[1] In this paper we describe measurements of the Martian dayglow obtained by SPICAM UV spectrograph on board Mars Express between October 2004 and March 2005. Typical spectra (of more than 24,000 individual spectra) display the main features of the dayglow already reported more than 30 years ago (Barth et al., 1971. The variations with altitude, solar zenith angle, and aerocentric longitude of the main emissions, namely, the CO Cameron band emissions (a 3 Å -X 1 S + ) from 180-260 nm, the CO 2 + ultraviolet doublet (B 2 S + -X 2 Å) emission near 289 nm, the O resonance line at 130.4 nm, and the H Lyman alpha line at 121.6 nm, are detailed. The average temperature deduced from the CO 2 + (B 2 S + -X 2 Å) emission scale height between 150 and 190 km is T 1 = 201 ± 10 K. The altitude of the peak of the dayglow emission varies from 120 km at low solar zenith angle ($30°) up to 132 km at high solar zenith angle ($75°). We also find 24 orbits in which the individual derived exospheric temperatures do not show significant variations with respect to solar zenith angle, aerocentric longitude, or longitude. The presence of crustal magnetic fields appears to correlate with a significant ($100 K) increase in exospheric temperature. We report also the first observations of emissions associated with the N 2 Vegard Kaplan band system in the Martian upper atmosphere.

In the high-latitude regions of Earth, aurorae are the often-spectacular visual manifestation of the interaction between electrically charged particles (electrons, protons or ions) with the neutral upper atmosphere, as they precipitate along magnetic field lines. More generally, auroral emissions in planetary atmospheres "are those that result from the impact of particles other than photoelectrons" (ref. 1). Auroral activity has been found on all four giant planets possessing a magnetic field (Jupiter, Saturn, Uranus and Neptune), as well as on Venus, which has no magnetic field. On the nightside of Venus, atomic O emissions at 130.4 nm and 135.6 nm appear in bright patches of varying sizes and intensities, which are believed to be produced by electrons with energy <300 eV (ref. 7). Here we report the discovery of an aurora in the martian atmosphere, using the ultraviolet spectrometer SPICAM on board Mars Express. It corresponds to a distinct type of aurora not seen before in the Solar System: it is unlike aurorae at Earth and the giant planets, which lie at the foot of the intrinsic magnetic field lines near the magnetic poles, and unlike venusian auroras, which are diffuse, sometimes spreading over the entire disk. Instead, the martian aurora is a highly concentrated and localized emission controlled by magnetic field anomalies in the martian crust.

Abstract. Auroral electron transport calculations are a critical part of auroral models. We evaluate a numerical solution to the transport and energy degradation problem. The numerical solution is verified by reproducing simplified problems to which analytic solutions exist, internal self-consistency tests, comparison with laboratory experiments of electron beams penetrating a collision chamber, and by comparison with auroral observations, particularly the emission ratio of the N2 second positive to N+2 first negative emissions. Our numerical solutions agree with range measurements in collision chambers. The calculated N22P to N+21N emission ratio is independent of the spectral characteristics of the incident electrons, and agrees with the value observed in aurora. Using different sets of energy loss cross sections and different functions to describe the energy distribution of secondary electrons that emerge from ionization collisions, we discuss the uncertainties of the solutions to the electron transport equation resulting from the uncertainties of these input parameters.

[1] We present a new set of observations of Martian aurorae obtained by Spectroscopy for the Investigation of the Characteristics of the Atmosphere of Mars (SPICAM) on board Mars Express (MEX). Using nadir viewing, several auroral events have been identified on the Martian nightside, all near regions of crustal magnetic fields. For most of these events, two to three consecutive events with variable intensities and separated by a few seconds to several tens of seconds have been observed, whereas simultaneous observations with Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS) and Analyzer of Space Plasma and Energetic Atoms (ASPERA-3) have been possible. In this paper, we present the data set for these events and discuss the possible correlation between the measured UV emission by SPICAM, the measured downward and/or upward flux of electrons by ASPERA-3 and the total electron content recorded by MARSIS. Despite the limited coverage of SPICAM ultraviolet spectrograph (UVS) on the Martian nightside (essentially in regions of high crustal magnetic fields), there is however a very good correlation between the regions with the locally smallest probability to be on closed crustal magnetic field lines, as derived from Mars Global Surveyor/ Electron Reflectometer (MGS/MAG-ER), and the position of an aurora event. This suggests that the crustal magnetic fields, when organized into cusp-like structure, can trigger the few aurorae identified by SPICAM UVS. It confirms also the good probability, in the cases where SPICAM UVS measured UV emissions, that the increase in the measured total electron content by MARSIS and the simultaneous measured precipitating electron flux by the ASPERA-3/Electron Spectrometer may be related to each other.

Abstract. The suprathermal particles, electrons and protons, coming from the magnetosphere and precipitating into the high-latitude atmosphere are an energy source of the Earth's ionosphere. They interact with ambient thermal gas through inelastic and elastic collisions. The physical quantities perturbed by these precipitations, such as the heating rate, the electron production rate, or the emission intensities, can be provided in solving the kinetic stationary Boltzmann equation. This equation yields particle fluxes as a function of altitude, energy, and pitch angle. While this equation has been solved through different ways for the electron transport and fully tested, the proton transport is more complicated. Because of charge-changing reactions, the latter is a set of two-coupled transport equations that must be solved: one for protons and the other for H atoms. We present here a new approach that solves the multistream proton/hydrogen transport equations encompassing the collision angular redistributions and the magnetic mirroring effect. In order to validate our model we discuss the energy conservation and we compare with another model under the same inputs and with rocket observations. The influence of the angular redistributions is discussed in a forthcoming paper.