Search for a command to run...
A review and critique of present‐day kinetic theory models of planetary exospheres is presented. Models of ionized exospheres, specifically the solar and the terrestrial polar wind, are also discussed. The objective of the paper is to point out the need for a rigorous kinetic theory treatment of the atmosphere in the altitude region between the thermosphere and the exosphere. This is the region where the atmosphere undergoes a transition from a collision‐dominated to a collisionless situation. The various aspects of the exospheric problem are introduced and developed around this main theme. The exospheric problem and its considerable overlap with several related problems in physics and chemistry are noted. The calculation of the velocity distribution function of exospheric constituents with the spherically symmetric collisionless model is then presented. Various modifications of this standard model with regard to planetary rotation, nonuniform exobasic density and temperature distributions, and time‐dependent effects are subsequently discussed and compared. The extent to which collisionless models suffice to explain observational data of planetary exospheres is assessed. This assessment is given in terms of a comparison of density profiles or reduction factors determined with hydrogen and/or helium cells. A comparison of experimental measurements of the diurnal variation of terrestrial atomic hydrogen densities and theoretical models to explain the asymmetry is presented. The collisionless approach is criticized on theoretical grounds for the neglect of a consideration of the transition region and the resulting artificial boundary between the thermosphere and the exosphere. The aim of collisional models with regard to escape‐induced non‐Maxwellian effects, modifications of the temperature structure, nonthermal escape processes, and satellite particle populations from Monte Carlo simulations and/or kinetic theory models based on a Boltzmann equation is discussed, and the various results are compared. The various attempts to introduce collisional concepts are shown to be incomplete, each considering the problem with different objectives. Throughout, the importance of a more rigorous collisional treatment is stressed, in particular with respect to the description of nonthermal escape processes and satellite particle populations which cannot be considered in a collisionless context. The effects of charge exchange reactions, photoionization, and solar radiation pressure on satellite particles are examined. With regard to nonthermal escape processes, the absence of a detailed kinetic theory of product velocity distributions based on a collisional theory is also noted.