Mercury's Atmosphere: A Surface-Bounded ExosphereAbstractIntroductionComposition: Sources, Sinks, and ProcessesStructure and Density DistributionsExosphere-Surface Boundary InteractionsSolar Wind and Magnetospheric InteractionsWhat's Next?SummaryAcknowledgementsReferencesSpace Sci Rev (2007) 131: 161–186DOI 10.1007/s11214-007-9260-9Mercury’s Atmosphere: A Surface-Bounded ExosphereDeborah L. Domingue ·Patrick L. Koehn ·Rosemary M. Killen ·Ann L. Sprague ·Menelaos Sarantos ·Andrew F. Cheng ·Eric T. Bradley ·William E. McClintockReceived: 28 August 2006 / Accepted: 7 August 2007 / Published online: 24 October 2007© Springer Science+Business Media B.V. 2007Abstract The existence of a surface-bounded exosphere about Mercury was discoveredthrough the Mariner 10 airglow and occultation experiments. Most of what is currentlyknown or understood about this very tenuous atmosphere, however, comes from ground-based telescopic observations. It is likely that only a subset of the exospheric constituentshave been identified, but their variable abundance with location, time, and space weatherevents demonstrate that Mercury’s exosphere is part of a complex system involving theplanet’s surface, magnetosphere, and the surrounding space environment (the solar wind andinterplanetary magnetic field). This paper reviews the current hypotheses and supporting ob-servations concerning the processes that form and support the exosphere. The outstandingquestions and issues regarding Mercury’s exosphere stem from our current lack of knowl-edge concerning the surface composition, the magnetic field behavior within the local spaceenvironment, and the character of the local space environment.Keywords Atmospheres · Exosphere · Mercury · Space physics · Space weathering ·MESSENGERD.L. Domingue () · A.F. ChengThe Johns Hopkins University Applied Physics Laboratory, Laurel, MD 20723, USAe-mail: [email protected]. L. Ko eh nDepartment of Physics and Astronomy, Eastern Michigan University, Ypsilanti, MI 48197, USAR.M. Killen · M. SarantosUniversity of Maryland, College Park, MD 20742, USAA.L. SpragueLunar and Planetary Laboratory, University of Arizona, Tucson, AZ 86721, USAE.T. Bradley · W.E. McClintockLaboratory for Atmospheric and Space Physics, University of Colorado, Boulder, CO 80303, USA162 D.L. Domingue et al.1 IntroductionThe discovery of an atmosphere, or more accurately an exosphere, around Mercury wasmade through the ultraviolet airglow and occultation experiments on the Mariner 10 space-craft during its three flybys of the planet in 1974 and 1975. The Mariner 10 occulta-tion experiment set an upper limit on Mercury’s atmospheric density of approximately105atoms/cm3, corresponding to a pressure of about 10−12bar (Broadfoot et al. 1976;Hunten et al. 1988), thus defining it as a collisionless exosphere with its exobase coinci-dent with Mercury’s surface: a surface-bounded exosphere. Ultraviolet (UV) emissions ofthe three atomic elements, hydrogen (H), helium (He), and oxygen (O), were detected withthe UV airglow spectrometer (Broadfoot et al. 1976;Kumar1976). Since the Mariner 10 fly-bys, exploration of Mercury’s exosphere has been conducted by means of ground-based tele-scopic observations. Three additional elements, sodium (Na), potassium (K), and calcium(Ca), have been detected through their resonance scattering emission lines (Potter and Mor-gan 1985, 1986; Bida et al. 2000). Because the combined pressures of the known species aremuch less than the total exospheric pressure measured by the Mariner 10 occultation experi-ment, other species are expected to exist in this tenuous atmosphere. Additional constituents,such as carbon (C), carbon monoxide (CO), carbon dioxide (CO2), lithium (Li), argon (Ar),neon (Ne), and xenon (Xe) have been sought but not detected (Broadfoot et al. 1976;Fink et al. 1974; Hunten et al. 1988; Sprague et al. 1996). Other species, such as hy-droxyl (OH) and sulfur (S), have been suggested (Slade et al. 1992; Butler et al. 1993;Sprague et al. 1995) and modeled (Killen et al. 1997; Koehn 2002; Koehn et al. 2002)as related to the radar-bright deposits near Mercury’s poles (Harmon and Slade 1992;Slade et al. 1992).Telescopic observations from the mid-1980s to today have shown that there is tempo-ral and spatial variability in Mercury’s exosphere. The elements have both high- and low-velocity components and are influenced by the thermal and radiative environments in ad-dition to the interstellar medium. For example, the variability in exospheric Na has beenmapped to variability in the solar wind (Killen et al. 1999, 2004a, 2004b) and its effects onMercury’s magnetosphere.Mariner 10 also made the first in situ measurements of the planet’s magnetic field (Nesset al. 1974; Simpson et al. 1974) and the space environment around Mercury (Ogilvie etal. 1977). During the first flyby the spacecraft passed through the magnetotail of the planetand provided the first hint that Mercury may have a magnetic field similar to, though oflower amplitude than, the Earth’s. The second flyby passed across the dayside of the planet,and the third again crossed the tail, this time closer to the planet’s surface. Analysis ofthese data showed that the planetary magnetic field was probably a dipole with a momentof 350 to 400 nT-R3M, oriented within 10◦of the rotational axis (Connerney and Ness 1988).Additional details concerning Mercury’s magnetosphere can be found in a companion paper(Slavin et al. 2007).The fundamental observation, however, was that the magnetic field of Mercury is ableto stand off the solar wind, at least under nominal solar wind conditions. This implies adynamical coupling to the planet that is mediated by magnetospheric current systems thatmust close near or within the planet. At Earth, the corresponding current systems close inEarth’s ionosphere, but Mercury has no ionosphere. How do the required current systemsclose at Mercury? One hypothesis is that Mercury’s exosphere provides a so-called “pick-up conductance,” derived from the ionization and electric field acceleration of atmosphericspecies, that enables the formation of an Earth-like magnetosphere despite the absence ofan ionosphere (Cheng et al. 1987;Ip1993). Another proposed mechanism for closing cur-rent systems is based on the assumption that the surface of Mercury is itself conductingMercury’s Atmosphere: A Surface-Bounded Exosphere 163(Janhunen and Kallio