1: \documentclass[12pt,preprint]{aastex}
2:
3: \shorttitle{OPTICAL VARIABILITY OF L DWARFS}
4: \shortauthors{Maiti et al.}
5: \newcommand{\beqna}{\begin{eqnarray}}
6: \newcommand{\eeqna}{\end{eqnarray}}
7: \begin{document}
8:
9: \title{OBSERVATION OF R-BAND VARIABILITY OF L DWARFS}
10: \author{M. Maiti\footnote{e-mail: mith@iiap.ernet.in}, S. Sengupta,
11: P. S. Parihar \and G. C. Anupama}
12: \affil{Indian Institute of Astrophysics, Koramangala, Bangalore-560 034}
13:
14: \begin{abstract}
15: We report, for the first time, photometric variability of L dwarfs in $R$ band.
16: Out of three L1 dwarfs (2MASS 1300+19, 2MASS 1439+19, and 2MASS
17: 1658+70) observed, we have detected R band variability in 2MASS 1300+19 and
18: 2MASS 1439+19. The objects exhibit variability of amplitude ranging from 0.01
19: mag to 0.02 mag. Object 2MASS 1658+70, turns out to be non-variable
20: in both $R$ and $I$ band. However, more observations are needed to infer its
21: variability. No periodic behaviour in the variability is found from the two
22: L1 dwarfs that are variable. All the three L1 dwarfs have either negligible
23: or no $H_{\alpha}$ activity. In the absence of any direct evidence for the
24: presence of sufficiently strong magnetic field, the detection of polarization
25: at the optical favors the presence of dust in the atmosphere of L dwarfs.
26: We suggest that the observed $R$ band photometric variability is most likely
27: due to atmospheric dust activity.
28:
29: \end{abstract}
30:
31: \keywords{stars:atmosphere --- stars: low-mass, brown dwarfs}
32:
33: \section{INTRODUCTION}
34: L dwarfs are ultra-cool objects with effective temperature
35: ranging between 2200 K and 1400K. They are characterized by the presence of
36: condensates in their atmosphere. Due to incomplete gravitational settling,
37: dust in the atmosphere of L dwarfs could be detectable in the optical. Dust
38: cloud can play a potential role in making the object variable. Time-resolved
39: photometric variability
40: of a large number of L dwarfs has been reported by Bailer-Jones \& Mundt
41: (2001a,b), Martin, Osorio, \& Lehto (2001), Gelino et al. (2002), Clarke,
42: Oppenheimer \& Tinney (2002a),
43: Osorio, Caballero, Bejar \& Rebolo (2003), Bailer-Jones \& Lamm (2003).
44: However, all these observations were made in the $I$ and $J$ bands.
45: Clarke, Tinney \& Covey (2002b) reported variability from ultra-cool dwarfs
46: by using a non-standard filter with the effective wavelength similar to that
47: of $I$ band. Enoch, Brown \& Burgasser (2003) reported evidence of
48: variability in $K_s$ band from a few
49: L and T dwarfs. These investigations provide much insight on the atmospheric
50: activities, especially the presence of dust clouds.
51:
52: Sengupta \& Krishan (2001) argued that the presence of dust
53: could give rise to a detectable amount of linear polarization in the optical
54: from L dwarfs. This was observationally verified by Menard, Delfosse \& Monin
55: (2002) who detected non-zero linear polarization at red (0.768 $\mu$~m) from
56: a few L dwarfs. In the absence of any direct evidence of sufficiently strong
57: magnetic field, observation of linear polarization strongly favours the
58: presence of dust in the atmosphere of L dwarfs and single dust scattering
59: model could explain the observed polarization (Sengupta 2003).
60: It should also be mentioned here that rigorous theoretical analysis
61: (see Burrows et al. 2001 and references therein; Tsuji et al. 2004) of the
62: continuum spectra implies the presence of dust in the visible region of L
63: dwarfs.
64:
65: Detection of non-zero polarization at the optical band that may arise from
66: dust scattering urges the necessity of investigating $R$ band variability by
67: dust activity in L dwarfs. In the present letter we, for the first time,
68: report differential photometric variability in $R$ band from a few L dwarfs.
69: The results could be a good complement to the polarization observations as
70: dust clouds play a crucial role in both cases. A detailed theoretical
71: investigation on polarization by single dust scattering of L dwarfs with fixed
72: rotational velocity (Sengupta \& Kwok 2004) shows that the degree of
73: linear polarization peaks at L1 spectral
74: type. This motivates us to concentrate on the observation of L1 dwarfs. We
75: have detected photometric variability from two L1 dwarfs. In the next
76: section we describe the observation and data reduction procedure followed.
77: The results are presented and discussed in section~3 followed by conclusions.
78:
79: \section{OBSERVATION AND DATA REDUCTION}
80: The photometric observations of selected L dwarfs were carried out
81: during 2004 January -- June
82: using the 2-m Himalayan Chandra Telescope (HCT) of the Indian Astronomical
83: Observatory (IAO) at Hanle, India, using the Himalaya Faint Object
84: Spectrograph Camera (HFOSC), equipped with a SITe $2\times 4$~K pixel
85: CCD. The central $2\times 2$~K region used for imaging corresponds to a field
86: of view of $10\,{\rm{arcmin}}\times 10\,{\rm{arcmin}}$ at 0.296~arcsec
87: pixel$^{-1}$. More details on the telescope and the instrument
88: may be obtained from {\it{http://www.iiap.res.in/$\sim$iao}}.
89:
90: We selected the targets from published spectroscopically determined
91: L dwarfs, specifically selecting those objects that have negligible or no
92: $H_{\alpha}$ effective line-width. This avoids the contribution of $H_{\alpha}$
93: line variability to the atmospheric variability in the $R$ band and hence any
94: variability observed could be attributed more convincingly, to the presence of
95: dust clouds. Table 1 presents the name, position, and the $R$ and $I$
96: magnitudes of those objects.
97:
98: Several exposures with times of 10~minutes and 5~minutes were
99: obtained in the $R$ and $I$ bands respectively. The central wavelengths of
100: the R band filter and the I band filter used are 0.6 and 0.805 micron
101: respectively. To minimize the effect of
102: improper flat fielding and any systematic error spatially associated with the
103: chip, we tried to confine the L dwarf to a particular CCD pixel in all
104: the frames. The observations were carried out during dark moon period,
105: and uninterrupted observations of a single object were obtained over 3 to 7
106: hours during different nights. The full width at half maximum (FWHM) of the
107: stellar profile was found to be about 1.5 to 1.8 arc-seconds.
108: The observing log is given in the Table \ref{tab2}.
109: The basic image processing such as bias subtraction and flat fielding were
110: done in the standard manner using the various tasks available within IRAF.
111: Atmospheric extinction and transformation coefficients were obtained from
112: observation of photometric standard stars (Landolt 1992). The $I$ frames were
113: affected by CCD fringing due to night sky emission lines. Fringe correction
114: was applied to all the $I$ frames using a master fringe frame created by
115: observing several blank night sky fields.
116:
117: The stellar magnitudes at varying apertures were then obtained using the
118: IRAF task \textit{phot}. Since accurate
119: sky background estimation is very crucial to faint object photometry, the
120: sky value was iteratively estimated by examining the growth curves of
121: isolated stars so that the growth curves were neither monotonically
122: decreasing (underestimated sky) nor increasing (overestimated sky).
123: Furthermore, the magnitudes of L dwarfs and faint stars
124: were first determined at the aperture having highest $S/N$ and then
125: aperture correction was made to the aperture size $4\times FWHM$ using
126: the correction term obtained from bright isolated stars.
127: The standard $R$ and $I$ magnitudes of L dwarfs and the field stars were
128: determined using system transformation coefficients.
129:
130: Differential photometry was performed following the ensemble photometry
131: technique (Gilliland \& Brown 1988; Everett \& Howell 2001). From the average
132: flux of the few fairly bright stars we determined the ensemble reference
133: magnitude, iteratively rejecting stars that were found to have either a
134: systematic variation or large errors.
135: The differential magnitude of L dwarfs with respect to the ensemble magnitude
136: was then computed using the relation
137: \begin{eqnarray}
138: \Delta m_{i,b} & = & \overline m - m_{i,b},
139: \end{eqnarray}
140: where $\overline m$ is the ensemble magnitude and $m_{i,b}$ is the magnitude of
141: L dwarfs.
142:
143: While analyzing the differential photometric data of L dwarfs, a linear
144: variation with respect to airmass was noticed.
145: This trend appears to be an effect of the second order extinction
146: coefficient. The spectral type of the observed ultra-cool dwarfs were L1
147: whereas our ensemble references were found to be near spectral type G.
