0707.0464/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: \newcommand{\tnm}{\tablenotemark}
3: \newcommand{\tnt}{\tablenotetext}
4: 
5: \shorttitle{Properties of M Dwarfs}
6: \shortauthors{T. N. Gautier et al.}
7: 
8: \begin{document}
9: 
10: \title{Far Infrared Properties of M Dwarfs}
11: 
12: 
13: \author{Thomas. N. Gautier III \altaffilmark{1},
14: G. H. Rieke \altaffilmark{2},
15: John Stansberry \altaffilmark{2},
16: Geoffrey C. Bryden \altaffilmark{1},
17: Karl R. Stapelfeldt \altaffilmark{1},
18: Michael W. Werner \altaffilmark{1},
19: Charles A. Beichman \altaffilmark{1},
20: Christine Chen \altaffilmark{3},
21: Kate Su \altaffilmark{2},
22: David Trilling \altaffilmark{2},
23: Brian M. Patten \altaffilmark{4} and 
24: Thomas L. Roellig \altaffilmark{5}}
25: 
26: 
27: \altaffiltext{1}{Jet Propulsion Laboratory, California 
28: Institute of Technology, Pasadena, California 91109; Thomas.N.Gautier@jpl.nasa.gov}
29: \altaffiltext{2}{Steward Observatory, University of Arizona, 933 North Cherry Avenue, 
30: Tucson, Arizona 85721}
31: \altaffiltext{3}{National Optical Astronomy Observatories, Tucson, Arizona 85719}
32: \altaffiltext{4}{Harvard-Smithsonian Center for Astrophysics, 60 Garden St., 
33:                  Cambridge, Massachusetts 02138}
34: \altaffiltext{5}{NASA Ames Research Center, MS 245-6, Moffett Field, California 94035}
35: 
36: 
37: \begin{abstract}
38: 
39:      We report the mid- and far-infrared properties of nearby M dwarfs. 
40: {\it Spitzer}/MIPS measurements were obtained for a sample of 62 stars 
41: at 24 \micron, with subsamples of 41 and 20 stars observed at 70 \micron\ 
42: and 160 \micron\ respectively. We compare the results with current models 
43: of M star photospheres and look for indications of circumstellar dust in 
44: the form of significant deviations of \hbox{K-[24 \micron]} colors and 
45: 70 \micron\ / 24 \micron\ flux ratios from the average M star values. At 
46: 24 \micron, all 62 of the targets were detected; 70 \micron\ detections were 
47: achieved for 20 targets in the subsample observed; and no detections were 
48: seen in the 160 \micron\ subsample. No clear far-infrared excesses were detected 
49: in our sample. The average far infrared excess relative to the photospheric 
50: emission of the M stars is at least four times smaller than the similar average 
51: for a sample of solar-type stars. However, this limit allows the average 
52: fractional infrared luminosity in the M-star sample to be similar to that for 
53: more massive stars. We have also set low limits ($10^{-4}$ to $10^{-9}$ 
54: M$_{Earth}$ depending on location) for the maximum mass of dust possible 
55: around our stars.
56: 
57: 
58: \end{abstract}
59: 
60: \keywords{Infrared: stars - stars: late-type - stars: 
61: fundamental parameters (colors) - planetary systems}
62: 
63: \section{Introduction}\label{intro}
64: 
65:      The study of dust around main sequence stars addresses intriguing questions 
66: in basic astrophysics while providing insights into the possible existence 
67: of planetary systems around these stars. The pioneering results from IRAS on 
68: infrared excess around stars like Vega \citep{au84} rapidly led to 
69: the realization that the circumstellar dust responsible for this emission 
70: could not survive for the lifetime of the star and that it must be 
71: replenished - from cometary or asteroid-type objects - just as the zodiacal 
72: cloud in our Solar System is replenished. The maintenance of this dust 
73: ultimately depends on collisions between asteroid-size objects, yielding 
74: fragments that collide with each other and eventually grind down to fine 
75: dust that is heated by the star and appears as an infrared excess. Thus the 
76: characteristics of circumstellar dust around a main sequence star (referred 
77: to hereafter as a ``debris disk'') reflect the planet-building and 
78: planet-destroying processes occurring around that star.
79: 
80:      The {\it Spitzer} Space Telescope \citep{wer04}, building on the 
81: foundation established by IRAS and ISO, has already produced major advances 
82: in our understanding of the debris disk phenomenon. These include 
83: demonstration of the stochastic nature of debris disk evolution 
84: \citep{gr05,su06,bei05a}, the first detection of debris disks around F, G, and 
85: K stars with known radial velocity companions \citep{bei05b} and investigations 
86: of the frequency of occurence of debris disks around solar type stars \citep{kim05,bry05}. Here 
87: we present the results from a {\it Spitzer} search for dust associated 
88: with M dwarfs.
89: 
90:      M dwarfs, with masses between 0.08 and 0.5 M$_\odot$, are the most 
91: common stars in the galactic plane. Thus the study of debris disks around them 
92: is needed to complete our understanding of the nature and incidence of debris 
93: disks in the solar neighborhood and the dependence of planet-forming processes 
94: on the mass of the central star. Since M stars are the most numerous 
95: stellar type, measurement of their debris disk frequency has large 
96: implications for the overall population of planetary systems in the galaxy. 
97: It may be that the nearest extrasolar planets to the Sun are orbiting M 
98: stars, a possibility that would be supported if M dwarfs frequently host 
99: debris systems.
100: 
101:      However, the timescales and processes for dust expulsion may be considerably 
102: different for M stars than for their more luminous cousins \citep{plav05}. 
103: Also, early results from precision radial velocity surveys have suggested that M stars 
104: have a lower frequency of giant planets than F and G stars \citep{but04}. It is 
105: important to determine if such a marked difference is also present in the relative 
106: frequency of debris disks between the M dwarfs and earlier types.
107: 
108:      Finally, besides extending our understanding of the evolution of planetary 
109: systems by investigating the frequency of debris disks around M dwarfs, 
110: these far-infrared measurements can determine basic parameters of the M 
111: stars themselves. We compare our data with current models of M dwarf 
112: photospheric emission and demonstrate how to refine determinations of M 
113: dwarf effective temperatures with the accurate 24 \micron\ data available 
114: with {\it Spitzer}.
115: 
116: \section{New Observations}\label{obs}
117: 
118:      We observed 62, 41, and 20 M dwarfs, respectively, at 24, 70, and 160$\mu$m, 
119: using the Multiband Imaging Photometer for {\it Spitzer} \citep[MIPS;][]{gr04}. 
120: Our target stars are listed in Table \ref{data_table}. Stars of spectral type M7 and 
121: earlier (a total of 40), plus one M9 star, are part of a MIPS program to 
122: obtain photometry of all stars lying within 5 pc of the sun ({\it Spitzer} program 
123: ID 52). Twenty one later-type stars (M6 - M9) lying at distances of 6 to 18 pc 
124: were generally selected to be the brightest members of their spectral classes 
125: ({\it Spitzer} program ID 56). These later-type stars were observed only at 24 \micron. 
126: While age determination of field M stars is notoriously difficult it is likely 
127: that the stars in our sample are nearly all older than 1 Gy.
128: 
129:      The 24 $\mu $m observations were made using the MIPS photometry observation 
130: template with the small dither size. For the nearby stars we generally used 
131: 1 cycle of the dither pattern (48s integration time). IL Aqr, LHS 3685 and LP 944-20 
132: were exceptions receiving 180s, 136s and 1500s integration time respectively. This strategy 
133: typically resulted in very high signal-to-noise ratios (SNRs) at 24 \micron. 
134: The more distant, later type, stars received from 1 to 9 dither 
135: cycles (48s to 1380s) at 24 \micron; many of them were measured only to SNR $\sim\! 10$.
136: 
137:      The 70 $\mu $m observations were also made using the MIPS photometry 
138: observation template with the small dither size and the default image scale 
139: \citep{gr04}. In this band, the required integration times were much longer 
140: than at 24 \micron\ and therefore automatically resulted in several cycles of 
141: the dither pattern. We limited the maximum integration time at 70 \micron\ to 
142: $\sim\! 1000$ seconds, both to fit the program within its time allocation and 
143: because the noise behavior of MIPS at longer integration times was unknown at 
144: the time these observations were scheduled (before the launch of {\it Spitzer}). 
145: Our goal of SNR $\sim\! 6$ on the photosphere was reached for many of the 
146: 70 \micron\ observations but the limitation on integration time resulted in low 
147: SNR detections or upper limits at 70 \micron\ for the fainter members of the sample. 
148: 
149:      To explore the possibility that these low-luminosity stars might possess 
150: very cold debris disks, 160 $\mu $m photometry was carried out for 20 stars 
151: in the sample. Detection of the stellar photospheres, whose median flux 
152: densities are expected to be a few mJy, was not achievable given {\it Spitzer's} 
153: measured extragalactic confusion limit of 8 mJy, 1$\sigma$ \citep{dol04}. 
154: The observations were therefore designed only to detect a source 3-5 
155: times the confusion noise level. Four cycles of 160 \micron\ photometry, 
156: totaling 84 sec of exposure time for each source, were used.
157: 
158: 
159: %\subsection{Data Reduction}\label{reduction}
160: 
161:      The data were processed using the MIPS instrument team Data Analysis Tool 
162: (DAT, version 2.73) described by \citet{gord05}. For the 24 \micron\ data, 
163: basic processing included slope fitting, flat-fielding, and corrections for 
164: droop and ``jailbar'' artifacts (DC offsets between adjacent detector columns 
165: that appear in measurements of bright sources). Further details are given by
166: \citet{eng06}. The 70 \micron\ data processing also used the DAT, but was similar to 
167: that of the {\it Spitzer} pipeline version S12. Details are given in 
168: \citet{gord04}.  Mosaics were constructed using pixels interpolated to about 
169: 1/2 the native pixel scale of the arrays. 
170: 
171:      Fluxes were measured using fairly small apertures to improve the 
172: signal-to-noise ratios of the measurements, particularly given 
173: the relatively high spatial variability of the sky background at 70 \micron. 
174: Most targets were measured using apertures 9.96\arcsec\ and 39.4\arcsec\ 
175: in diameter at 24 and 70 \micron. The observed PSF full width at half maximum 
176: is 6.4\arcsec\ and 19.3\arcsec\ , at 24 and 70 \micron\ respectively. In a few 
177: cases nearby sources contributed significant flux at the target location, so we 
178: used apertures 25\% smaller than those just described to reduce contamination. 
179: In one case (LHS 58 \& 59, identified in the notes to Table \ref{data_table}), 
180: we deblended two sources at 70 \micron\ using the model PSFs. 
181: The $\sim10$\arcsec\ aperture at 24 \micron\  corresponds to disk radii of  
182: 7-90 AU at the various distances of our stars. At 70 \micron, where only stars 
183: within 5 pc were observed, the $\sim40$\arcsec\ aperture corresponds to disk radii 
184: of 36-200 AU.
185: 
186: After extraction, the measured fluxes were corrected to infinite aperture. 
187: Aperture corrections were computed using STinyTim model PSFs \citep{krist05}
188: for a 3000 K blackbody source. The model PSFs were smoothed until they provided 
189: good agreement with observed stellar PSFs out through the first Airy maximum.
190: 
191:      Targets were usually identified in the 24 \micron\ images using their J2000 
192: coordinates and proper motions from SIMBAD. In most instances the agreement 
193: between the predicted and observed position was excellent (see Notes to 
194: Table \ref{data_table} for cases where there was a significant disagreement). 
195: However, many of our later M type stars are rather dim, even at 24 \micron, and 
196: special care was needed to ensure that the correct object was measured among 
197: many others in these fields. We confirmed a number of identifications with 
198: IRAC 3.6 \micron\ images from program 35 (Multiplicity and Infrared Colors of Nearby MLT 
199: Dwarfs, \citet{pat06}). The 24 \micron\ photometric aperture was centered on the photocenter 
200: of the target. We used the same procedure at 70 \micron\ when the source was 
201: clearly detected and there were no confusing sources nearby. In cases where 
202: the 70 \micron\ centroid position disagreed by more than 5\arcsec\  with the 
203: 24 \micron\ position, we forced the 70 \micron\ aperture to be centered at the 
204: 24 \micron\ position.
205: 
206:      The fluxes we report are based on conversion factors of 1.048 $\mu$Jy
207: /arcsec$^{2}$/(DN/s) and 16.5 mJy/arcsec$^{2}$/U70 at 24 and 70 \micron, 
208: respectively (U70 = MIPS 70 \micron\ unit). Table \ref{data_table} lists the 
209: observed fluxes, their formal 1$\sigma$ uncertainties (not including systematic 
210: calibration uncertainties) and 3$\sigma$ upper limits where no flux could 
211: be determined.
212: 
213:      For most of the observed stars, the error in the 24 \micron\ 
214: photometry is dominated by calibration uncertainty, which contributes 
215: systematic 24 \micron\ errors at the level of $\sim$5\% \citep[see e.g.][]{bei05a}. 
216: However, we do not depend on the absolute calibration of the measurements 
217: to identify debris disks. We are instead limited by the scatter in the 
218: predictions of photospheric flux density from shorter wavelengths \citep[see][]{su06}. 
219: Background contamination by galactic cirrus or by extragalactic confusion is 
220: of relatively little importance. Dim sources, however, are limited by photon 
221: noise and by flat fielding uncertainties created by latent dark images. The 
222: effective noise level for the dimmest sources is $\sim\!\! 0.05$ mJy. Except
223: for the faintest sources, we can detect excesses down to $\sim$15\%
224: above the photosphere in this band.
225: 
226:      At 70 \micron\ the flux error is a combination of calibration uncertainty and 
227: background noise. The variable sky background, a combination of galactic 
228: cirrus and extragalactic confusion, presents a noise floor that cannot be 
229: improved with increases in integration time. This confusion noise is reached 
230: with integrations of 100's to 1000's of seconds depending on the local level 
231: of galactic cirrus brightness \citep{bry05}. This variation of the sky background 
232: within each frame, when convolved with our chosen beam size, is listed as the 
233: 70\micron\ flux error in Table 1. On top of this background noise, we have included 
234: a 10\% systematic error at 70 \micron, consistent with the uncertainty found within 
235: photospheric measurements of FGK stars \citep{bei05a,bry05}. A thorough discussion 
236: of the uncertainties in the 70 \micron\ data can be found in these papers and in 
237: \citet{gord04}. Excesses of 30\% above the predicted photosphere for bright stars 
238: at 70 $\mu $m should be readily detectable. 
239: 
240: At 160 \micron, the DAT was used to calibrate the data, and mosaic the images at a 
241: pixel scale of 8\arcsec/pixel (half the native  scale). The data were flux-calibrated 
242: using the standard conversion factor of 42.6 MJy/sr/MIPS160, where MIPS160 are the 
243: instrumental units that result after the data are normalized by the stim flashes. 
244: The uncertainty on the absolute calibration is 12\% \citep{stan07}. The MIPS 
245: 160 \micron\ array suffers from a near-IR spectral leak that produces a strong ghost 
246: image overlapping the true 160 \micron\ image for sources with stellar temperatures 
247: and J magnitudes brighter than 5.5 (MIPS data handbook, http://ssc.spitzer.caltech.edu/mips/dh/).  
248: Only four of our brightest stars have a possibility of being effected by this leak. 
249: One of these stars (LHS 70) is the single case where one of our targets was detected 
250: at 160 \micron, the spectral leak was detected and not the true 160 \micron\ emission. 
251: This detection is ignored and we give $3\sigma$ upper limits on the 160 \micron\ fluxes 
252: for all observed targets.
253: 
254: In most cases these limits are estimated from the scatter of pixels falling within a 
255: sky annulus with radii of 64\arcsec\ -- 128\arcsec, and normalized  to an aperture 
256: radius of 24\arcsec. In some cases we also estimated the sky noise by dropping multiple 
257: 24\arcsec\ apertures on the images (away from the source position), and using the scatter 
258: of the fluxes in those apertures instead. The limits are corrected for the finite aperture 
259: size using a correction factor of 2.37, derived from STiny Tim model PSFs \citep{krist05} 
260: fitted to observed PSFs \citep{stan07}. The resulting 3$\sigma$ upper limits 
261: are presented in Table \ref{data_table}. There is considerable variation in the sensitivity 
262: achieved, due to the effects of galactic cirrus emission. The median upper limit is 74 mJy, 
263: but the limits for individual stars range from 53 to 963 mJy.
264: 
265: \section{Interpretation}\label{interp}
266: 
267:      Because the intrinsic mid- and far-infrared properties of M stars are not well known, 
268: we take the approach of using our photometry and existing K band photometry to compare the 
269: M stars among themselves to look for evidence of excess emission. Extrapolations from K$_S$ 
270: to [24] require an accurate understanding of the photospheric colors that we determine in 
271: Section \ref{colorsnxs}. We find no convincing evidence for any 24 \micron\ excesses above 
272: the photospheric outputs. In section \ref{70xs} we use the ratio of our measured 70 \micron\ 
273: flux to a 70 \micron\ flux predicted from our 24 \micron\ measurement as our diagnostic tool. 
274: Again, we find no convincing evidence for a 70 \micron\ excess 
275: at greater than the 3$\sigma$ level. The 160 \micron\ upper limits are discussed in 
276: section \ref{160ul}. Finally, in sections \ref{dust} and \ref{comp}, we compare 
277: the results on dust excesses around M stars with those obtained by previous {\it Spitzer} 
278: studies of main sequence stars in the solar neighborhood.
279: 
280: \subsection{Photospheric colors and 24 \micron\ excess}\label{colorsnxs}
281: \subsubsection{Photospheric colors}\label{colors}
282: 
283:      Table \ref{color_table} lists the K$_S$ band magnitudes and the 
284: colors derived from the infrared photometry. The K$_S$ band data were 
285: obtained mostly from the 2MASS catalog \citep{cut03a} but in some cases 
286: K measurements were obtained from other sources and transformed to 2MASS 
287: K$_S$ (see notes to Table \ref{color_table}). 
288: 
289:      We show a color-color plot of K$_S$ - [24] vs. V - K$_S$ in Figure \ref{V-K_fig}. 
290: The V magnitudes were obtained from NStars (http://nstars.nau.edu/), where possible, 
291: with SIMBAD as a secondary 
292: source. Eleven of our 62 objects have no accurate V photometry available. 
293: The K$_S$ - [24] colors are greater than zero, reflecting the fact 
294: that the spectral energy distributions of M stars do not transition to the 
295: Rayleigh-Jeans regime until well beyond 2 \micron. They also become increasingly 
296: red with later spectral type, reflected by larger V - K$_S$. With its long
297: wavelength baseline, the V - K$_S$ color should be monotonic in spectral type/effective
298: temperature. This plot should therefore allow us to identify excess candidate
299: stars as outliers toward red K$_S$ - [24] from the overall trend. In fact, there is
300: one such object, V1581 Cyg B. However, the figure also shows significant
301: scatter, substantially more than expected purely from the errors in V, 
302: K$_S$ or [24], suggesting that there are significant variations in V - K$_S$
303: outside a purely monotonic relationship. 
304: 
305: \subsubsection{Identification of Excesses Using IRFM for Stellar Temperatures}
306: 
307:      Although accurate photometry for the entire sample is available in the near
308: infrared, the issues shown in Figure 1 cannot be addressed with these measurements
309: because M dwarf spectral type is not a sufficiently strong monotonic function of infrared colors. 
310: To improve over color-color plots, we have used a modified form of the infrared flux method \citep[and references 
311: therein]{blac86} to determine effective temperatures for the stars in the sample, 
312: designated T$_{ir\kern-0.1em fm}$. We can then plot K$_S$ - [24] as a function of T$_{ir\kern-0.1em fm}$
313: rather than V - K$_S$. 
314: 
315: The effective temperature can be expressed as:
316: 
317: \[
318: 10log\left( {T_{ir\kern-0.1em fm}}  \right) = 42.32 - 5{\kern 1pt} log\left( {9.43 \times 
319: 10^{15}\left( {\frac{{S_{\nu}  \left( {24} \right)}}{{\left( {\int {S_{\nu}  
320: \,d\nu} }  \right)^{1/4}\alpha \left( {T} \right)\,\beta \left( {T} 
321: \right)}}} \right)^{2/3}} \right) - \left( {m_{K} + BC_{K}}  \right),
322: \]
323: 
324: \noindent
325: where $\alpha$(T) is a term of order unity that is the ratio of the Planck 
326: function to the Rayleigh Jeans approximation at 24 \micron, and $\beta$(T) 
327: is another term of order unity ($\beta$(T) = 0.76 independent of 
328: temperature for these stars) that is the ratio of the flux received through 
329: the 24 \micron\ filter to the bolometric flux from the star, divided by the 
330: similar ratio for a blackbody at the effective temperature of the star. For 
331: stars of type M6 and later, we take bolometric corrections from \citet{golim04}. 
332: For stars earlier than M6, we used 
333: 
334: \[
335: BC_{K} = 0.42 + 1.45\left( {I - K} \right) - 0.20\left( {I - K} 
336: \right)^{2}.
337: \]
338: 
339: \noindent
340: This expression is based on that given by \citet{rh00}, but has been modified 
341: to join better to the expression of \citet{golim04} at the M5.5 to M6 transition. 
342: Our derived values of T$_{ir\kern-0.1em fm}$ are listed in Table \ref{color_table}.
343: 
344:      Figure 2 shows the trend of K$_S$ -[24] color vs. 
345: T$_{ir\kern-0.1em fm}$. The scatter in the observed colors at a fixed temperature 
346: is small, consistent with it arising purely from the measurement errors. 
347: 
348:      We have compared the colors of the low mass stars with synthetic colors 
349: generated from a grid of models to be published by Brott et al. \citep[see][]
350: {brot05}. These models represent dusty atmospheres and are termed the ``GAIA grid" 
351: because they were calculated in support of planning for the GAIA mission. 
352: They incorporate advances in the models published by \citet{all01}, through 
353: improvements in parameters such as opacities and mixing lengths. The theoretical 
354: spectra were convolved with the wavelength-dependent instrumental response for the 
355: 2MASS K$_S$ filter and the MIPS 24 \micron\ filter. Similar convolutions were 
356: performed for a Kurucz model A-star spectrum, and the two were ratioed to obtain 
357: colors relative to A-star colors for the low mass stars. 
358: 
359:      The synthetic colors are also compared with the measured ones in
360: Figure \ref{K-24_fig}. The observations follow roughly the theoretical curve, 
361: although the observed colors are systematically up to 20\% less red in the 2500 - 3500 K 
362: range.  Among the contributors to this discrepancy could be the effects 
363: of dust in the atmospheres, which can modify the absorption feature 
364: strengths in the K-band as well as affect the radiation transfer and hence the 
365: emitted flux in the 24 \micron\  band. These new observations should help provide a basis 
366: for improvement of models.
367: 
368:      Figure \ref{K-24_fig} also plots the positions of the stars using a conventional 
369: spectroscopic temperature calibration. We took spectral types from the NStars tabulation as of 
370: December 31, 2006 wherever possible. In the remaining cases, the types are from 
371: SIMBAD as of the same date. In the latter instances, we consulted the primary references 
372: to verify the tabulated values. We used a type of M7.5 for LP 044-162, where the SIMBAD 
373: classification of M6.5 does not agree with the M7.5 found in the literature 
374: \citep{giz00,cruz03}. From M0 through M6, the temperature calibration is from 
375: \citet{tok00}, while from M6.5 through M9.5, we have assumed a linear relation 
376: with a slope of 200K per stellar sub-type. This relation joins the stellar and brown dwarf 
377: temperatures smoothly and is consistent in approach with the temperature scales for L 
378: dwarfs \citep{bur02}. Arrows indicate the shifts in effective temperature from 
379: these values to those determined through the IRFM in this work. If an object has a 
380: deficiency in flux at 24 \micron\ relative to the conventionally determined temperature, 
381: then the IRFM calculation will indicate a higher temperature; if the object has a higher 
382: flux, then the IRFM will return a lower one. It can be seen that most of the shifts are 
383: toward {\it higher} temperatures. 
384: 
385: If these stars had a combination of underestimated temperature plus an excess at 24 
386: \micron, then the corrected temperatures should show considerable scatter in the plot 
387: (because the bolometric corrections on which the plot depends are based on a broad range 
388: of wavelengths). Instead, these points come to a well determined relationship with only 
389: small scatter. Therefore, we look for excesses among the stars where the IRFM shifts the 
390: temperature {\it lower} than the conventional calibration. An example is V1581 Cyg A , 
391: the star with a spectroscopic temperature of 2940K, adjusted to 2756K by the IRFM. 
392: However, this star has a low mass companion V 1581 Cyg C that is not resolved in our 
393: beam. This companion is faint enough in the visible not to bias the spectrum 
394: significantly, but it can contribute significantly at 24 \micron\ \citep[e.g.,][]{dr00}. 
395: A satisfactory fit to the properties of V 1581 Cyg A/C is a M6 primary 
396: and L1 secondary. The predicted brightness difference between A and C is 3 magnitudes at 
397: R, while the predicted difference at [24] is only 0.8 magnitudes. The observed value is 
398: 3.3 magnitudes at V (red version of V with an effective wavelength of 0.583 \micron; 
399: \citet{hen99}); a difference of 0.8 magnitudes at [24] explains 
400: the behavior in the figure. Thus, the behavior in both bands is explained
401: by the M6/L1 binary hypothesis. Although the excess in this case is not related to a 
402: debris disk, the temperature shift illustrates how a debris disk could be identified. 
403: 
404: The second object where we derive a lower temperature is V 1581 Cyg B, the single
405: star suggested to have an excess in the color-color plot in Figure 1. It is not known 
406: to have a companion, but our results suggest it would be worthwhile to look for one. 
407: The third strongest example, LHS 3814, is also a double that is unresolved in our beam 
408: at 24 \micron\ (we have plotted it at the 
409: temperature of the hotter member). The fourth strongest example, LP 944-20, is thought 
410: to be only $\sim$300 Myr old, much younger than a typical object in the sample 
411: \citep{rib03}. Assuming it has lower surface gravity than the other members of the sample, the 
412: molecular bands in the 2 \micron\ window will be strengthened consistent with its color 
413: deviation. All of the remaining objects fall in the region where no infrared excess is 
414: indicated.
415: 
416: \subsubsection{24 \micron\ excesses}\label{24xs}
417: 
418:      Any star with a 24 \micron\ excess should lie above the photospheric color 
419: locus in Figure 2 or be shifted downward in temperature with our IRFM
420: estimates.  Although one could hypothesize that V1581 Cyg B has a debris-disk 
421: excess, since 24 \micron\ excesses are very uncommon around FGK stars \citep{bry05} 
422: and since we find no large excess at 70 \micron, this possibility is unlikely, and we 
423: will not consider it further. We therefore interpret Figure \ref{K-24_fig} to show 
424: that there are no debris-disk-associated excesses at 24 \micron.
425: 
426:      We have repeated our analysis with artificial excesses at 24 \micron. An excess 
427: of 20\% above the stellar photosphere results in a downward adjustment of the indicated 
428: stellar temperature by about 200 K. We can compare this value with the scatter in stellar 
429: colors. The thick solid line in Figure \ref{K-24_fig} can be fitted by
430: 
431: $$ {\rm K}_S-[24]= -0.2915 + {1 \over{1.267\times 10^{-4}\;{\rm T} + 
432:                    (2.389\times 10^{-4}\;{\rm T})^2 + (2.2\times 10^{-4}\;{\rm T})^3}}$$
433: 
434: \noindent A predicted 24 \micron\ flux density (P$_{24}$ in Table \ref{color_table}) was 
435: produced from this K$_S$ - [24] and the measured K$_S$. Then the ratio of the measured 
436: 24 \micron\ flux density to P$_{24}$ was formed for each star (F$_{24}$/P$_{24}$ in 
437: Table \ref{color_table}). This ratio has a mean of 1.04 and a standard deviation of 
438: 0.04 over our sample of 55 stars with derived T$_{ir\kern-0.1em fm}$.  This standard 
439: deviation corresponds to about 100 K in temperature. Excepting the two cases discussed 
440: in the preceding section, all the stars in this sample either have spectroscopically 
441: determined temperatures lower than T$_{ir\kern-0.1em fm}$, or in agreement with this 
442: temperature. Therefore, the limits on excesses are of the order of 20\%, 2$\sigma$.
443: 
444: \subsection{70 $\mu $m Excesses}\label{70xs}
445: 
446:      Thirteen of the 20 stars with detected 70 \micron\ fluxes have high 
447: quality measurements (SNR $>4$ and no interference from nearby objects or cirrus) 
448: that allow a good determination of 70 \micron\ excess. 
449: We formed the ratio, given in Table \ref{color_table} as F$_{70}$/P$_{70}$, 
450: of the measured 70 \micron\ flux to a predicted 70 \micron\ flux derived from 
451: a black body extrapolation of our measured 24 \micron\ flux using T$_{ir\kern-0.1em fm}$ as 
452: the black body temperature. In Figure 3 we have plotted this 
453: ratio  vs.~T$_{ir\kern-0.1em fm}$ for these 13 stars.  Objects with only photospheric 
454: emission appear at a value of one on this plot, while those with excess emission 
455: at 70 \micron\ would appear at values above unity. None of the 13 stars shows a 
456: 70 \micron\ excess larger than the 3$\sigma$ uncertainty in its (F$_{70}$/P$_{70}$) 
457: ratio. A 70 \micron\ excess of 100\% of the photospheric emission should be apparent 
458: in any of these stars. This null result is just at an interesting level of significance 
459: (see section 3.5.1 for discussion).
460: 
461:      We also looked for 70 \micron\ excesses by comparing our 70 \micron\ detections 
462: among themselves. Any star with an excess should appear out of family with the, 
463: presumed, majority of non-excess stars. A $\pm 3\sigma$ range for F$_{70}$/P$_{70}$ 
464: of (0.59, 2.18) was obtained for the stars in Figure \ref{70_ratio_fig} by finding 
465: the mean and standard deviation of $log({\rm F}_{70}/{\rm P}_{70})$. None of the 13 
466: stars in Figure \ref{70_ratio_fig} exceed this range, confirming the conclusion of the previous 
467: paragraph. Among all twenty 70 \micron\ detections (SNR $\ge 3.0$) only AD Leo, YZ Cet and LHS 39 
468: fall outside of $0.59<{{\rm F}_{70}/{\rm P}_{70}}<2.18$, the range indicating no significant excess. 
469: AD Leo is anomalously low and may indicate a problem with the measurement of this star. 
470: YZ Cet may be contaminated by a background galaxy and should not be considered as having an excess 
471: without confirmation. LHS 39 is tantalizingly high, indicating the possibility of a 70 \micron\ excess, 
472: but its 70 \micron\ detection is only barely at 3.0 $\sigma$ significance and additional 
473: uncertainties entering the F$_{70}$/P$_{70}$ ratio result in too low a significance to 
474: claim an excess.
475: 
476: 
477: \subsection{160 $\mu $m Upper Limits}\label{160ul}
478: 
479:      The 160 \micron\ band is more sensitive to the emission from colder dust, 
480: though the constraints vary greatly from star to star depending on the quality 
481: of the 160 \micron\ data. Lalande 21185 and V1216 Sgr represent the range of 160 \micron\ 
482: limits within our data. We discuss the 160 \micron\ upper limits along with
483: the shorter wavelength measurements in the following section.
484: 
485: \subsection{Fractional dust luminosities}\label{dust}
486: 
487:      Figure \ref{dust_fig} illustrates the limits on dust luminosity derived from our flux 
488: measurements. The upper limits on the ratio of dust luminosity to star luminosity 
489: (L$_{d}$/L$_{*}$) are shown as a function of dust temperature for two stars, 
490: Lalande 21185 and V1216 Sgr, representing the range of limits obtained for stars 
491: detected at 70 \micron. For each dust temperature, we calculate the maximum 
492: blackbody emission that is consistent with the observed 3$\sigma$ limits. 
493: Each of the curved segments in Figure \ref{dust_fig} corresponds to an individual observation 
494: at a specific wavelength, with the temperatures that an observation is most sensitive 
495: to depending on that wavelength. The 70 \micron\ observations are particularly 
496: useful for constraining the emission from dust with temperatures $\sim$50 - 100 K, 
497: for which fractional luminosities greater than $\sim\!\! 10^{-4}$ are generally excluded.
498: 
499:      The constraints on L$_{d}$/L$_{*}$ in Figure \ref{dust_fig} can be translated into upper 
500: limits on dust mass, for an assumed dust size and orbital location. Assuming 
501: micron-size dust, the 160 \micron\ upper limits correspond to dust mass 
502: limits of $\sim\!\! 10^{-6}$ to $10^{-4}$ M$_{Earth}$ of 20 K dust at 50 AU. 
503: Observations with moderate S/N at 70 \micron\ translate to a maximum of 
504: $\sim\!\! 10^{-7}$ M$_{Earth}$ of 50 K dust at 5 AU, while 24 \micron\ observations 
505: limit warm, 250 K dust to $\sim\!\! 10^{-9}$ M$_{Earth}$ at 0.2 AU (again 
506: assuming micron-size dust in both cases).
507: 
508:      While fractional dust luminosities similar to that of $\beta$ Pic 
509: (L$_{d}$/L$_{*} = 3\times 10^{-3}$) would have been detected in all stars with 
510: their photospheres detected at 70 \micron, luminosities as low as that of Vega 
511: (L$_{d}$/L$_{*}$$\sim 2\times 10^{-5}$) could only be detected around stars 
512: with optimal data such as Lalande 21185.
513: 
514: \subsection{Comparison of M stars with F, G, and K Stars}\label{comp}
515: 
516: \subsubsection{Previous Work}\label{prev_work}
517: 
518:      Debris disk excesses in M stars have been elusive, partly because of the 
519: increase in detection difficulty with decreasing stellar luminosity (see 
520: Section \ref{quant_comp}). \citet{song02} and \citet{plav05} have searched 
521: for excesses from M stars from IRAS data complemented by ground based 
522: measurements. Both studies find no credible excesses around mature M stars, 
523: that is those not associated with known young clusters nor showing spectroscopic signs 
524: of youth. Our results confirm this conclusion, using the more sensitive and accurate 
525: {\it Spitzer} data. Using reprocessed IRAS data, \citet{back86}, 
526: reported an infrared excess from Ross 128 at 60 and 100 \micron. This was 
527: not confirmed by subsequent processing in the IRAS Faint Source Survey, and 
528: is also contradicted by the {\it Spitzer}/MIPS results. The 0.12 Jy excess they 
529: reported at 60 \micron\ is 8 times larger than our 3 sigma upper limit at 
530: 70~\micron, and is therefore confidently excluded by our data. Using {\it Spitzer} 
531: spectroscopy, \citet{jur04} report no excess for Ross 128. This leaves no 
532: known example of a mature main sequence M dwarf known to possess a 
533: far-infrared excess.
534: 
535:      It is interesting to compare the present results with similar studies of 
536: stars of other spectral types to see whether M star debris disks 
537: are less common than disks around earlier type stars. As a comparison 
538: sample, we have used both the study of F5-K5 main-sequence field stars 
539: discussed by \citet{bry05} and the similar set of stars studied by 
540: \citet{kim05}. We combine all the data without regard to whether they apply
541: to single stars, binaries, or stars with radial velocity planets. 
542: There are 69 stars in Bryden et al. and 35 in Kim et al., 
543: for a total of 104. Of these stars, 12 have detected excesses. We need
544: to add $\sim$2 more detections to allow for the weighted contribution of the
545: four stars with bright excesses detected by IRAS and not re-observed 
546: \citep[see][]{bry05}. We find that $13 \pm 4$\% of these stars have excesses 
547: greater than 100\%. Only one of the 69 stars in \citet{bry05}, HD 69830, 
548: shows evidence for a significant 24 \micron\ photometric excess more than 15\% 
549: above the photosphere. \citet{kim05} included only stars with cold excesses, 
550: so we cannot derive an independent statistic in this regard. However, it is 
551: clear that 24 \micron\ excesses are uncommon in solar type main sequence field stars. 
552: We find no convincing 24\micron\ excess in a sample of 56 M stars. 
553: 
554:      The current sample of 20 M dwarfs with $3\sigma$ or better 70 \micron\ detections is quite 
555: small for purposes of drawing confident statistical inferences of debris disk 
556: frequency. Binomial statistics indicate that our result of no disk detections 
557: in the 20 stars only allows us to derive an upper limit of 14\%, with 2$\sigma$ 
558: confidence, to the incidence of disks among M dwarfs within 5pc of the Sun. This upper limit
559: is right at the measured incidence level for more massive stars. 
560: Our upper limit is also consistent with a value of 13\% reported by \citet{les06}\ for 
561: 20 to 200 Myr old field M dwarfs. However, we note that our 5 pc sample 
562: is likely to contain a large fraction of stars older than 200 Myr and may not be directly 
563: comparable.
564: 
565:      \citet{song02} find that a number of extremely young M stars have 
566: detectable excesses (i.e., Hen 3-600 and AU Mic, both less than 15 M yr 
567: old). \citet{low05} report sensitive {\it Spitzer} measurements in the young 
568: TW Hya association where, of the 24 stars observed, 5 have significant 
569: excess at 24 \micron\ and 6 have excess at 70 \micron. \citet{chen05} confirm 
570: the excess in AU Mic. Thus, excesses are not uncommon in late-type stars, 
571: including M-type ones, at an age of $\sim$10 Myr. However, they virtually 
572: completely disappear in mature low-mass stars.
573: 
574: 
575: \subsubsection{Approach for Quantitative Comparison}\label{quant_comp}
576: 
577:      To expand on the results in the preceding section, we need to use an appropriate 
578: metric to characterize debris systems. We will adopt the 70 \micron\ fractional 
579: excess, F$_{excess}$/F$_{photosphere}$, where F$_{excess}$ is the flux 
580: density of the infrared excess and F$_{photosphere}$ that of the star itself, 
581: both at 70 \micron. We use the 70 \micron\ fractional excess as a proxy for the 
582: fractional luminosity, L$_{IR}$/L$_{star}$, because it should be representative of 
583: disk luminosity for objects with no excess at shorter wavelengths. 
584: 
585:      As stars become less luminous and of lower temperature, it becomes 
586: increasingly difficult to measure infrared excesses to a given limit in 
587: fractional luminosity. This issue can be understood by considering a 
588: set of stars with varying temperature. The total luminosity of these objects, 
589: and thus the dust luminosity, will vary as ${\rm T}^4 {\rm R}^2$ (where R is 
590: the stellar radius), but their photospheric outputs in the far infrared will be 
591: in the Rayleigh-Jeans regime and will only go as ${\rm T} {\rm R}^2$. Thus, for 
592: a given threshold for excess detection above the photosphere, the fractional 
593: luminosity threshold will go as T$^{-3}$.
594: 
595: \subsubsection{Application to M Star Sample}\label{Mstars}
596: 
597:      To obtain a quantitative measure of the average level of excess 
598: in the M star sample, we proceed in a manner analogous to stacking, a 
599: technique often used to enhance the average signal to noise ratio with 
600: observations of a suite of similar sources. Our various M stars 
601: have different expected signal levels and, as a result, the measurements 
602: are of intrinsically differing ratios of signal to noise. 
603: Therefore, we stack by taking the weighted 
604: average of the measured excess ratios at 70\micron. We eliminate from 
605: this average all objects with a nominal SNR $<$ 4. We adopt for the 
606: remaining sources a net error consisting of the statistical error 
607: indicated in Table 1, plus a systematic error of 6\% to reflect the 
608: scatter in measurements of bright objects at 70 $\mu $m with MIPS 
609: \citep[e.g.][]{gord05}. The result is a weighted average excess ratio 
610: at 70 \micron\ of $1.02 \pm 0.05$; that is, on average, there is no excess 
611: from these stars. We place a $2\sigma$ upper limit of 1.12 to the 
612: average excess ratio, or 0.12 to the excess (relative to the photospheric 
613: emission). We also compute an average effective temperature for this sample 
614: as a weighted average of the individual stellar effective temperatures 
615: where the weights are those used for the excess ratios. The result is 
616: T = 3555 K. 
617: 
618:      As a comparison sample, we use the FGK stars in \citet{bry05}. We add 
619: in the four additional stars listed by Bryden et al. but measured in 
620: other {\it Spitzer} programs at half weight, since Bryden et al. are 
621: reporting on about half of their sample. We also add in DY Eri and eta Cas 
622: from this program. The result is an average excess ratio of $1.37 \pm 0.033$. 
623: The average temperature of this sample computed in the same way 
624: as for the M stars is 5763 K. If we scale the average excess for this sample 
625: as T$^3$, we predict that the average for the M stars if they have the 
626: same fractional luminosity in infrared excess emission is 0.09, just within 
627: the upper limit from our observations of 0.12. 
628: 
629:      Another effect that might lead to more prominent debris excesses 
630: in M stars is the scaling of dust mass to produce a given excess level, 
631: as a function of stellar luminosity. For a given fractional luminosity upper 
632: limit at a given wavelength, the minimum mass limits derived from the M star 
633: observations reported here are lower than would be the corresponding limits from 
634: more luminous stars. The reason is that the fractional dust luminosity is 
635: proportional to the total solid angle subtended by the dust as viewed from 
636: the star and that the temperature of the dust increases with the star's 
637: luminosity but decreases with distance from the star. Thus, at a fixed dust 
638: temperature, an increase in stellar luminosity requires that the dust be more 
639: distant which, in turn, requires more dust to subtend enough solid angle to 
640: keep L$_{d}$/L$_{*}$ constant. In the simple case of a single fixed grain 
641: size, the number of grains to achieve a certain fractional luminosity at a 
642: given temperature scales with the luminosity of the star. Thus the minimum 
643: dust mass associated with an M star with $\sim 0.01\;{\rm L}_\odot$ solar luminosities 
644: would be $\sim 1$\% of the corresponding minimum mass associated with a 
645: solar mass star, again assuming the same limit to the fractional luminosity 
646: at a given wavelength. For typical M stars in the present sample, therefore, 
647: it is not surprising to note that the limits on 24 \micron\ excesses correspond to 
648: masses as low as 10$^{-9}$ earth masses, while at 70 \micron, the non-detections 
649: could imply masses below 10$^{-7}$ earth masses. Similar {\it Spitzer}/MIPS 
650: surveys for IR excess around FGK stars \citep[e.g.][]{bry05} can detect 
651: dust with lower luminosity relative to the central star 
652: (L$_{dust}$/L$_{star}$ as low as $5 \times 10^{-6}$), but are less sensitive in 
653: terms of overall dust mass. The corresponding limit for 
654: non-detections of 70 \micron\ excess around FGK stars is $>10^{-6}$ M$_{Earth}$, an 
655: order of magnitude above that for M stars. 
656: 
657: \subsubsection{Discussion}\label{complex}
658: 
659:      The fractional luminosity is the 
660: appropriate metric if we assume that hypothetical debris disks would have 
661: the same covering fraction independent of stellar type, thus absorbing and 
662: reradiating in the far infrared an equivalent fraction of the stellar output. 
