ + 0.78logM$_{BH}$.

63: 

64: One of the problems with discovering such relationships is the nature of the sample.  In

65: particular, distance has often introduced a number of biases that need to be dealt with (e.g., the

66: Malmquist effect).  Many biases can be avoided by choosing volume-limited samples, such as

67: clusters of stars or galaxies.  However, AGNs and Galactic black hole binaries are not clustered on small scales,

68: so if a sample is to be constructed, one is forced to combine objects that have very different

69: distances.  When the range of the square in the distance becomes greater than the range in the

70: fluxes, the resulting luminosities are naturally stretched in luminosity-luminosity space, leading

71: to an apparent correlation.  The effect of distances or sample selection can obscure or modify the

72: relationship that one seeks to uncover \citep{chan83}.  The analysis presented here shows that 

73: this distance problem may be partly or entirely responsible for the 

74: claimed fundamental plane of black hole activity and

75: that such a relationship has not yet been uncovered.

76: 

77: \section{The Fundamental Plane}

78: 

79: Obtaining a spurious relationship is an effect that can occur when dealing with a flux-limited sample, provided that the multiplier (i.e., the square of the distance) is large compared to

80: the range of the fluxes.  This spurious relationship will occur even 

81: if the fluxes are initially uncorrelated, so it can

82: obscure a true relationship that may exist.  The existing samples of black

83: holes in papers discussing a fundamental plane have a variety of selection effects

84: that lead them to having a flux-limited nature rather than complete volume-limited samples.  The

85: objects that made it into the samples are largely those that were detected in the radio and X-ray

86: bands, with some estimate for the black hole mass, so most had to be detected and studied in the

87: optical as well.  Most galaxies at

88: 100 Mpc (a typical distance) would not be detected to have nuclear point sources with a

89: typical Chandra observation.  Therefore, a detection will be usually within a couple of orders of

90: magnitude of the detection threshold of the instrument.  A similar situation exists in the radio

91: band, although the detection rate of nuclear radio sources may be a bit higher with the VLA.  The

92: more distant objects must be more luminous if they are to be detected, and although they are

93: rarer, the search volume is substantially larger.  Most of the galactic supermassive black hole

94: sources in the samples are at distances from 1 Mpc to 1 Gpc, so excluding the upper and lower 5\% of objects on

95: the distance scale, the remaining 90\% of the sources have a distance range of 100, and as this

96: enters in the luminosity as distance squared, or a factor of 10$^4$.  The 90\% range in the X-ray and

97: radio fluxes is about a factor of 3000.  The range in d$^2$ is greater, but not vastly greater than the

98: other ranges, yet it can lead to a spurious relationship or distort the nature of the true relationship.

99: 

100: To illustrate this, we take the sample of Merloni, but scramble the radio fluxes so that

101: they are randomly assigned to objects (the X-ray flux, distance, and mass are not scrambled).  We

102: calculate new radio ``luminosities'' and find that plotting these objects fits the fundamental plane

103: about as well as the original data (Figure 1, 2).  Without scrambling fluxes, another test that one

104: can perform is to consider the objects that have upper limits to their radio fluxes

105: rather than detections (this is a minority of objects).  

106: In this case, the fluxes contain almost no information about the objects, merely reflecting the rms of the

107: instrument used.  Yet the resulting correlation has less scatter than the original fundamental

108: plane relationship (Figure 3).

109: 

110: These two illustrations do not prove that a relationship does not exist between L$_X$, L$_R$, and

111: M$_{BH}$, but that one must be concerned as to whether this relationship has yet been found.  The

112: experience in astronomy is that within a class of objects, a source brighter in one waveband tends

113: to be brighter in other wavebands as well.  So the nominal expectation is a near-linear

114: relationship between different luminosities.  The effect of a range of distances stretches a sample

115: along exactly the same relationship, making the two effects degenerate.  Below, we discuss one

116: approach of overcoming this distance problem.

117: 

118: There are distance-independent relationships that can be considered and that may provide

119: important insights.  For the subsample of supermassive black hole systems, we see that there is a

120: relationship between L$_R$/L$_X$ and the black hole mass (Figure 4).  Using the bisector linear

121: fitting method for detected sources only \citep{isobe90,feig92}, we find that L$_R$/L$_X$ $\propto$ M$_{BH}$$^{1.54 \pm 0.13}$, with a

122: Pearson correlation coefficient of 0.44 for 78 degrees of freedom; in fitting y-on-x, L$_R$/L$_X$ $\propto$ M$_{BH}$$^{0.78 \pm 0.19}$. 

123: Although there is significant scatter, reflected in the low correlation coefficient,

124: the relationship is significant at beyond the 99\% confidence level.  

125: The upper and lower limits fall nearly equally on both sides of the regression 

126: line so the relationship still exists when such objects are taken into account.

127: 

128: The implication of this result is that the radio luminosity rises more rapidly than the X-ray

129: luminosity.  However, without a clear sample selection criteria, we cannot be certain that this is

130: not at least partly an artificial result.  For example, some of the sources were originally

131: discovered by radio studies (e.g., 3C390.3 in the 3C survey), which found radio-bright AGNs,

132: often at considerable distances.  The galactic hosts of these AGNs are usually optically luminous

133: early-type galaxies, and as there is a relationship between optical luminosity and black hole mass,

134: these are also systems of preferentially high M$_{BH}$.  Consequently, this type of selection might

135: bias the sample toward the highest L$_R$ at large M$_{BH}$.  Until the sample selection is better

136: understood, we would not place too much significance on this relationship.

