Dust in the photospheric environment. III. A fundamental element in the characterization of ultracool dwarfs

Recent photometry of L and T dwarfs revealed that the infrared colors show a large variation at a given effective temperature, and within the framework of our unified cloudy model (UCM), this result can be interpreted as due to a sporadic variation of the critical temperature (T-cr), which is essentially a measure of the thickness of the dust cloud. In our previous applications of the UCMs, we assumed that T-cr is constant at about 1800 K in all the L and T dwarfs, but in view of the new observing result, we now allow T-cr to vary between the surface temperature (T-0) and the condensation temperature (T-cond) at given T-eff and log g. Then, the two-color diagram and the color-magnitude diagram can be well explained by the effects of T-eff, log g, and T-cr, but not by the effects of T-eff and log g alone. This result implies that T-cr is one of the important parameters needed for characterization of dusty dwarfs. The effects of T-eff and T-cr on individual spectra, however, are difficult to discriminate, since T-eff at fixed T-cr on one hand and T-cr at fixed T-eff on the other essentially have the same effect on the spectra. We show that the degeneracy of T-eff and T-cr can be removed to some extent by the analysis of the spectral energy distribution on an absolute scale. The reanalysis of a selected sample of spectra revealed that the L-T spectral sequence may not necessarily be a sequence of T-eff, but may reflect a change in the thickness of the dust cloud, represented by T-cr in our UCM. Although this unexpected result is based on a limited sample, an odd "brightening" of the absolute J magnitudes plotted against the L-T spectral types may also be an indication that the L-T spectral sequence is not necessarily a temperature sequence. This is because M-bol based on the same photometry data also shows a similar brightening. Thus, the "J-brightening" might not be due to any atmospheric effect and hence should not be a problem to be solved by model atmospheres, including the UCMs. Thus, almost all the available observed data are reasonably well interpreted by the UCMs, whose full capability emerges once T-cr is introduced as the fifth parameter in addition to the usual four parameters (i.e., chemical composition, T-eff, log g, and microturbulent velocity) needed to characterize stellar spectra in general.