148: Therefore, within our observing band they would
149: have very different effective wavelengths. In order to check how severe the
150: second order extinction effect would be, we first determined the effective
151: wavelengths of the L dwarfs and G type reference stars using equation:
152: \begin{eqnarray}
153: \lambda_{eff} & = & \frac{ \int \lambda~F_{\lambda}~S_{\lambda}~d\lambda}
154: {\int F_{\lambda}~S_{\lambda}~d\lambda}
155: \end{eqnarray}
156: \noindent
157: where $F_{\lambda}$ is spectral energy distribution and $S_{\lambda}$ is
158: system response. The L dwarf and reference star spectra were
159: retrieved from digital spectral libraries (Martin, Delfosse \& Basri 1999;
160: Le Borgne et al. 2003). The computed effective wavelengths for L
161: dwarf and G type star are 6308~\AA\ and 7148~\AA\ respectively.
162: The average spectroscopic extinction at IAO
163: are $0.081~mag$ and $0.04~mag$ at these two wavelengths.
164: The difference of these two extinction values is nothing but
165: the second order extinction correction at the unit airmass i.e.
166: $k^{''}_{R}\Delta(R-I)$. The observed color difference $(R-I)$ of
167: L dwarf and G type reference star was found to be $2.1~mag$ and
168: that predicts $k^{''}_{R}\approx -0.02~mag$.
169: The second order extinction coefficient was independently estimated using the
170: observations of the photometric standards and found to be
171: $k^{''}_R \approx -0.02$, similar to the above estimation . The second order
172: extinction correction was made to each observed L dwarf data using the relation
173: \begin{eqnarray}
174: (\Delta m_{i,b})_o & = & \Delta m_{i,b} - k^{''}_R \Delta (R-I) X_i
175: \end{eqnarray}
176: \noindent where $X_i$ is airmass of the frames.
177:
178:
179: \section{RESULTS AND DISCUSSIONS}
180:
181: The light curves of the three L1 dwarfs observed are presented in figure 1
182: and in figure 2. Figure~1 shows the light curves of 2MASS 1300+19 and
183: 2MASS 1439+19 in $R$ band, while figure 2 shows the light curves of
184: 2MASS 1658+70 in both $R$ and $I$ bands. The light curves in $R$ band for
185: all the three objects indicate a possible variability.
186:
187: In order to verify any variability, we employ the procedure given by
188: Martin et al.~(2001). In figure~3a-e , we show the standard deviation,
189: $\sigma_R$ with respect to the standard R magnitudes of all the field
190: stars for different fields at different nights along with that for the three
191: targets.
192: Figure 3a-d show that although the programme stars 2MASS 1300+19 and 2MASS
193: 1439+19 lie at the variable side of the
194: $\sigma_R$ vs the R magnitude diagram, no overwhelming evidence for variability
195: in any of the two objects is found. On the other hand Figure 3e shows that
196: the programme object 2MASS 1658+70 lies at the non-variable side of the
197: diagram for 19 May implying no variability in R band. A similar inference
198: can be made from the other observing nights for R band as well as for
199: the I-band observation of the same object.
200:
201: From figure 3a-e, we notice that for the same field there are very less
202: variation of systematic error at different nights. Also, for different fields
203: the relation between $\sigma_R$ with standard R magnitudes and its 1$\sigma$
204: scatter do not change significantly. In order to have
205: objects well distributed across the entire magnitude range we combine the
206: different fields. This should provide a statistically
207: more reliable result for the $\sigma_R$ vs R magnitude relationship.
208: The result with combined fields is presented in figure~3f.
209: For this case, the relation between $\sigma_R$ vs standard R magnitudes can be
210: written as
211: \begin{eqnarray}\label{rel}
212: \sigma_R = 0.712139 - 0.0854801~R + 0.00256972~R^2
213: \end{eqnarray}
214: with a scatter of
215: 0.004 ($1\sigma$). However, the results do not differ from that obtained by
216: using the individual fields. The objects
217: 2MASS 1300+19 and 2MASS 1439+19 are found to be nearly $2\sigma$ and
218: $1\sigma$ away from the mean relation respectively.
219:
220: On the other hand, if we determine the systematic error from the relation
221: between $\sigma_R$ vs standard R magnitude and consider its large $1\sigma$
222: error as well, the systematic error becomes about double the photometric
223: error which is calculated using ensemble references and program star
224: photometric errors. Even if we add the scatter due to color effect, such a
225: large error is not expected. However, if we discard the objects that are
226: fairly out of fit in figure \ref{fig3}(f) as well as having systematic trend
227: in their light curve and consider rest of the field objects instead, the
228: scatter of distribution of error reduces to 0.002 ($1\sigma$). Consequently,
229: the objects 2MASS 1300+19 and 2MASS 1439+19 are found to be situated at about
230: 5$\sigma$ and 3$\sigma$ away from the mean relation respectively. Therefore,
231: our analysis implies variability of 2MASS 1300+19 and 2MASS 1439+19 in R band.
232:
233: The statistical significance of the observed variability is also checked by
234: computing the $\chi^2$ by using the formula:
235: \begin{eqnarray}
236: \chi^2 = \sum_{i = 1}^n \left( \frac{\overline {\Delta m_b } -
237: \Delta m_{i,b}}{\sigma_{i,b}} \right)^2
238: \end{eqnarray}
239: \noindent
240: \noindent where \textit{n} is the number of observed data points and
241: $\sigma_{i,b}$ is the error associated with the L dwarfs at their respective
242: magnitudes, as obtained from the standard deviation versus standard magnitudes
243: relation given by equation (\ref{rel}).
244: However, it is worth mentioning that the ratio between the variance of L
245: dwarf data and the variance read from the fitted curve may not have a
246: $\chi^2$ distribution and hence it should be considered as an assumption.
247: Table~\ref{tab2} gives the
248: results for each object for individual nights of observation, the number of good
249: frames taken for final analysis, number of references taken to make the mean
250: standard, the standard deviation of the points from the mean level
251: ($\sigma_{rms}$), the average $\sigma _{rms}$ used for $\chi^{2}$ test and
252: the probability that the L dwarf is variable ($p$).
253: The L dwarfs 2MASS 1439+19 and 2MASS 1300+19 indicate variability with about
254: 99\% probability for all the observed nights. Note that $\sigma_{i,b}$ are
255: calculated from the $\sigma_R$ vs R magnitude relation by considering all the
256: field stars including those that show systematic trend in their light curve.
257:
258:
259:
260: A period analysis program based on the widely used Scargle formalism
261: (Scargle 1982) was used to search for any periodicity in our time series
262: photometric data. However, we have not obtained any significant periodicity in
263: the variations from either L dwarfs.
264:
265: The third object, 2MASS 1658+70, does not show any
266: variability both in $R$ and $I$ bands. However an inspection of the light
267: curves (see figure~\ref{fig2}) of the 20th of May and the 16th of June
268: although look like scatter plot, there is a systematic trend in the light
269: curves of the 19th and the 21st May 2004 suggesting a possibility of variation.
270: Further, Gelino et al (2002) reported this object to be variable in $I$ band.
271: It is therefore possible that the variability in this object is transient due
272: to the dust-active variation. More observations with improved temporal
273: coverage are required to establish the variable nature of 2MASS 1658+70.
274:
275: \section{CONCLUSIONS}
276: $R$ band differential photometry of three L1 dwarfs 2MASS 1300+19, 2MASS
277: 1439+19, and 2MASS 1658+70 are presented here. The first two objects show
278: variability. However, no periodicity in the
279: variation is obtained in any of these objects. The light curves indicate
280: transient activity of short time scale. Since rotationally related
281: variability should show rather smooth light curves, any co-relation
282: with the rotation of the objects is unlikely. The third object,
283: 2MASS 1658+70, which was also observed in the $I$ band does not show any
284: variability either in the $R$ or in the $I$ band although it was reported to
285: be a variable in the $I$ band by Gelino et al.~(2002). The light
286: curves of this object presented here, however, provide an indication of
287: possible flux variation. More observations of this object are required before
288: its variable nature can be established.
289:
290: The observation of linear polarization in the red by Menard, Delfosse \&
291: Monin (2002) favours the presence of dust cloud in the atmosphere of L dwarfs.
292: The synthetic spectra of L dwarfs also favours the presence of dust in the
293: atmosphere. We propose that the photometric variability in the $R$ band
294: reported here arises due to the dust activity in the atmosphere.
295:
296: On the other hand, if the variability observed is due to the dust activity,
297: then we predict non-zero polarization from L1 dwarfs at R band
298: by dust scattering. However, photospheric variability, if caused by dust
299: cloud, needs sufficiently optically thick dust layer. In an optically thick
300: medium, polarization would arise by multiple scattering of photons. It is
301: known that multiple scattering reduces the degree of polarization as
302: compared to that by single scattering mechanism (Sengupta 2001, 2003). Hence,
303: a variable L dwarf should show less amount of polarization as compared to that
304: of non-variable or weakly variable objects which might have optically thin
305: dust layer.
306:
307: Further observations of L dwarfs of different spectral types
308: and in both $R$ and $I$ bands could tell if there is any co-relation in the
309: variability at $R$ and $I$ bands with the spectral type, which in turn could
310: provide significant insight on the distribution of dust in the atmosphere of
311: L dwarfs.
312:
313: \acknowledgments
314: We are thankful to the referee for many useful suggestions, comments and
315: constructive criticism.
316: Thanks are due to A. V. Raveendran, A. Saha, and T. P. Prabhu for several
317: discussions.
318:
319: The Observations reported in this letter were obtained using the 2-m
320: Himalayan Chandra Telescope at Mt. Saraswati, Hanle, Indian Astronomical
321: Observatory, the high altitude station of the Indian Institute of
322: Astrophysics, Bangalore.
323: We thank the staff at IAO and at the remote control station at CREST,
324: Hosakote for assistance during the observations.
325:
326: \begin{thebibliography}{}
327: \bibitem[Bailer-Jones and Mundt (2001)]{bj01a}Bailer-Jones, C. A. L., and
328: Mundt, R. 2001a, A \& A 367, 218
329: \bibitem[Bailer-Jones and Mundt (2001)]{bj01b}Bailer-Jones, C. A. L., and
330: Mundt, R. 2001b, A \& A 374, 1071
331: \bibitem[]{294} Burrows, A., Hubbard, W. B., Lunine, J. I., \& Liebert, J. 2001,
332: Rev. Mod. Phys. 73, 719
333: \bibitem[]{400} Clarke, F. J., Oppenheimer, B. R., \& Tinney, C. G., 2002a,
334: MNRAS, 335, 1158
335: \bibitem[]{402} Clarke, F. J., Tinney, C. G., \& Covey, K. R. 2002b,
336: MNRAS, 332, 361
337: \bibitem[]{300} Enoch, M. L., Brown, M. E., \& Burgasser, A. J. 2003, AJ, 126,
338: 1006
339: \bibitem[Everett \& Howell (2001)]{how01} Everett, M. E., Howell, S. B.
340: 2001 \pasp, 113, 1428
341: \bibitem[Gelino et al.(2002)]{gal02} Gelino, C. R., Marley, M. S., Holtzman,
342: J. A., Akerman, A. S. and Lodderes, K. 2002, \apj, 557, 433
343: \bibitem[Gilliland \& Brown 1988]{gill88} Gilliland, R. L., Brown,
344: T. M. 1988, \pasp, 100, 754
345: \bibitem[Landolt (1992)]{land92} Landolt, A. U. 1992, AJ, 104, 340
346: \bibitem[Le Borgne et al. (2003)] {LeBor03} Le Borgne, J. F., Bruzual, G.,
347: Pello, R., et al. 2003 \aap , 402 433
348: \bibitem[Martin et al.(2001)]{mart01} Martin, E. L., Osorio, M. R. Z. \&
349: Lehto, H. J. 2001, \apj, 557, 822
350: \bibitem[Martin et al.(1999)]{mart99}Martin, E. L., Delfosse, X., Basri, G.,
351: et al. 1999, AJ, 118, 2466
352: \bibitem[]{415} Menard, F., Delfosse, X., \& Monin, J. 2002, A \& A, 396, L35
353: \bibitem[]{416} Osorio, M. R. Z, Caballero, J. A., Bejar, V. J. S., \& Rebolo,
354: R. 2003, A \& A, 408, 663
355: \bibitem[]{418} Scargle, J. D. 1982, ApJ, 263, 835.
356: \bibitem[]{419} Sengupta, S. \& Kwok, S. 2004, ApJ, (submitted)
357: \bibitem[]{420} Sengupta, S. 2003, \apj, 585, L155
358: \bibitem[]{421} Sengupta, S. \& Krishan, V. 2001, ApJ, 561, L123
359: \bibitem[]{322} Tsuji, T., Nakajima, T., \& Yanagisawa, K. 2004, ApJ, 607, 511
360: \end{thebibliography}
361:
362:
363: \begin{figure}[hb]
364: \includegraphics[scale = .9, trim = 0 150 00 00,clip]{f1.eps}
365: \caption{The $R$ band light curves of L1 dwarfs 2MASS1300+19 (top) and
366: 2MASS1439+19 (bottom) obtained on different nights.\label{fig1}}
367: \end{figure}
368:
369: \clearpage
370:
371: \begin{figure}[hb]
372: \includegraphics[scale = .9, trim = 0 150 00 00,clip]{f2.eps}
373: \caption{The $R$ and $I$ band light curves of the L1 dwarf 2MASS1658+70
374: obtained on different nights. \label{fig2}}
375: \end{figure}
376:
377: \clearpage
378: \begin{figure}
379: \includegraphics[scale = .9, trim = 0 0 00 00,clip]{f3.eps}
380: \caption{Standard deviation ($\sigma_R$) versus $R$ band magnitudes for
381: the field stars for three different fields for individual nights.
382: A second-order polynomial fit to the $\sigma_R$
383: is shown as a continuous curve. The $\sigma_{rms}$ on each night
384: for all the three L1 dwarfs are also shown as filled symbols. The result with
385: combined data of May21, April 17 and May 19 is given in plot f.
386: \label{fig3}}
387: \end{figure}
388: \clearpage
389:
390: \clearpage
391:
392: \begin{table}
393: \begin{center}
394: \caption{BASIC DATA OF VARIABLE L1 DWARFS. \label{tab1}}
395: \vskip5mm
396: \begin{tabular}{ccccccc}
397: \tableline\tableline
398: Name & RA(2000) & DEC(2000) & $R~mag$ & $I~mag$ & $H_{\alpha}$ Emission \\
399: \tableline
400: 2MASS 1300+19 & 13 00 42.5 & +19 12 35 & 18.7 & 16.0 & ... \\
401: 2MASS 1439+19 & 14 39 28.4 & +19 29 15 & 18.7 & 16.0 & $< 0.03$ \\
402: 2MASS 1658+70 & 16 58 03.7 & +70 27 01 & 19.2 & 16.6 & ... \\
403: \tableline
404: \end{tabular}
405: \end{center}
406: \end{table}
407: \clearpage
408: \begin{table}[h]
409: \begin{center}
410: \caption{Observing log and results \label{tab2} }
411: \begin{tabular}{cccccccc}
412: \tableline\tableline
413: Name & Dates & Frames & Ref. & Filter
414: & $\sigma_{rms}$\tablenotemark{a} of & $\sigma_{rms}$\tablenotemark{b} used &
415: p(\%) \\
416: & & (Used/Total) & stars & & L dwarfs & in $\chi^2$ & \\
417: \tableline
418: 2MASS1300+19 & 19/02/04 & 13/15 & 5 & R & 0.019 & 0.012 & $>$99.0\\
419: 2MASS1300+19 & 21/05/04 & 21/24 & 5 & R & 0.020 & 0.012 & $>$99.0\\
420: \hline
421: 2MASS1439+19 & 17/04/04 & 20/24 & 5 & R & 0.017 & 0.012 & $>$99.0\\
422: 2MASS1439+19 & 18/04/04 & 21/24 & 5 & R & 0.016 & 0.012 & $>$98.5\\
423: \hline
424: 2MASS1658+70 & 19/05/04 & 24/25 & 5 & R & 0.014 & 0.018 & 10.1\\
425: 2MASS1658+70 & 20/05/04 & 15/15 & 5 & R & 0.012 & 0.018 & 4.5\\
426: 2MASS1658+70 & 21/05/04 & 9/9 & 5 & R & 0.011 & 0.018 & 6.1\\
427: 2MASS1658+70 & 16/06/04 & 24/24 & 3 & I & 0.005 & 0.007 & 3.1\\
428: \tableline
429: \end{tabular}
430: \tablenotetext{a}{$\sigma_{rms}$ about the mean differential magnitude of the
431: L dwarf.}
432: \tablenotetext{b}{$\sigma_{rms}$ used in $\chi^2$ test, obtained from standard
433: deviation ($\sigma_R$) vs. standard R magnitudes relation.}
434: \end{center}
435: \end{table}
436:
437: \end{document}
438:
439:
440:
441: