Spatial variations of the [S II]/H alpha and [N II]/H alpha line intensity
ratios observed in the gaseous halo of the Milky Way and other galaxies are
inconsistent with pure photoionization models. They appear to require a su
pplemental heating mechanism that increases the electron temperature at low
densities, n(e). This would imply that in addition to photoionization, whi
ch has a heating rate per unit volume proportional to n(e)(2), there is ano
ther source of heat with a rate per unit volume proportional to a lower pow
er of n(e). One possible mechanism is the dissipation of interstellar plasm
a turbulence, which, according to Minter & Spangler, heats the ionized inte
rstellar medium in the Milky Way at a rate of similar to 1 x 10(-25)n(e) er
gs cm(-3) s(-1). If such a source were present, it would dominate over phot
oionization heating in regions where n(e) less than or similar to 0.1 cm(-3
), producing the observed increases in the [S II]/H alpha and [N II]/H alph
a intensity ratios at large distances from the galactic midplane as well as
accounting for the constancy of [S II]/[N II], which is not explained by p
ure photoionization. Other supplemental heating sources, such as magnetic r
econnection, cosmic rays, or photoelectric emission from small grains, coul
d also account for these observations, provided they supply similar to 10(-
5) ergs s(-1) per square centimeter of the Galactic disk to the warm ionize
d medium.