36th DPS Meeting, 8-12 November 2004
Session 37 Mars Atmosphere
Poster II, Thursday, November 11, 2004, 4:15-7:00pm, Exhibition Hall 1A

[Previous] | [Session 37] | [Next]

[37.07] The Assimilation of Ground- and Space-based Observations of the Martian Water Cycle

H. Houben (Bay Area Environmental Res. Inst.), A. L. Sprague (LPL, University of Arizona), M. D. Smith, J. C. Pearl (NASA-GSFC)

We have developed a forward-trajectory semi-Lagrangian transport model to assimilate observations of martian water vapor from both ground- and space-based instruments. While the Viking Mars Atmospheric Water Detector and Mars Global Surveyor Thermal Emission Spectrometer (TES) have generated an excellent description of the annual variation of the atmospheric water vapor column, a number of issues remain. Because all of the usable TES observations have been made around 2 pm, measurements of the diurnal cycle must come from other sources. Ground-based observations provide such information, albeit at low spatial resolution. These observations suggest a much larger variation in the diurnal water vapor column than has been heretofore generated in numerical models. Whether this variation is due to interactions with clouds, the surface (e.g., regolith exchange), or some other process is unknown. In any case, understanding the cause will require knowledge of the vertical distribution of water vapor as well as other meteorological variables. Our approach is to assimilate the available water vapor observations along with TES 15-micron observations that constrain the temperatures and winds. The diabatic circulation (i.e., the Hadley cell) that is so important to meridional transport on Mars is digitally filtered to eliminate non-physical gravity wave perturbations. The semi-Lagrangian tracer transport proceeds in three steps. First, tracers propagate downstream from model grid points (preserving the positive definiteness of the fields). The resulting fields are then remapped to the model grid. Finally, condensation and precipitation redistribute the water within a column. This method is especially suited to our purpose since data assimilation is precisely the remapping of the proper linear combination of calculated and observed fields. We pay careful attention to the relative calibration of the ground- and space-based observations, as the diurnal variation in observing geometry can exaggerate or mask true diurnal variations when the calibration is in error.

[Previous] | [Session 37] | [Next]

Bulletin of the American Astronomical Society, 36 #4
© 2004. The American Astronomical Soceity.