663: Although the fractional luminosity provides a reasonable first-order metric 
664: to compare excesses in different stellar populations, there are many other 
665: considerations. For example, one would expect that cool, low-mass stars 
666: might be less effective at heating circumstellar dust than hotter ones, even 
667: after allowing for the difference in luminosity. The degree to which this 
668: expectation is realized depends on the optical properties of the debris disk 
669: particles. They are likely to be relatively large grain aggregates, whose 
670: behavior may differ substantially from that of typical interstellar grains. 
671: The grain loss mechanisms - photon pressure, Poyting-Robertson drag,
672: and corpuscular winds - are generally different in low mass stars than 
673: in high mass ones, and may cause significant differences in their disks
674: (e.g., strong winds may destroy them quickly). Also, our study 
675: probes debris disks at 70 \micron\ and hence at a 
676: temperature of about 50 K. The equilibrium distance from a star of luminosity 
677: L for material at a given temperature scales as the square root of L. Given 
678: the steep dependence of stellar luminosity on mass, we are probing 
679: distinctly different physical regions around the M stars in this sample than 
680: the regions probed in observations of solar-like stars. 
681: 
682: \section{Conclusions}\label{conclude}
683: 
684:      We observed a sample of 62 M stars to search for debris disk dust and to 
685: study photospheric models. We demonstrate a modified infrared flux method to 
686: determine the effective temperatures of these stars, taking advantage of the 
687: high accuracy measurements available at 24 \micron\ with {\it Spitzer}. Using these 
688: effective temperatures, the K$_S$ - [24] colors of the stars agree 
689: reasonably well with the theoretical models of Brott et al. (private 
690: communication). None of the stars in our sample has a high-weight detection 
691: of a mid- or far-infrared excess. In particular, the sample lacks the 
692: occasional stars with large excess that typifies samples of more massive 
693: stars. We compute a weighted average of the observed excess ratios at 70 \micron\ 
694: as a metric to characterize the excesses in this and other samples 
695: of stars. We find that the average excess in the M stars is at least four 
696: times less than the average in a sample of FGK stars. This limit is just at 
697: the level for equal fractional infrared luminosity in the two samples. 
698: We have set low limits ($10^{-4}$ to $10^{-9}$ M$_{Earth}$ depending on location) 
699: for the maximum mass of dust possible around our stars.
700: 
701: \acknowledgements
702: 
703:      This work is based, in part, on observations made with the {\it Spitzer} Space 
704: Telescope, which is operated by the Jet Propulsion Laboratory, California 
705: Institute of Technology under contract with NASA. Support for this work 
706: was provided by NASA through contract 1255094. 
707: This work also makes use of data products from the Two 
708: Micron All Sky Survey, which is a joint project of the University of 
709: Massachusetts and the Infrared Processing and Analysis Center/California 
710: Institute of Technology, funded by the National Aeronautics and Space 
711: Administration and the National Science Foundation. We would further like to 
712: thank Peter Hauschildt for supplying us with results from the latest GAIA model
713: spectra of late type stars.
714: 
715: \begin{thebibliography}{}
716: 
717: \bibitem[Allard, et al.(2001)]{all01} Allard, F., Hauschildt, P., Alexander, D. R., Tamanai, A., \& Schweitzer, A. 2001, \apj, 556, 357
718: \bibitem[Aumann, et al.(1984)]{au84} Aumann, H. H., et al. 1984, \apj, 278, L23
719: \bibitem[Backman, et al.(1986)]{back86} Backman, D. E., Gillett, F. C. \& Low, F. J. 1986, Advances in Space Research, 6, 43.
720: \bibitem[Beichman, et al.(2005a)]{bei05a} Beichman, C. A., et al. 2005a, \apj, 626, 1061
721: \bibitem[Beichman, et al.(2005b)]{bei05b} Beichman, C. A., etal. 2005b, \apj, 622, 1160
722: \bibitem[Bryden, et al.(2006)]{bry05} Bryden, G., et al. 2006, \apj, 636, 1098
723: \bibitem[Blackwell, et al.(1986)]{blac86} Blackwell, et al. 1986, \mnras, 221, 427
724: \bibitem[Brott \& Hauschildt(2005)]{brot05} Brott, I., \& Hauschildt, P. H. 2005, ``The Three-dimensional 
725: Universe with GAIA", (ESA SP-576), ed. C. Turon, K. E. O'Flaherty, M. A. C. Perryman, p. 565.
726: \bibitem[Burgasser, et al.(2002)]{bur02} Burgasser, A. T., et al. 2002, \apj, 564, 421
727: \bibitem[Butler, et al.(2004)]{but04} Butler, R. P., et al. 2004, \apj, 617, 580
728: \bibitem[Chen, et al.(2005)]{chen05} Chen, C., et al. 2005, \apj, 634, 1372
729: \bibitem[Cruz, et al.(2003)]{cruz03} Cruz K. L., et al. 2003, \aj, 126, 2421
730: \bibitem[Cutri, et al.(2003a)]{cut03a} Cutri, R. M., et al. 2003a,``The 2MASS All-Sky Catalog of Point Sources'', University of Massachusetts and Infrared Processing and Analysis Center (IPAC/California Institute of Technology); VizieR On-line Data Catalog: II/246
731: \bibitem[Cutri, et al.(2003b)]{cut03b} Cutri, R. M., et al. 2003b. ``The 2MASS All-Sky Data Release Explanatory 
732: Supplement: VI.4 2MASS Photometric System'', University of Massachusetts and Infrared Processing and 
733: Analysis Center (IPAC/California Institute of Technology)
734: \bibitem[Dawson \& de Robertis(2000)]{dr00} Dawson, P. C., and de Robertis, M. M. 2000, \aj, 120, 1532
735: \bibitem[Dole, et al.(2004)]{dol04} Dole, H. et al. 2004 \apjs, 154, 93
736: \bibitem[Engelbracht, et al.(2007)]{eng06} Engelbracht, et al. 2007 \pasp, in press
737: \bibitem[Dullemond \& Dominik(2005)]{dd05} Dullemond, C. P., \& Dominik, C. 2005 \aap, 434, 971
738: \bibitem[Gizis, et al.(2000)]{giz00} Gizis, J. E., et al. 2000 \aj, 120, 1085
739: \bibitem[Goldreich, et al.(2004)]{gold04} Goldreich, P., Lithwick, Y., \& Sari, R. 2004, \apj, 614, 497
740: \bibitem[Golimowski, et al.(2004)]{golim04} Golimowski, D. A., Leggett, S. K., Marley, M. S., Fan, X., Geballe, T. R., Knapp, G. R., Vrba, F. J. et al. 2004, \aj, 127, 3516
741: \bibitem[Gordon, et al.(2004)]{gord04} Gordon, K. D., et al. 2004, \procspie, 5487, 177
742: \bibitem[Gordon, et al.(2005)]{gord05} Gordon, K. D., et al. 2005, \pasp, 117, 503
743: \bibitem[Henry, et al.(1999)]{hen99} Henry, T. J., et al. 1999, \apj, 512, 864
744: \bibitem[Johnson (1983)]{jon83} Johnson, H. M. 1983, \apj, 273, 702
745: \bibitem[Jura, et al.(2004)]{jur04} Jura, M., et al. 2004, \apjs, 154, 453
746: \bibitem[Kalas, et al.(2004)]{kal04} Kalas, P., Liu, M. C. \& Matthews, B.C. 2004, Science, 303, 1990
747: \bibitem[Kenyon \& Bromley(2004)]{kb04} Kenyon, S. J., \& Bromley, B. C. 2004, \aj, 127, 513
748: \bibitem[Kim, et al.(2005)]{kim05} Kim, J. S., et al. 2005, \apj, 632, 659
749: \bibitem[Krist(2005)]{krist05} Krist, J. E. 2005, http://ssc.spitzer.caltech.edu/archanaly/contributed/stinytim.tar.gz
750: \bibitem[Lang(1991)]{lang91} Lang, K. R. 1991, Astrophysical Data, Springer-Verlag, New York
751: \bibitem[Leinhardt \& Richardson(2005)]{lr05} Leinhardt, Z. M., \& Richardson, D. C. 2005, \apj, 625, 427
752: \bibitem[Liu, et al.(2004)]{liu04} Liu, M.C., Matthews, B.C., Williams, J. P., and Kalas, P. G. 2004, \apj, 608, L526
753: \bibitem[Lestrade, et al.(2006)]{les06} Lestrade, J. F., Wyatt, M. C., Bertoldi, F., Dent, W. R. F. \& Menten, K. M. 2006, \aap, 460, 733
754: \bibitem[Low, et al.(2005)]{low05} Low, F. J., Smith, P. S., Werner, M. W., Chen, C., Krause, K., Jura, M., and 
755: Hines, D. C. 2005, \apj, 631, 1170
756: \bibitem[Mould \& Hyland(1976)]{mh76} Mould, J. R. \& Hyland, A. R. 1976, \apj, 208, 399
757: \bibitem[Patten, et al. (2006)]{pat06} Patten, B. M., et al. 2006, \apj, 651, 502
758: \bibitem[Persson, et al.(1977)]{per77} Persson, S. E., Aaronson, M. \& Frogel, J. A. 1977, \aj, 82, 729
759: \bibitem[Plavchan, et al.(2005)]{plav05} Plavchan, P., Jura, M. \& Lipscy, S. J. 2005, \apj, 631, 1161
760: \bibitem[Reid \& Hawley(2000)]{rh00} Reid, N. \& Hawley, S. L. 2000, ``New Light on Dark Stars," (Springer: New York)
761: \bibitem[Ribas(2003)]{rib03}Ribas, I. 2003, \aap, 400, 297
