Mark S. Marley

We present a low-resolution (R = 90), 5.5-38 micron spectral sequence of a sample of M, L, and T dwarfs obtained with the Infrared Spectrograph (IRS) onboard the Spitzer Space Telescope. The spectra exhibit prominent absorption bands of H_2O at 6.27 microns, CH_4 at 7.65 microns, and NH_3 at 10.5 microns and are relatively featureless at lambda > 15 microns. Three spectral indices that measure the strengths of these bands are presented; H_2O absorption features are present throughout the MLT sequence while the CH_4 and NH_3 bands first appear at roughly the L/T transition. Although the spectra are, in general, qualitatively well matched by synthetic spectra that include the formation of spatially homogeneous silicate and iron condensate clouds, the spectra of the mid-type L dwarfs show an unexpected flattening from roughly 9 to 11 microns. We hypothesize that this may be a result of a population of small silicate grains that are not predicted in the cloud models. The spectrum of the peculiar T6 dwarf 2MASS J0937+2931 is suppressed from 5.5-7.5 microns relative to typical T6 dwarfs and may be a consequence of its mildly metal-poor/high surface gravity atmosphere. Finally, we compute bolometric luminosities of a subsample of the M, L, and T dwarfs by combining the IRS spectra with previously published 0.6-4.1 micron spectra and find good agreement with the values of Golimowski et al. who use L'- and M'-band photometry and to account for the flux emitted at lambda > 2.5 microns.

Clouds and hazes are important throughout our solar system and in the atmospheres of brown dwarfs and extrasolar giant planets. Among the brown dwarfs, clouds control the colors and spectra of the L-dwarfs; the disappearance of clouds helps herald the arrival of the T-dwarfs. The structure and composition of clouds will be among the first remote-sensing results from the direct detection of extrasolar giant planets.

We report on the results of an I-band photometric variability survey of eighteen L dwarfs. We find that seven exhibit statistically significant variations above the 95.4% confidence level with root-mean-square scatter (including photometric errors) between 0.010 and 0.083 mag. Another five targets have variability probabilities ~80%, suggesting that these are likely variable objects. Three of the variable objects display significant peaks in a CLEAN periodogram that are several times higher than the noise. The period for 2MASS 0345+25 is clearly not intrinsic to the object and can be dismissed. The periods found for 2MASS 0746+20AB and 2MASS 1300+19 are unique but longer than those periods likely from rotation velocity measurements and they do not represent periodic behavior in the light curve that persists through the entire data set. These observations suggest that we are not observing the rotation modulation of a long-lived albedo feature. Instead, rapid evolution of atmospheric features is likely causing the non-periodic variability. The remaining variable objects show no prominent features in their light curves, suggesting even more rapid evolution of atmospheric features. We argue against the existence of magnetic spots in these atmospheres and favor the idea that non-uniform condensate coverage is responsible for these variations. The magnetic Reynolds number in the atmosphere of L dwarfs is too small to support the formation of magnetic spots. In contrast, silicate and iron clouds are expected to form in the photospheres of L dwarfs. Inhomogeneities in such cloud decks and the evolution of the inhomogeneities can plausibly produce the observed photometric variations.

Clouds of metal-bearing condensates play a critical role in shaping the emergent spectral energy distributions of the coolest classes of low-mass stars and brown dwarfs, L and T dwarfs. Because condensate clouds in planetary atmospheres show distinct horizontal structure, we have explored a model for partly cloudy atmospheres in brown dwarfs. Our model successfully reproduces the colors and magnitudes of both L and T dwarfs for the first time, including the unexpected brightning of the early- and mid-type T dwarfs at J-band, provided that clouds are rapidly removed from the photosphere at T_eff ~ 1200 K. The clearing of cloud layers also explains the surprising persistence and strengthening of gaseous FeH bands in early- and mid-type T dwarfs. The breakup of cloud layers is likely driven by convection in the troposphere, analogous to phenomena observed on Jupiter. Our results demonstrate that planetary-like atmospheric dynamics must be considered when examining the evolution of free-floating brown dwarfs.

We present a method to calculate vertical profiles of particle size distributions in condensation clouds of giant planets and brown dwarfs. The method assumes a balance between turbulent diffusion and sedimentation in horizontally uniform cloud decks. Calculations for the Jovian ammonia cloud are compared with results from previous methods. An adjustable parameter describing the efficiency of sedimentation allows the new model to span the range of predictions made by previous models. Calculations for the Jovian ammonia cloud are consistent with observations. Example calculations are provided for water, silicate, and iron clouds on brown dwarfs and on a cool extrasolar giant planet. We find that precipitating cloud decks naturally account for the characteristic trends seen in the spectra and colors of L- and T-type ultracool dwarfs.Comment: 33 pages including 7 figures; AASTex; Accepted for publication in Astrophysical Journal, tentatively scheduled for v556 n1 July 20, 2001 . Plotting error in Fig 5 corrected; slight modification to Fig 5 dicussion in tex

The recent discovery of photometric variations in L dwarfs has opened a discussion on the cause of the variations. We argue against the existence of magnetic spots in these atmospheres and favor the idea that non-uniform condensate coverage (i.e. clouds) is responsible for the variations. The magnetic Reynolds number (Rm) in the atmosphere of L dwarfs, which describes how well the gas couples with the magnetic field, is too small (<<1) to support the formation of magnetic spots. In constrast silicate and iron clouds form in the photospheres of L dwarfs. Inhomogeneities in such cloud decks can plausibly produce the observed photometric variations. Further evidence in support of clouds is the tendency for variable L dwarfs to be bluer in J-Ks than the average L dwarf of a given spectral type. This color effect is expected if clear holes appear in an otherwise uniform cloud layer.

We present a method to calculate vertical profiles of particle size distributions in condensation clouds of giant planets and brown dwarfs. The method assumes a balance between turbulent diffusion and sedimentation in horizontally uniform cloud decks. Calculations for the Jovian ammonia cloud are compared with results from previous methods. An adjustable parameter describing the efficiency of sedimentation allows the new model to span the range of predictions made by previous models. Calculations for the Jovian ammonia cloud are consistent with observations. Example calculations are provided for water, silicate, and iron clouds on brown dwarfs and on a cool extrasolar giant planet. We find that precipitating cloud decks naturally account for the characteristic trends seen in the spectra and colors of L- and T-type ultracool dwarfs.