The planetary effective surface temperature alone is insufficient to characterize exoplanet atmospheres and their stability or evolution. Considering the star-planet system as a whole is necessary, and a critical component of the system is the photoionizing stellar extreme ultraviolet emission (EUV; 100-912 {\AA}). EUV photons drive atmospheric mass loss through thermal and nonthermal processes, and an accurate accounting of the EUV energy deposition in a planet's energy budget is essential, especially for terrestrial habitable zone planets and close-in gaseous planets. Direct EUV observations of exoplanet host stars would require a new, dedicated observatory. Archival observations from the $\textit{EUVE}$ satellite, models, and theory alone are insufficient to accurately characterize EUV spectra of the majority of exoplanet host stars, especially for low-mass stars.

The EUV (100-912 A) is a spectral region notoriously difficult to observe due to attenuation by neutral hydrogen gas in the interstellar medium. Despite this, hundreds to thousands of nearby stars of different spectral types and magnetic activity levels are accessible in the EUV range. The EUV probes interesting and complicated regions in the stellar atmosphere like the lower corona and transition region that are inaccessible from other spectral regions. In this white paper we describe how direct EUV observations, which require a dedicated grazing-incidence observatory, cannot yet be accurately substituted with models and theory. Exploring EUV emission from cool dwarf stars in the time domain can make a major contribution to understanding stellar outer atmospheres and magnetism, and offers the clearest path toward detecting coronal mass ejections on stars other than the Sun.

We have studied how young Mars lost its atmospheric and exospheric oxygen and carbon into the solar wind when ionized atmospheric oxygen and carbon atoms are picked up by the solar wind convection electric field. Analysis was made by a 3D kinetic hybrid model which treats ions as particles and electrons as a massless charge neutralizing fluid. The hybrid simulation allows to study self-consistently the motion of ions in the Martian magnetosphere and to estimate the total ion escape rate.

We study the Type III solar bursts simultaneously recorded by radio experiments onboard Cassini, Ulysses and Wind. Those radio bursts cover a large frequency range from about 14 MHz to a few kHz. The corresponding source locations are mainly in the solar corona and the interplanetary medium. The empirical electron density models provide different distances depending on the emission mode, fundamental or harmonic. A real trouble arises due to the distance discrepancies, as inferred from the models. Also the Archimedean spiral trajectories of the electrons, at the origin of the Type III bursts, are another difficulty to correctly estimate the source locations. We show in our analysis that the stereoscopic observations are essential to reduce the source location inaccuracy. We finally discuss the relationship between the Type III beams, the emission modes and the source locations.