137: 

138: \section{Seeking the Relationship Between L$_R$, L$_X$, and M$_{BH}$}

139: 

140: For single sources, it is possible to determine the relationship between L$_R$ and L$_X$, and

141: this has been the basis of several programs.  An impressive case is that of V404 Cyg 

142: \citep{han92}, where all the radio and X-ray luminosities are consistent with L$_R$ $\propto$

143: L$_X$$^{0.7}$ (see also \citealt{gall03}).  Other objects show much broader and more

144: complicated variations between these luminosities, especially at high luminosities

145:  \citep{gall03, corb03}.

146: 

147: In order to study the relationship to black hole mass, along with L$_R$ and L$_X$, one needs

148: a statistically sound sample of objects for which the range in d$^2$ is less than L$_R$ or L$_X$.  

149: For supermassive black holes, we

150: suggest that the safest approach is to use a volume-limited sample of galaxies.  In doing so, one

151: might limit the galaxies by optical luminosity to exclude dwarf galaxies.  The luminosities of

152: interest are those from the black hole, so contaminating sources need to be avoided.  In the X-ray

153: band, X-ray binaries are common in galaxies, but the resolution with the Chandra Observatory

154: may be sufficient to exclude most binaries and obtain L$_X$ from the black hole or an upper limit

155: to L$_X$.  Ideally, the radio observations should be obtained at about the same time as the X-ray

156: data and at sufficient spatial resolution to exclude extended emission.  This would be an

157: ambitious observing program in which many observations would result in upper limits.  However,

158: methods exist to include upper limits to the analysis.  Clearly, investigations should be

159: carried out in the future to provide further insight into this important issue.

160: 

161: \acknowledgments

162: 

163: The author would like to thank Jon Miller, David Spergel, Jimmy Irwin, Renato Dupke, and Eric

164: Feigelson for their comments and advice.  JNB would like to acknowledge support from NASA

165: under LTSA grant NAG5-10765.

166: 

167: \begin{thebibliography}{} 

168: 

169: \bibitem[Chanan(1983)]{chan83} Chanan, G.~A.\ 1983, \apj, 275, 

170: 45 

171: \bibitem[Corbel et al.(2003)]{corb03} Corbel, S., Nowak, 

172: M.~A., Fender, R.~P., Tzioumis, A.~K., \& Markoff, S.\ 2003, \aap, 400, 

173: 1007 

174: \bibitem[Falcke et al.(2004)]{falc04} Falcke, H., K{\"o}rding, 

175: E., \& Markoff, S.\ 2004, \aap, 414, 895 

176: \bibitem[Feigelson \& Babu(1992)]{feig92} Feigelson, E.~D., \& 

177: Babu, G.~J.\ 1992, \apj, 397, 55

178: \bibitem[Gallo et al.(2003)]{gall03} Gallo, E., Fender, R.~P., 

179: \& Pooley, G.~G.\ 2003, \mnras, 344, 60 

180: \bibitem[Han \& Hjellming(1992)]{han92} Han, X., \& 

181: Hjellming, R.~M.\ 1992, \apj, 400, 304 

182: \bibitem[Isobe et al.(1990)]{isobe90} Isobe, T., Feigelson, 

183: E.~D., Akritas, M.~G., \& Babu, G.~J.\ 1990, \apj, 364, 104

184: \bibitem[Merloni et al.(2003)]{merl03} Merloni, A., Heinz, S., 

185: \& di Matteo, T.\ 2003, \mnras, 345, 1057 

186: 

187: \end{thebibliography}

188: 

189: \clearpage

190: 

191: \begin{figure}

192: \plotone{f1.eps}

193: \caption{The objects with detections from \citet{merl03} along with

194: their fit to the data.

195: }

196: \end{figure}

197: 

198: \begin{figure}

199: \plotone{f2.eps}

200: \caption{The same detected objects from \citet{merl03} but with the

201: radio fluxes scrambled.  The result is a very strong correlation that

202: is also fit by the fundamental plane relationship.

203: }

204: \end{figure}

205: 

206: \begin{figure}

207: \plotone{f3.eps}

208: \caption{The objects with radio upper limits from \citet{merl03} along with

209: their fundamental plane fit.

210: }

211: \end{figure}

212: 

213: 

214: \begin{figure}

215: \plotone{f4.eps}

216: \caption{The ratio of the radio to the X-ray luminosity vs the black hole mass

217: for the supermassive black holes objects (AGNs).  The solid points are detections, 

218: the downward pointing triangles have upper limits to their radio luminosities and

219: the upward pointing triangles have upper limits to their X-ray luminosities.

220: The dashed line is the bisector fit to the detected objects, with a slope of

221: 1.54 $\pm$ 0.13 while the dotted line is the y-on-x fit, with a slope of 

222: 0.78 $\pm$ 0.19.

223: }

224: \end{figure}

225: 

226: \end{document}

227: