Infrared imaging of late-type stars

Infrared imaging properties of dusty winds around late-type stars are investigated in detail, employing a self-consistent model that couples the equations of motion and radiative transfer. Because of general scaling properties, the angular profiles of surface brightness are self-similar. In any given star, the profile shape is determined essentially by overall optical depth at each wavelength and it is self-similarly scaled by the size of the dust condensation zone. We find that the mid-IR is the best wavelength range to measure directly the angular size of this zone, and from IRAS data we identify the 15 best candidates for such future observations. We also show that the visibility function at short wavelengths (less than or similar to 2 mu m) directly determines the scattering optical depth, and produce theoretical visibility curves for various characteristic wavelengths and the entire parameter range relevant to late-type stars. The infrared emission should display time variability because of cyclical changes in overall optical depth, reflecting luminosity induced movement of the dust condensation point. Calculations of the wavelength dependence of photometric amplitudes and time variability of envelope sizes are in agreement with observations; envelopes are bigger and bluer at maximum light.