Morphology of the near-terminator Martian ionosphere: A comparison of models and data

[ 1] We have constructed low and high solar activity models of the Martian thermosphere/ ionosphere for solar zenith angles from 60 to 90 degrees in 5 degree increments. The solar fluxes that we have adopted are those from the Solar 2000 v2.22 models of Tobiska ( 2004), without enhancements of the soft X-ray fluxes. The background neutral density and temperature profiles are similar to those that we have recently presented ( Fox, 2004). We compute the density profiles for 14 ions and nine neutral species. For all the models, we present altitude profiles of the photoionization rates, electron impact ionization rates, total ion production rates, and the predicted electron density profiles. Each model exhibits both an F-1 peak and an E peak, although the latter usually appears as a shoulder, rather than as a separate peak. The altitudes of the model peaks are found to be slightly too high. We fit the model peak densities to the equation n(max,chi)(i) = A( cos chi)(k), where, for an ideal Chapman layer, A is the value of the subsolar peak density, n(max,0)(i), and the exponent k is 0.5. We compare the behavior of the model electron density profiles to that of a theoretical Chapman layer and to the values of A and k obtained by fitting the Mars Global Surveyor (MGS) radio science electron density profiles for occultation seasons 1, 2, and 4. We also compare our results to those of previous investigators who have analyzed data from earlier Mars missions and those from MGS and from the Mars Express spacecraft. We find that our model best fit values of k for the F-1 peak and those derived from the MGS data are less than the Chapman value of 0.5. We note, however, that the use of spherical geometry alone reduces the value of k below the Chapman value for large solar zenith angles, but the deviation from the experimental values also indicates that there are changes in the neutral atmosphere as the terminator is approached. Our peak densities and predicted subsolar peak densities for both the F-1 and E peaks are somewhat smaller than those derived from the data. This is attributed to the use of the S2K v2.22 solar flux models, rather than the S2K v1.24 models or those from Hinteregger ( 1981). We also evaluate the neutral, ion, and plasma pressure scale heights at the peaks, 33 km above the peaks and at 250 km for all the models. We find that the solar activity variation of our peak densities are in substantial agreement with those determined by other investigators. We argue that the peaks near 90 degrees solar zenith angle are above the photochemical equilibrium region and fitting these peaks to a Chapman profile is therefore inappropriate.