762: \bibitem[Rieke, et al.(2004)]{gr04} Rieke, G. H., et al. 2004, \apjs, 154, 25
763: \bibitem[Rieke, et al.(2005)]{gr05} Rieke, G. H., et al. 2005, \apj, 620, 1010
764: \bibitem[Song, et al.(2002)]{song02} Song, I., Weinberger, A. J., Becklin, E. E., Zuckerman, B. \& Chen, C. 2002, \aj, 124, 514
765: \bibitem[Stansberry, et al.(2007)]{stan07}Stansberry, J., et al. 2007, \pasp, in press
766: \bibitem[Su, et al.(2006)]{su06} Su, K., et al. 2006, \apj, 653, 675
767: \bibitem[Tokunaga(2000)]{tok00} Tokunaga, A. T. 2000, in \textit{Allen's Astrophysical Quantities}, 4th 
768: edition, ed. A.N. Cox, Springer-Verlag, NY, p. 143
769: \bibitem[Werner, et al.(2004)]{wer04} Werner, M. W., et al. 2004, \apjs, 154, 1
770: \end{thebibliography}
771: 
772: \clearpage
773: 
774: \begin{deluxetable}{llrllccc}
775: \tabletypesize{\scriptsize}
776: \tablecaption{Observational Data \label{data_table}}
777: \tablewidth{0pt}
778: \tablehead{
779: \colhead{Name} & \colhead{Gliese}& \colhead{LHS}& \colhead{SpT\tnm{1}}& \colhead{F$_{24}$(mJy)}& 
780: \colhead{F$_{70}$(mJy)\tnm{*}}& \colhead{F$_{160}$(mJy,3$\sigma $)}
781: }
782: \startdata
783: % name                      gliese     LHS    SpT           F24                      F70                160
784: AX Mic                  & GJ 825    & 66    & M0    & 504.00 $\pm$ 0.57       & 61.8 $\pm$ 3.8       & $<$75    \\
785: LHS 38                  & GJ 412 A  & 38    & M1    & 123.94 $\pm$ 0.21       & 15.8 $\pm$ 3.9\tnm{a}& $<$67    \\
786: GX And                  & GJ 15 A   & 3     & M1.5  & 249.45 $\pm$ 0.29       & 27.2 $\pm$ 2.2       & $<$102    \\
787: Kapteyn's Star          & GJ 191    & 29    & M1.5  &  98.46 $\pm$ 0.14       &  4.6 $\pm$ 2.4\tnm{b}& \nodata  \\
788: LHS 70                  & GJ 887    & 70    & M1.5  & 436.96 $\pm$ 0.43       & 47.3 $\pm$ 3.9       & $<$53    \\
789: Lalande 21185           & GJ 411    & 37    & M2    & 481.55 $\pm$ 0.40       & 43.0 $\pm$ 3.5       & $<$62    \\
790: LHS 1                   & GJ 1      & 1     & M3    & 159.13 $\pm$ 0.28       & 16.2 $\pm$ 3.2       & \nodata  \\
791: AD Leo                  & GJ 388    & 5167  & M3    & 151.62 $\pm$ 0.20       &  8.4 $\pm$ 2.3       & \nodata  \\
792: V2306 Oph               & GJ 628    & 419   & M3    & 104.83 $\pm$ 0.23       & 19.2 $\pm$ 2.8       & $<$154   \\
793: LHS 449                 & GJ 674    & 449   & M3    & 120.51 $\pm$ 0.22       & 10.4 $\pm$ 7.3\tnm{c}& $<$963   \\
794: LHS 450                 & GJ 687    & 450   & M3    & 160.17 $\pm$ 0.21       & 28.1 $\pm$ 3.4       & $<$83    \\
795: LHS 58                  & GJ 725 A  & 58    & M3    & 183    $\pm$ 3          & 31.9 $\pm$ 3.3\tnm{d}& $<$73\tnm{d}\\
796: LHS 3685                & GJ 832    & 3685  & M3    & 163.15 $\pm$ 0.18       & 19.5 $\pm$ 4.1       & $<$57    \\
797: LHS 3814                & GJ 860 A  & 3814  & M3    & 146.0  $\pm$ 0.4        & 117  $\pm$  61\tnm{e}& \nodata  \\
798: GQ And                  & GJ 15 B   &       & M3.5  &  49.46 $\pm$ 0.20       &  5.3 $\pm$ 1.9\tnm{f}& \nodata  \\
799: Luyten's Star           & GJ 273    & 33    & M3.5  & 127.16 $\pm$ 0.22       & 11.4 $\pm$ 3.7       & $<$71    \\
800: LHS 59                  & GJ 725 B  & 59    & M3.5  & 115    $\pm$ 3          & 11.0 $\pm$ 3.3\tnm{d}& \nodata\tnm{d}\\
801: V1216 Sgr               & GJ 729    & 3414  & M3.5  &  83.67 $\pm$ 0.27       & 17.0 $\pm$ 5.7\tnm{g}& $<$318   \\
802: IL Aqr                  & GJ 876 A  & 530   & M3.5  & 108.27 $\pm$ 0.16       &  9.2 $\pm$ 2.2       & $<$67    \\
803: Ross 128                & GJ 447    & 315   & M4    &  66.32 $\pm$ 0.24       &  7.5 $\pm$ 5.0       & $<$61    \\
804: Barnard's Star          & GJ 699    & 57    & M4    & 188.45 $\pm$ 0.25       & 29.1 $\pm$ 4.4       & $<$141    \\
805: YZ Cet                  & GJ 54.1   & 138   & M4.5  &  33.69 $\pm$ 0.19       &  9.0 $\pm$ 2.0\tnm{h}& \nodata  \\
806: TZ Ari                  & GJ 83.1   & 11    & M4.5  &  27.71 $\pm$ 0.21       &               \tnm{i}& \nodata  \\
807: DY Eri C                & GJ 166 C  & 25    & M4.5  &  49.05 $\pm$ 0.14       &  4.0 $\pm$ 2.0       & \nodata  \\
808: V577 Mon A              & GJ 234 A  & 1849  & M4.5  &  77.97 $\pm$ 0.20       & 12.2 $\pm$ 1.9\tnm{j}& $<$128    \\
809: LHS 451                 & GJ 682    & 451   & M4.5  &  62.15 $\pm$ 0.23       &               \tnm{k}& \nodata  \\
810: EZ Aqr                  & GJ 866 A  & 68    & M5    &  78.73 $\pm$ 0.19       & -4.2 $\pm$ 3.5       & \nodata  \\
811: LHS 39                  & GJ 412 B  & 39    & M5.5  &   9.89 $\pm$ 0.16       & 10.1 $\pm$ 3.4\tnm{l}& \nodata  \\
812: FL Vir                  & GJ 473A   & 333   & M5.5  &  49.4  $\pm$ 0.4        &  3.0 $\pm$ 2.0       & \nodata  \\
813: Proxima Cen             & GJ 551    & 49    & M5.5  & 235.06 $\pm$ 0.37       &               \tnm{m}& \nodata  \\
814: Ross 248                & GJ 905    & 549   & M5.5  &  52.78 $\pm$ 0.16       & 10.7 $\pm$ 2.0\tnm{n}& \nodata  \\
815: LHS 2                   & GJ 1002   & 2     & M5.5  &  14.60 $\pm$ 0.18       &  4.1 $\pm$ 2.6       & \nodata  \\
816: LHS 1565                & GJ 1061   & 1565  & M5.5  &  29.65 $\pm$ 0.22       &               \tnm{p}& $<$63    \\
817: V1581 Cyg A             & GJ 1245 A & 3494  & M5.5  &  27.0  $\pm$ 1.6        & -12  $\pm$  10\tnm{q}& $<$387   \\
818: LHS 1375                & GJ 3146   & 1375  & M5.5 s& 3.31 $\pm $ 0.13        & \nodata              & \nodata  \\
819: LHS 288                 & GJ 3618   & 288   & M5.5  &  10.71 $\pm$ 0.14       &               \tnm{r}& \nodata  \\
820: UV/BL Cet               & GJ 65 B   & 10    & M6    &  95.74 $\pm$ 0.22\tnm{r}&  5.6 $\pm$ 1.9\tnm{s}& $<$68    \\
821: GL 316.1                & GL 316.1  & 2034  & M6   s& 1.35 $\pm $ 0.11        & \nodata              & \nodata  \\
822: Wolf 359                & GJ 406    & 36    & M6    &  49.90 $\pm$ 0.30       &               \tnm{t}& \nodata  \\
823: V1581 Cyg B             & GJ 1245 B & 3495  & M6    &  18.0  $\pm$ 0.7        & -12  $\pm$  10\tnm{q}& \nodata  \\
824: GJ 283B                 & GJ 283 B  & 234   & M6.5 s& 1.05 $\pm $ 0.13        & \nodata              & \nodata  \\
825: DX Cnc                  & GJ 1111   & 248   & M6.5  &  17.18 $\pm$ 0.22       &  6.0 $\pm$ 3.0\tnm{u}& $<$101    \\
826: LHS 292                 & GJ 3622   & 292   & M6.5  &   9.41 $\pm$ 0.13       &  1.1 $\pm$ 1.9\tnm{v}& \nodata  \\
827: LHS 2930                & GJ 3855   & 2930  & M6.5 s& 1.63 $\pm $ 0.1         & \nodata              & \nodata  \\
828: GJ 644C                 & GJ 644 C  & 429   & M7    & 4.17 $\pm $ 0.12        & \nodata              & \nodata  \\
829: LHS 3003                & GJ 3877   & 3003  & M7    & 3.75 $\pm $ 0.11        & \nodata              & \nodata  \\
830: 2MASS J09522188-1924319 &           &       & M7   s& 0.53 $\pm $ 0.07        & \nodata              & \nodata  \\
831: 2MASS J23062928-0502285 &           &       & M7.5 s& 1.35 $\pm $ 0.12        & \nodata              & \nodata  \\
832: LP 044-162              &           &       & M7.5\tnm{2}  & 1.08 $\pm $ 0.08 & \nodata              & \nodata  \\
833: LP 349-025              &           &       & M8   s& 2.31 $\pm $ 0.12        & \nodata              & \nodata  \\
834: LP 326-021              &           &       & M8   s& 0.89 $\pm $ 0.09        & \nodata              & \nodata  \\
835: LP 412-031              &           &       & M8   s& 0.74 $\pm $ 0.07        & \nodata              & \nodata  \\
836: LP 771-021              &           &       & M8   s& 0.41 $\pm $ 0.05        & \nodata              & \nodata  \\
837: LHS 2397 A              & GJ 3655   & 2397A & M8   s& 0.99 $\pm $ 0.06        & \nodata              & \nodata  \\
838: ESO 207-61              &           &       & M8   s& 0.16 $\pm $ 0.04        & \nodata              & \nodata  \\
839: TVLM 513-46546          &           &       & M8.5 s& 1.03 $\pm $ 0.1         & \nodata              & \nodata  \\
840: LP 944-20               & \nodata   &       & M9 s  &   3.95 $\pm$ 0.05       &  7   $\pm$   3\tnm{w}& \nodata  \\
841: LHS 2065                & GJ 3517   & 2065  & M9    & 1.9  $\pm $ 0.09        & \nodata              & \nodata  \\
842: TVLM 868-110639         &           &       & M9   s& 0.68 $\pm $ 0.03        & \nodata              & \nodata  \\
843: BRI 0021-0214           &           &       & M9.5 s& 1.26 $\pm $ 0.05        & \nodata              & \nodata  \\
844: 2MASSW J1733189+463359  &           &       & M9.5 s& 0.28 $\pm $ 0.03        & \nodata              & \nodata  \\
845: DENIS-P J0021.0-4244    &           &       & M9.5 s& 0.26 $\pm $ 0.04        & \nodata              & \nodata  \\
846: \enddata
847: 
848: \tnt{*}{Entries with an ellipsis (\nodata) in the 70 \micron\ column are from {\it Spitzer} program 56.}
849: \tnt{1}{Spectral types are from the NStars data base except for those marked ``s'' which are from Simbad.}
850: \tnt{2}{See text for LP 044-162 spectral type source.}
851: \tnt{a}{{L}HS 38: Some cirrus at 70 \micron}
852: \tnt{b}{Kapteyn's: No source apparent at 70 \micron. Photometry at reference position.}
853: \tnt{c}{{L}HS 449: Bad cirrus confusion at 70 \micron}
854: \tnt{d}{{L}HS 58 \& 59: 14\arcsec\  separation. 70 \micron\ fluxes from deblending. 160 \micron\ upper limit is for the combination.}
855: \tnt{e}{{L}HS 3814: Bad cirrus, especially at 70 \micron}
856: \tnt{f}{GQ And: 35\arcsec\  separation from GX And: possible slight blending at 70 \micron.}
857: \tnt{g}{V1216 Sgr: Bad Cirrus at 70 \micron}
858: \tnt{h}{YZ Cet: Diffuse source, probable background contamination}
859: \tnt{i}{TZ Ari: Bright background source 20\arcsec\  north accounts for flux}
860: \tnt{j}{V577 Mon A: 2\arcsec\  separation from GJ23B. Photometry includes both sources in both bands}
861: \tnt{k}{{L}HS 451: Bad cirrus confusion: no source apparent at 70 \micron}
862: \tnt{l}{{L}HS 39: 70 \micron\ photometry done at 24 \micron\ position. 32\arcsec\  separation from LHS 38}
863: \tnt{m}{Proxima Cen: Very bad cirrus at 70 \micron.}
864: \tnt{n}{Ross 248: 70 \micron\ source offset slightly from 24 \micron\ source; possible contamination.}
865: \tnt{p}{{L}HS 1565: Bright background galaxy. No 70 \micron\ photometry possible}
866: \tnt{q}{V1581 Cyg A \& B: 8\arcsec\ separation and bad cirrus. 70 \micron\ photometry includes both sources}
867: \tnt{r}{{L}HS 288: Not on the 70 \micron\ image (error in requested pointing)}
868: \tnt{s}{UV Cet/BL Cet: 2\arcsec\  separation. Photometry is for both sources in both bands}
869: \tnt{t}{Wolf 359: Not on the 70 \micron\ image (error in requested pointing)} % same as note "e"
870: \tnt{u}{DX Cnc: No source apparent at 70 \micron. Photometry at reference position.}
871: \tnt{v}{{L}HS 292: 70 um photometry at 24 \micron\ position. Nearby 70 \micron\ source.}
872: \tnt{w}{{L}P 944-20: Faint source at 70 \micron; another 70 \micron\ source 50\arcsec\  north}
873: \end{deluxetable}
874: \clearpage
875: 
876: 
877: \begin{deluxetable}{lccccccccccc}
878: \tabletypesize{\scriptsize}
879: \rotate
880: \tablecaption{Colors and Derived Quantities \label{color_table}}
881: \tablewidth{0pt}
882: \tablehead{
883: \colhead{Name} & \colhead{Dist(pc)}& \colhead{SpT}& \colhead{T$_{eff}$}& \colhead{T$_{ir\kern-0.1em fm}$}& 
884: \colhead{B.C.}& \colhead{K$_S$\tnm{*}} & \colhead{K$_S$-[24]\tnm{\dag}} & \colhead{P$_{24}$\tnm{\ddag}} & 
885: \colhead{F$_{24}$/P$_{24}\tnm{\ddag}$} & \colhead{P$_{70}$\tnm{\ddag}} & \colhead{F$_{70}$/P$_{70}$\tnm{\ddag}}
886: }
887: \startdata
888: %Name               Dist(pc)   SpT   Teff   Tirfm   B.C.        K_s$^1$            K_s-[24]$^2$       p24     f/p24     p70     f/p70
889: 
890: AX Mic             & 3.95   & M0   & 3800 &  3987 & 2.370 &  $3.07\pm 0.02$\tnm{a}& $0.19\pm 0.06$ & 502.04 &  1.00 &  58.37 &   1.06   \\
891: LHS 38             & 4.85   & M1   & 3680 &  3591 & 2.530 &  $4.77\pm 0.02$       & $0.37\pm 0.06$ & 116.58 &  1.06 &  14.44 &   1.09   \\
892: GX And             & 3.56   & M1.5 & 3605 &  3646 & 2.470 &  $4.02\pm 0.02$       & $0.38\pm 0.06$ & 228.78 &  1.09 &  29.03 &   0.94   \\
893: Kapteyn's Star     & 3.92   & M1.5 & 3605 &  3554 & 2.474 &  $5.05\pm 0.02$       & $0.40\pm 0.06$ &  88.49 &  1.11\tnm{d} &  11.46 & $< 0.6$\tnm{d}  \\
894: LHS 70             & 3.29   & M1.5 & 3605 &  3636 & 2.530 &  $3.36\pm 0.02$\tnm{a}& $0.33\pm 0.06$ & 421.41 &  1.04 &  50.86 &   0.93   \\
895: Lalande 21185      & 2.54   & M2   & 3530 &  3469 & 2.530 &  $3.35\pm 0.02$\tnm{b}& $0.42\pm 0.06$ & 432.61 &  1.11\tnm{d} &  56.10 &   0.77\tnm{d}   \\
896: LHS 1              & 4.36   & M3   & 3380 &  3563 & 2.530 &  $4.52\pm 0.2 $       & $0.39\pm 0.21$ & 148.05 &  1.07 &  18.55 &   0.87   \\
897: AD Leo             & 4.89   & M3   & 3380 &  3360 & 2.648 &  $4.59\pm 0.02$       & $0.41\pm 0.06$ & 148.76 &  1.02 &  17.73 &   0.47   \\
898: V2306 Oph          & 4.24   & M3   & 3380 &  3265 & 2.710 &  $5.08\pm 0.02$       & $0.49\pm 0.06$ &  99.15 &  1.06 &  12.28 &   1.56   \\
899: LHS 449            & 4.54   & M3   & 3380 &  3471 & 2.580 &  $4.86\pm 0.02$       & $0.42\pm 0.06$ & 112.58 &  1.07 &  14.07 & $< 1.6$  \\
900: LHS 450            & 4.54   & M3   & 3380 &  3332 & 2.710 &  $4.55\pm 0.02$       & $0.43\pm 0.06$ & 155.96 &  1.03 &  18.74 &   1.50   \\
901: LHS 58             & 3.53   & M3   & 3380 &  3369 & 2.650 &  $4.43\pm 0.02$       & $0.45\pm 0.06$ & 171.81 &  1.07 &  21.40 &   1.49   \\
902: LHS 3685           & 4.93   & M3   & 3380 &  3657 & 2.470 &  $4.47\pm 0.02$\tnm{c}& $0.37\pm 0.06$ & 150.66 &  1.08 &  18.98 &   1.03   \\
903: LHS 3814           & 4.03   & M3   & 3380 &  3211 & 2.648 &  $4.78\pm 0.03$       & $0.56\pm 0.06$ & 132.40 &  1.10 &  17.13 & $<10.7$  \\
904: GQ And             & 3.56   & M3.5 & 3280 &  3157 & 2.708 &  $5.95\pm 0.02$       & $0.55\pm 0.06$ &  46.12 &  1.07 &   5.81 & $< 1.0$  \\
905: Luyten's Star      & 3.79   & M3.5 & 3280 &  3273 & 2.710 &  $4.86\pm 0.02$       & $0.49\pm 0.06$ & 119.93 &  1.06 &  14.90 &   0.77   \\
906: LHS 59             & 3.53   & M3.5 & 3280 &  3239 & 2.710 &  $5.00\pm 0.02$       & $0.52\pm 0.06$ & 106.87 &  1.08 &  13.48 &   0.82   \\
907: V1216 Sgr          & 2.97   & M3.5 & 3280 &  3215 & 2.710 &  $5.37\pm 0.02$       & $0.54\pm 0.06$ &  76.76 &  1.09 &   9.81 & $< 1.8$  \\
908: IL Aqr             & 4.70   & M3.5 & 3280 &  3240 & 2.760 &  $5.01\pm 0.02$       & $0.46\pm 0.06$ & 105.85 &  1.02 &  12.69 &   0.72   \\
909: Ross 128           & 3.35   & M4   & 3180 &  3130 & 2.760 &  $5.65\pm 0.02$       & $0.57\pm 0.06$ &  61.53 &  1.08 &   7.79 & $< 1.9$  \\
910: Barnard's Star     & 1.83   & M4   & 3180 &  3039 & 2.850 &  $4.52\pm 0.02$       & $0.57\pm 0.06$ & 181.72 &  1.04 &  22.19 &   1.31   \\
911: YZ Cet             & 3.72   & M4.5 & 3105 &  3052 & 2.810 &  $6.42\pm 0.02$       & $0.60\pm 0.06$ &  31.38 &  1.07 &   3.97 &   2.27   \\
912: TZ Ari             & 4.45   & M4.5 & 3105 &  2995 & 2.810 &  $6.65\pm 0.02$       & $0.62\pm 0.06$ &  26.11 &  1.06 &   3.27 & \nodata  \\
913: DY Eri C           & 5.03   & M4.5 & 3105 &  3058 & 2.810 &  $5.96\pm 0.02$       & $0.55\pm 0.06$ &  47.80 &  1.03 &   5.77 & $< 1.0$  \\
914: V577 Mon A         & 4.09   & M4.5 & 3105 &  3074 & 2.810 &  $5.49\pm 0.02$       & $0.59\pm 0.06$ &  73.14 &  1.07 &   9.17 &   1.33   \\
915: LHS 451            & 5.01   & M4.5 & 3105 &  3302 & 2.710 &  $5.61\pm 0.02$       & $0.46\pm 0.06$ &  59.43 &  1.05 &   7.28 & \nodata  \\
916: EZ Aqr             & 3.45   & M5   & 3030 &  2855 & 2.943 &  $5.54\pm 0.02$       & $0.65\pm 0.06$ &  78.21 &  1.01 &   9.31 & $< 1.1$  \\
917: LHS 39             & 4.85   & M5.5 & 2940 &  2811 & 2.943 &  $7.84\pm 0.03$       & $0.69\pm 0.06$ &   9.65 &  1.03 &   1.17 & $< 9.0$  \\
918: FL Vir             & 4.39   & M5.5 & 2940 &  2866 & 2.943 &  $6.04\pm 0.02$       & $0.63\pm 0.06$ &  49.04 &  1.00 &   5.80 & $< 1.0$  \\
919: Proxima Cen        & 1.30   & M5.5 & 2940 &  2855 & 2.943 &  $4.35\pm 0.02$\tnm{a}& $0.64\pm 0.06$ & 234.04 &  1.00 &  27.80 & \nodata  \\
920: Ross 248           & 3.16   & M5.5 & 2940 &  2928 & 2.940 &  $5.93\pm 0.02$       & $0.60\pm 0.06$ &  52.46 &  1.01 &   6.23 &   1.72   \\
921: LHS 2              & 4.69   & M5.5 & 2940 &  2792 & 2.943 &  $7.44\pm 0.02$       & $0.72\pm 0.06$ &  14.10 &  1.04 &   1.73 & $< 4.5$  \\
922: LHS 1565           & 3.69   & M5.5 & 2940 &  2879 & 2.940 &  $6.61\pm 0.02$       & $0.66\pm 0.06$ &  28.80 &  1.03 &   3.50 & \nodata  \\
923: V1581 Cyg A        & 4.54   & M5.5 & 2940 &  2756 & 2.940 &  $6.85\pm 0.02$       & $0.79\pm 0.08$ &  24.82 &  1.09 &   3.20 & $<10.3$  \\
924: LHS 1375           & 8.50   & M5.5 & 2940 &  2821 & 3.034 &  $8.98\pm 0.05$       & $0.65\pm 0.09$ &   3.36 &  0.99 &        &          \\
925: LHS 288            & 4.49   & M5.5 & 2940 &  2832 & 2.943 &  $7.73\pm 0.03$       & $0.67\pm 0.06$ &  10.54 &  1.02 &   1.27 & \nodata  \\
926: UV/BL Cet          & 2.68   & M6   & 2850 &  2874 & 2.940 &  $5.34\pm 0.02$       & $0.66\pm 0.06$ &  93.03 &  1.03 &  11.32 & $< 0.5$  \\
927: GL 316.1           & 14.06  & M6   & 2850 &  2724 & 3.034 & $10.05\pm 0.05$       & $0.74\pm 0.12$ &   1.33 &  1.02 &        &          \\
928: Wolf 359           & 2.39   & M6   & 2850 &  2857 & 2.943 &  $6.08\pm 0.02$       & $0.69\pm 0.06$ &  47.51 &  1.05 &\nodata & \nodata  \\
929: V1581 Cyg B        & 4.54   & M6   & 2850 &  2602 & 3.030 &  $7.39\pm 0.02$       & $0.88\pm 0.07$ &  16.89 &  1.07 &   2.14 & $<15.4$  \\
930: GJ 283B            & 8.90   & M6.5 & 2750 &\tnm{e}&\nodata&  $9.26\pm 0.05$       &$-0.32\pm 0.15$ &   2.48 &  0.42\tnm{e} &        &          \\
931: DX Cnc             & 3.63   & M6.5 & 2750 &  2746 & 3.030 &  $7.26\pm 0.02$       & $0.71\pm 0.06$ &  17.12 &  1.00 &   2.04 & $< 4.4$  \\
932: LHS 292            & 4.54   & M6.5 & 2750 &  2675 & 3.069 &  $7.93\pm 0.03$       & $0.73\pm 0.06$ &   9.67 &  0.97 &   1.12 & $< 5.1$  \\
933: LHS 2930           & 9.63   & M6.5 & 2750 &  2711 & 3.069 &  $9.79\pm 0.05$       & $0.69\pm 0.10$ &   1.70 &  0.96 &        &          \\
934: GJ 644C            & 6.45   & M7   & 2650 &  2700 & 3.060 &  $8.82\pm 0.05$       & $0.74\pm 0.08$ &   4.19 &  0.99 &        &          \\
935: LHS 3003           & 6.40   & M7   & 2650 &  2704 & 3.060 &  $8.93\pm 0.05$       & $0.73\pm 0.08$ &   3.78 &  0.99 &        &          \\
936: 2MASS J0952-1924   &\nodata & M7   & 2650 &\tnm{e}&\nodata& $10.87\pm 0.05$       & $0.55\pm 0.16$ &   0.67 &  0.79\tnm{e} &        &          \\
937: 2MASS J2306-0502   &\nodata & M7.5 & 2550 &  2479 & 3.070 & $10.30\pm 0.05$       & $0.99\pm 0.12$ &   1.26 &  1.07 &        &          \\
938: LP 044-162         & 12.5   & M7.5 & 2750 &  2589 & 3.070 & $10.40\pm 0.05$       & $0.85\pm 0.11$ &   1.06 &  1.02 &        &          \\
939: LP 349-025         & 7.8    & M8   & 2450 &  2570 & 3.100 &  $9.57\pm 0.05$       & $0.84\pm 0.09$ &   2.30 &  1.00 &        &          \\
940: LP 326-021         &\nodata & M8   & 2450 &  2559 & 3.100 & $10.62\pm 0.05$       & $0.86\pm 0.13$ &   0.88 &  1.01 &        &          \\
941: LP 412-031         & 14.64  & M8   & 2450 &  2728 & 3.056 & $10.64\pm 0.05$       & $0.68\pm 0.13$ &   0.77 &  0.96 &        &          \\
942: LP 771-021         & 16.23  & M8   & 2450 &  2591 & 3.100 & $11.42\pm 0.05$       & $0.82\pm 0.15$ &   0.41 &  0.99 &        &          \\
943: LHS 2397 A         & 14.29  & M8   & 2450 &  2380 & 3.100 & $10.74\pm 0.05$       & $1.09\pm 0.10$ &   0.92 &  1.08 &        &          \\
944: ESO 207-61         & 18.48  & M8   & 2450 &\tnm{e}&\nodata& $12.11\pm 0.05$       & $0.49\pm 0.28$ &   0.25 &  0.65\tnm{e} &        &          \\
945: TVLM 513-46546     & 9.82   & M8.5 & 2350 &  2404 & 3.120 & $10.71\pm 0.05$       & $1.11\pm 0.13$ &   0.92 &  1.12 &        &          \\
946: LP 944-20          & 4.97   & M9   & 2250 &  2123 & 3.150 &  $9.55\pm 0.02$       & $1.41\pm 0.06$ &   3.58 &  1.10 &   0.48 & $<18.8$  \\
947: LHS 2065           & 8.53   & M9   & 2250 &  2420 & 3.150 &  $9.94\pm 0.05$       & $1.00\pm 0.09$ &   1.85 &  1.03 &        &          \\
948: TVLM 868-110639    & 17.39  & M9   & 2250 &  2193 & 3.150 & $11.35\pm 0.05$       & $1.30\pm 0.09$ &   0.63 &  1.08 &        &          \\
949: BRI 0021-0214      & 12.12  & M9.5 & 2150 &  2284 & 3.180 & $10.54\pm 0.05$       & $1.16\pm 0.08$ &   1.21 &  1.04 &        &          \\
950: 2MASSW J1733+4633  &\nodata & M9.5 & 2150 &  2512 & 3.180 & $11.89\pm 0.05$       & $0.87\pm 0.14$ &   0.28 &  0.99 &        &          \\
951: DENIS-P J0021.0-4244&\nodata & M9.5 & 2150 &\tnm{d}&\nodata& $12.30\pm 0.05$       & $1.20\pm 0.18$ &   0.22 &  1.21\tnm{d} &        &          \\
952: \enddata
953: 
954: \tnt{*}{K$_S$ is the 2MASS magnitude. K magnitudes footnoted $^a$, $^b$ and $^c$ were converted to K$_S$ using the 
955:         2MASS-CIT transformation in \citet{cut03b}}
956: \tnt{\dag}{24um zero point magnitude taken as 7.14 Jy.}
957: \tnt{\ddag}{Predicted values of the flux at 24 and 70 \micron, P$_{24}$ and P$_{70}$ are derived as described in the text, 
958: except that footnoted values use T$_{eff}$ instead of T$_{ir\kern-0.1em fm}$ Upper limits are at the 2 $\sigma$ level.}
959: \tnt{a}{Photometry from Mould, J.R. \& Hyland, A.R., 1976, \apj, 308, 399} 
960: \tnt{b}{Photometry from Persson et al., 1977, \aj, 82, 9}
961: \tnt{c}{Photometry from Leggett, S.K. 1992, \apjs, 82, 351}
962: \tnt{d}{Excluded from K$_S$-[24] calculation due to low SNR, but fits the trend.}
963: \tnt{e}{Excluded from K$_S$-[24] calculation due to low SNR, and is highly discrepant.}
964: 
965: \end{deluxetable}
966: \clearpage
967: 
968: \begin{figure}
969: \epsscale{1.0}
970: \plotone{f1}
971: \caption{
972: K$_S$-[24] colors vs. V - K$_S$.  The large symbol is for V1581 Cyg B, the only object
973: with a possible excess at 24$\mu$m identified through this test. 
974: \label{V-K_fig}}
975: \end{figure}
976: \clearpage
977: 
978: 
979: \begin{figure}
980: \epsscale{1.0}
981: \plotone{f2}
982: \caption{
983: K$_S$-[24] colors vs. T$_{eff}$ and T$_{ir\kern-0.1em fm}$.  The open circles show T$_{eff}$ from the 
984: spectral type, while the filled boxes are the result of deriving T$_{ir\kern-0.1em fm}$ from the 
985: infrared flux method, as described in the text. Unresolved binaries are indicated by larger 
986: circles and young objects by open squares. Cases with large unexplained differences are 
987: connected with arrows from the spectral estimate to the infrared flux method one.
988: The thin solid line shows the prediction from the atmospheric models of \citet{brot05}. 
989: The thick solid line is the fit to K$_S$-[24]vs. T$_{ir\kern-0.1em fm}$ derived in this paper. The 
990: dashed line shows the color of a black body.
991: \label{K-24_fig}}
992: \end{figure}
993: \clearpage
994: 
995: \begin{figure}
996: \epsscale{1.0}
997: \plotone{f3}
998: \caption{
999: The ratio of measured to predicted 70\micron\ flux vs. the effective temperature of the 
1000: sources determined from the infrared flux method. Only sources that were unambiguously 
1001: detected at 70 \micron\ are included (i.e. SNR $> 3$ and no obvious indications of 
1002: contamination from background sources; see notes to Table \ref{color_table}). 
1003: The 70 \micron\ prediction is extrapolated from the measured 24 \micron\ flux using a 
1004: black body at T$_{ir\kern-0.1em fm}$ for each star, as described in the text.
1005: \label{70_ratio_fig}}
1006: \end{figure}
1007: \clearpage
1008: 
1009: 
1010: \begin{figure}
1011: \epsscale{1.0}
1012: %\plotone{f4}
1013: \includegraphics[scale=0.6, angle=-90]{f4}
1014: \caption{
1015: Detection limits for circumstellar dust. The MIPS detection levels are 
1016: shown for two stars as a function of dust luminosity and temperature. 
1017: The light grey region is ruled out by the detection upper limits for 
1018: Lalande 21185 (solid line), a relatively bright star with a clean 
1019: 160 \micron\ image, while the dark grey region is ruled out for V1216 Sgr 
1020: (limits shown as a dotted line), a star located in a very noisy 160 \micron\ 
1021: field. Both stars have 1$\sigma$ errors of $\sim $5\% at 24 \micron, but their 
1022: relative accuracies diverge at longer wavelengths; 1$\sigma$ errors at 
1023: 160 \micron\ are 0.9 and 70 times the stellar photosphere for Lalande 21185 
1024: and V1216 Sgr respectively. In both cases, 3 $\sigma $ limits are plotted. 
1025: \label{dust_fig}}
1026: \end{figure}
1027: 
1028: 
1029: \end{document}
1030: