Observations of H and D densities and escape fluxes from the upper atmosphere
of Mars with the MAVEN IUVS echelle channel
J.T. Clarke, Center for Space Physics, Boston University, Boston USA (jclarke@bu.edu), M. Mayyasi CSP /
BU, D. Bhattacharyya Univ of Illinois UC, Urbana USA, J-Y Chaufray LATMOS France, and the MAVEN
IUVS team Univ. of Colorado Boulder USA
from late MY 32 to mid MY 35.
Introduction: The history of water on Mars is
of great interest to understand terrestrial planet evolution. The constituent atoms of water are known to
escape the weak Martian gravity into space, while
heavier water molecules remain in the atmosphere.
A key indicator of water loss is the ratio of the isotopes deuterium to hydrogen. The lighter H escapes
faster than D leading to a gradual increase in the D/H
ratio of the remaining water. The present D/H ratio
is proportional to the total water Mars has lost in its
lifetime, but deriving the depth of a primordial Martian ocean requires understanding the physical processes that control escape today to accurately extrapolate back in time. Observations of UV resonance
line emissions from H and D with the MAVEN and
Hubble Space Telescope (HST) missions provide the
key atomic H and D densities and escape rates.
Observations: H and D atoms in the Martian
upper atmosphere are separately measured by resolving their UV Lyman-α emissions at high spectral
resolution. H Lyman-α emission from Mars is produced by resonant scattering of solar photons by H
and D atoms in the upper atmosphere. The incident
sunlight penetrates to a depth where it is absorbed by
CO2, roughly 80-100 km altitude depending on the
angle of incidence. Both D and H atoms in the Martian upper atmosphere scatter photons in the spectrally broad solar emission line, with a wavelength shift
between the D and H lines corresponding to the
small difference in energy levels in the isotopes.
Relating the emission brightnesses to densities requires the application of a radiative transfer model
that accounts for multiple scattering of photons in
the atmosphere. This leads to complications involving a degeneracy between derived values of the
atomic density and temperature, which have been
carefully studied and reported. In this work we have
made a fit to upper atmospheric temperature measurements by the MAVEN NGIMS instrument to
remove the degeneracy.
Another complication
comes in relating local measurements to global average values to derive the total loss rates. In this report
functional relationships for the variation in temperature and density with solar zenith angle are applied
to the data. Measured values at each location on the
planet have been converted to the values at the subsolar point, and then integrated across the Martian
globe with the assumed functional relations for temperature and density with solar zenith angle to derive
global escape fluxes. Observations will be presented

Results: As previously reported there are strong
repeated seasonal increases in the densities and escape fluxes of both H and D atoms each Martian
year near perihelion, consistent with a strong
upwelling of water tied to lower atmosphere dynamics needed to supply the atoms. Around perihelion
there appear dramatic changes on time scales of a
couple of weeks, providing a means to estimate the
loss processes and fractions. A process in addition
to thermal escape is required to explain the observed
changes in D density, likely from superthermal processes that produce hot atoms. This implies that at
present D and H escape at close to their ratio in water, providing strong evidence for the scenario of a
warm, wet primordial Mars with a large depth of
water on its surface. In addition the D/H ratio appears elevated compared with the ratio in Martian
water, consistent with rapid escape of H atoms from
the upper atmosphere. Changes in the atomic D/H
ratio over each year result in changes in the fractionation factor that describes the ratio of D/H in the
escaping atoms compared with the D/H ratio in Martian water. These results will be presented and the
implications will be discussed.