R. C. Woodward (U Wisc), W. H. Smyth (AER)
The complex structure of the Io plasma torus makes the task of modeling it difficult and computationally expensive; several equilibrium assumptions and/or simplifications are therefore commonly used to make the problem more tractable. Because many of these assumptions were untested, we made them optional in our semi-empirical model of the torus [B.A.A.S. 26, 1139 (1994)]. Here we explore the effects of three of these assumptions on the modeled latitudinal emission brightnesses of the torus. First: Perhaps the least-often questioned assumption is that the distribution of atomic states of an ion species is in local equilibrium with the surrounding electrons, based on the fact that the inverse Einstein A coefficients of most observed lines are short compared to the ion's travel time along a field line of ~ 5000 s from one extreme of electron density to another. However, the commonly observed [S II] 6716,6731Å doublet has much smaller Einstein A coefficients [1/(5780 s) and 1/(1890 s), respectively], suggesting that their upper states may not fully equilibrate. We have therefore performed a full time-evolved nonequilibrium calculation of emissions from the S+ ion. While the differences from the corresponding equilibrium calculations are not huge, neither are they negligible: we show that the nonequilibrium 6716,6731Å calculated emissions are latitudinally more extended and, consequently, that parallel ion temperatures estimated from equilibrium models are significantly too high. Second: Although there is strong evidence of a second electron population in the torus, hotter but less dense than the first, this population is frequently neglected in emission brightness calculations for computational ease. Third: The various plasma species, while not necessarily in thermal equilibrium with one another, are generally each regarded as being in a purely thermal (i.e. Maxwellian) velocity distribution. Recent data from Ulysses, however, do not support this--the quasi-thermal "kappa" distribution appears to be a better description of the data--but thermal distributions are still commonly used. We present comparisons of modeled emission brightnesses calculated with and without each of these simplifications.