astro-ph0308359/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: %\documentclass[preprint]{aastex}
3: %\documentclass[]{aastex}
4: \documentclass{emulateapj}
5: \usepackage{apjfonts}
6: 
7: \begin{document}
8: 
9: \shortauthors{Zheng et al.}
10: \shorttitle{$I$-band Luminosity Function}
11: 
12: \title{M Dwarfs from {\it Hubble Space Telescope} Star Counts. V. The $I$-band
13: Luminosity Function\altaffilmark{1}}
14: 
15: \author{Zheng Zheng\altaffilmark{2}, Chris Flynn\altaffilmark{3},
16:         Andrew Gould\altaffilmark{2}, John N. Bahcall\altaffilmark{4},
17:         Samir Salim\altaffilmark{5}} 
18: 
19: \altaffiltext{1}{Based on observations with the NASA/ESA Hubble Space 
20:                  Telescope, obtained at the Space Telescope Science Institute, 
21:                  which is operated by the Association of Universities for 
22:                  Research in Astronomy, Inc. under NASA contract 
23:                  No. NAS5-26555.}
24: \altaffiltext{2}{Department of Astronomy, Ohio State University, Columbus, 
25:                  OH 43210; zhengz@astronomy.ohio-state.edu,  
26:                  gould@astronomy.ohio-state.edu.}  
27: \altaffiltext{3}{Tuorla Observatory, Turku University, 
28:                  V\"{a}is\"{a}l\"{a}ntie 20, FIN-21500, Piikki\"{o},
29:                  Finland; cflynn@astro.utu.fi.}
30: \altaffiltext{4}{Institute for Advanced Study, Princeton, NJ 08540;
31:                  jnb@ias.edu.}
32: \altaffiltext{5}{Department of Physics and Astronomy, University of California
33:                  at Los Angeles, Los Angeles, CA 90095; samir@astro.ucla.edu.} 
34: 
35: \begin{abstract}
36: We derive the disk $I$-band luminosity function from the Zheng et al. sample 
37: of $\sim 1400$ disk M dwarfs observed with the {\it Hubble Space Telescope}. 
38: We adopt a Galactic-height-dependent color-magnitude relation to account
39: for the metallicity gradient above the Galactic plane. The resultant $I$-band 
40: luminosity function peaks at $M_I \sim 9.5$ and drops sharply toward 
41: $M_I \sim 10.5$.
42: \end{abstract}
43: \keywords{stars: late-type -- stars: low-mass -- stars: luminosity function 
44:        -- stars: statistics -- surveys}
45: 
46: 
47: \section{Introduction}
48: 
49: The stellar luminosity function (LF) measures the number density of 
50: stars as a function of luminosity. It is traditionally expressed in the 
51: $V$-band. However, sometimes it is more convenient to use the $I$-band LF for 
52: planning and modeling observations. Here we present the $I$-band LF of disk M 
53: dwarfs from {\it Hubble Space Telescope} ({\it HST}) star counts.
54: 
55: \citet{Zheng01} studied a sample of about 1400 disk M dwarfs found in 
56: 148 fields observed with the Wide Field Camera 2 (WFC2) on the {\it HST} and 
57: 162 fields observed with pre-repair Planetary Camera 1 (PC1). The $V$-band LF
58: and the Galactic disk parameters were derived simultaneously using the method 
59: of maximum likelihood. If the luminosity of M dwarfs depended only on their 
60: color, as is the case for example for the solar neighborhood color-magnitude 
61: relation (CMR) determined by \citet{Reid91}, the conversion from the $V$-band 
62: LF to the $I$-band LF would be straightforward. However, since many stars in 
63: their sample lie far from the Galactic plane, \citet{Zheng01} introduced a 
64: more realistic CMR [CMR (2) in that paper], which makes a first-order 
65: correction for the metallicity effect: the luminosity then depends not only on 
66: the color but also on the height, $z$, above the Galactic plane. With this 
67: more complex CMR, the transformation from the $V$-band LF to the $I$-band LF 
68: is not trivial. 
69: 
70: In this paper, we apply the method of maximum likelihood to the same set of 
71: {\it HST} observations as in \citet{Zheng01} and incorporate the 
72: Galactic-height-dependent CMR to derive the $I$-band LF of M dwarfs. This
73: direct method avoids the uncertainty and the difficulty in the transformation 
74: from the $V$-band LF to the $I$-band LF. For details on the observations, the 
75: CMR, and the maximum likelihood method, we refer readers to \citet{Zheng01} and 
76: references therein. In \S~2, we present our results after a brief description 
77: of models.
78: 
79: \section{Models and Results}
80: 
81: The LF of disk M dwarfs is a function of the Galactic position ($R$, $z$). We 
82: parameterize it as the product of a vector of 12 absolute $I$-magnitude bins 
83: and a Galactic density law,
84: \begin{equation}
85: \Phi(i,z,R)=\Phi_i\nu(z)\exp\biggl(-{R-R_0 \over H}\biggr),
86: \end{equation}
87: where $\Phi_i$ is the solar neighborhood LF for the {\it i}th magnitude bin,
88: $R_0=8$ kpc is the Galactocentric distance of the Sun, and $H$ is the scale
89: length of the disk. The vertical density profile $\nu(z)$ is assumed to have
90: either a ``sech$^2$'' form,
91: \begin{equation}
92: \nu_s(z)=(1-\beta){\rm sech}^2 {z \over h_1} +\beta \exp 
93: \biggl(-{|z| \over h_2}\biggr),
94: \end{equation}
95: or a ``double-exponential'' form,
96: \begin{equation}
97: \nu_e(z)=(1-\beta) \exp\biggl(-{|z| \over h_1}\biggr) +\beta \exp 
98: \biggl(-{|z| \over h_2}\biggr).
99: \end{equation}
100: The magnitude bins are centered at $M_I=$ 6.5, 7.0, 7.5, 8.0, 8.5, 9.0,
101: 9.5, 10.0, 10.5, 11.0, 12.0, and 13.5, respectively. The size of each bin
102: is 0.5 mag except the last two (1.5 mag). Given a choice of disk scale length 
103: $H$, the local LF, $\Phi_i$, and the three disk profile parameters ($h_1$, 
104: $h_2$, $\beta$) are found by maximizing the likelihood. The scale length
105: is determined by finding the maximum in the maximized likelihoods of
106: an ensemble of solutions using different values of $H$. 
107: 
108: As in \citet{Zheng01}, we select stars below $z=$ 2400 pc. The absolute 
109: $I$-magnitude of stars should satisfy $6.25<M_I<14.25$. The faint boundary 
110: is set to prevent contamination by spheroid giants. When translated into
111: color, the boundary we use here is slightly bluer than that in 
112: \citet{Zheng01}, but it is still acceptable since our results show virtually 
113: no change if we move the boundary to a higher absolute $I$-magnitude. 
114: Altogether, 1403 stars satisfy our selection criteria and they constitute 
115: our sample. This sample has 30 more stars than the one used by \citet{Zheng01}, 
116: which results primarily from the small difference in the blue boundary. 
117: 
118: We solve for the LF and disk profile parameters by maximizing the likelihood
119: and find that the resultant disk profile parameters, for either the sech$^2$
120: model or the double-exponential model, are almost identical to those derived 
121: in \citet{Zheng01}. This consistency of the results implies that the disk 
122: profile derived from the maximum likelihood method is robust. 
123: We therefore decide to fix the disk parameters at the values in 
124: \citet{Zheng01}, namely $h_1=$ 270 (156) pc, $h_2=$ 440 (439) pc, 
125: $\beta=$ 56.5\% (38.1\%), and $H=$ 2.75 kpc for the sech$^2$ 
126: (double-exponential) model. 
127: 
128: Similar to the the case in $V$-band, due to the lack of local stars in the 
129: sample, the best-fit LF for the sech$^2$ model and that for the 
130: double-exponential model have nearly the same shape but differ from each 
131: other in normalization, and the normalization difference is compensated by 
132: the vertical density profile. We adopt a linear combination of the LFs of
133: these two models as our final result, with the same combination coefficient 
134: derived by \citet{Zheng01} by matching the local normalization to the 
135: well-established luminous end of the solar neighborhood M-dwarf $V$-band LF.
136:  
137: The $I$-band LF from our analysis is listed in Table 1 and shown in Figure 1.
138: As expected, the overall shape of the $I$-band LF, with a peak at $M_I\sim 9.5$
139: and a sharp drop toward $M_I\sim 10.5$, mimics the $V$-band LF in 
140: \citet{Zheng01}.
141: 
142: \acknowledgments
143: 
144: We thank B. Paczy\'nski, whose request for an $I$-band LF was the genesis
145: of this paper. Work by Z.Z. and A.G. was supported in part by grant AST
146: 02-01266 from the NSF. Z.Z. received additional support from a Presidential
147: Fellowship from the Graduate School of The Ohio State University. C. F. has 
148: been supported by the Academy of Finland through its support of the ANTARES 
149: program for space research.
150: 
151: \begin{deluxetable}{rr}
152: \tablecaption{The $I$-band Luminosity Function}
153: \tablehead{M$_I$ & $\phi$ \\
154: (mag) & (10$^{-3}$pc$^{-3}$$I$-mag$^{-1}$)}
155: \startdata
156:  6.5 &  4.5$\pm$1.1\\
157:  7.0 &  5.3$\pm$1.2\\
158:  7.5 &  6.0$\pm$1.2\\
159:  8.0 & 10.1$\pm$1.5\\
160:  8.5 & 13.0$\pm$1.8\\
161:  9.0 & 19.5$\pm$2.1\\
162:  9.5 & 23.1$\pm$2.2\\
163: 10.0 & 14.8$\pm$1.8\\
164: 10.5 &  7.6$\pm$1.5\\
165: 11.0 &  8.4$\pm$1.6\\
166: 12.0 &  3.4$\pm$0.6\\
167: 13.5 &  3.3$\pm$0.9\\
168: \enddata
169: \end{deluxetable}
170: 
171: \begin{figure}[h]
172: \plotone{f1.eps}
173: \caption[]{The $I$-band luminosity function based on 1403 M dwarfs 
174: observed with {\it HST}.}
175: \end{figure}
176: 
177: \begin{thebibliography}{}
178: 
179: \bibitem[Reid(1991)]{Reid91}
180: Reid, I.\ N.\ 1991, \aj, 102, 1428
181: 
182: \bibitem[Zheng et al.(2001)]{Zheng01} 
183: Zheng, Z., Flynn, C., Gould, A., Bahcall, J.\ N., \& Salim, S.\ 2001, \apj,
184: 555, 393
185: 
186: \end{thebibliography}
187: 
188: \end{document}
189: 
190: