The binary brown dwarf WISE J104915.57$-$531906.1 (also Luhman 16AB), composed of a late L and early T dwarf, is a prototypical L/T transition flux reversal binary located at only 2 pc distance. Luhman 16B is a known variable whose light curves evolv
e rapidly. We present spatially resolved spectroscopic time-series of Luhman 16A and B covering 6.5 h using HST/WFC3 at 1.1 to 1.66 $mu$m. The small, count-dependent variability of Luhman 16A at the beginning of the observations likely stems from instrumental systematics; Luhman 16A appears non-variable above $approx$0.4%. Its spectrum is well fit by a single cloud layer with intermediate cloud thickness (f_sed=2, Teff=1200 K). Luhman 16B varies at all wavelengths with peak-to-valley amplitudes of 7-11%. The amplitude and light curve shape changes over only one rotation period. The lowest relative amplitude is found in the deep water absorption band at 1.4 $mu$m, otherwise it mostly decreases gradually from the blue to the red edge of the spectrum. This is very similar to the other two known highly variable early T dwarfs. A two-component cloud model accounts for most of the variability, although small deviations are seen in the water absorption band. We fit the mean spectrum and relative amplitudes with a linear combination of two models of a warm, thinner cloud (Teff=1300 K, fsed=3) and a cooler, thicker cloud (Teff=1000-1100 K, f_sed=1), assuming out-of-equilibrium atmospheric chemistry. A cloud as for Luhman 16A but with holes cannot reproduce the variability of Luhman 16B, indicating more complex cloud evolution through the L/T transition. The projected separation of the binary has decreased by $approx$0.3 in 8 months.
Most directly imaged giant exoplanets are fainter than brown dwarfs with similar spectra. To explain their relative underluminosity unusually cloudy atmospheres have been proposed. However, with multiple parameters varying between any two objects, it
remained difficult to observationally test this idea. We present a new method, sensitive time-resolved Hubble Space Telescope near-infrared spectroscopy, to study two rotating L/T transition brown dwarfs (2M2139 and SIMP0136). The observations provide spatially and spectrally resolved mapping of the cloud decks of the brown dwarfs. The data allow the study of cloud structure variations while other parameters are unchanged. We find that both brown dwarfs display variations of identical nature: J- and H-band brightness variations with minimal color and spectral changes. Our light curve models show that even the simplest surface brightness distributions require at least three elliptical spots. We show that for each source the spectral changes can be reproduced with a linear combination of only two different spectra, i.e. the entire surface is covered by two distinct types of regions. Modeling the color changes and spectral variations together reveal patchy cloud covers consisting of a spatially heterogenous mix of low-brightness, low-temperature thick clouds and brighter, thin and warm clouds. We show that the same thick cloud patches seen in our varying brown dwarf targets, if extended to the entire photosphere, predict near-infrared colors/magnitudes matching the range occupied by the directly imaged exoplanets that are cooler and less luminous than brown dwarfs with similar spectral types. This supports the models in which thick clouds are responsible for the near infrared properties of these underluminous exoplanets.
We present H-band VLT/NACO polarized light images of the Herbig Ae/Be star HD169142 probing its protoplanetary disk as close as ~0.1 to the star. Our images trace the face-on disk out to ~1.7 (~250 AU) and reveal distinct sub-structures for the first
time: 1) the inner disk (<20 AU) appears to be depleted in scattering dust grains; 2) an unresolved disk rim is imaged at ~25 AU; 3) an annular gap extends from ~40 - 70 AU; 4) local brightness asymmetries are found on opposite sides of the annular gap. We discuss different explanations for the observed morphology among which ongoing planet formation is a tempting - but yet to be proven - one. Outside of ~85 AU the surface brightness drops off roughly r^{-3.3}, but describing the disk regions between 85-120 AU / 120-250 AU separately with power-laws r^{-2.6} / r^{-3.9} provides a better fit hinting towards another discontinuity in the disk surface. The flux ratio between the disk integrated polarized light and the central star is ~4.1 * 10^{-3}. Finally, combining our results with those from the literature, ~40% of the scattered light in the H-band appears to be polarized. Our results emphasize that HD169142 is an interesting system for future planet formation or disk evolution studies.
