Unveiling the dust nucleation zone of IRC+10216 with ALMA

We report the detection in IRC+10216 of lines of HNC $J$=3-2 pertaining to 9 excited vibrational states with energies up to $sim$5300 K. The spectrum, observed with ALMA, also shows a surprising large number of narrow, unidentified lines that arise in the vicinity of the star. The HNC data are interpreted through a 1D--spherical non--local radiative transfer model, coupled to a chemical model that includes chemistry at thermochemical equilibrium for the innermost regions and reaction kinetics for the external envelope. Although unresolved by the present early ALMA data, the radius inferred for the emitting region is $sim$0.06 (i.e., $simeq$ 3 stellar radii), similar to the size of the dusty clumps reported by IR studies of the innermost region ($r <$ 0.3). The derived abundance of HNC relative to H$_2$ is $10^{-8} <$ $chi$(HNC) $< 10^{-6}$, and drops quickly where the gas density decreases and the gas chemistry is dominated by reaction kinetics. Merging HNC data with that of molecular species present throughout the inner envelope, such as vibrationally excited HCN, SiS, CS, or SiO, should allow us to characterize the physical and chemical conditions in the dust formation zone.

Classification of Stellar Spectra with LLE

We investigate the use of dimensionality reduction techniques for the classification of stellar spectra selected from the SDSS. Using local linear embedding (LLE), a technique that preserves the local (and possibly non-linear) structure within high dimensional data sets, we show that the majority of stellar spectra can be represented as a one dimensional sequence within a three dimensional space. The position along this sequence is highly correlated with spectral temperature. Deviations from this stellar locus are indicative of spectra with strong emission lines (including misclassified galaxies) or broad absorption lines (e.g. Carbon stars). Based on this analysis, we propose a hierarchical classification scheme using LLE that progressively identifies and classifies stellar spectra in a manner that requires no feature extraction and that can reproduce the classic MK classifications to an accuracy of one type.