The Mars Odyssey Gamma Ray Spectrometer presented the surprising finding of high hydrogen concentrations in near equatorial regions such as Tharsis. This was usually explained as water in chemically bounded form, not free water or ice:

Global Distribution of Neutrons from Mars: Results from Mars Odyssey.
"So, although our data do not rule out the possibility of subsurface ice deposits at low to middle latitudes, they do not require them. We suspect instead
that the NS data from these latitudes may be revealing the quantity of physically and chemically bound hydrogen in the upper tens of centimeters of martian soil that is buried
beneath a 15- to 20-cm-thick layer of hydrogen-poor soil."
Science, vol. 297, July 5, 2002, p.78
http://www.lpl.arizona.edu/~umpire/professional/papers/75.pdf

However, the report given at the December AGU meeting ascribes this hydrogen to former high atmospheric water vapor amounts during periods of Mars' high obliquity:

The weatherman of Mars.
"According to Los Alamos' principal investigator on the project, Bill Feldman, this spatial distribution of hydrogen deposits cannot be explained by the general north-south latitude gradient of water vapor that is in the present Martian atmosphere, thereby requiring different climatic conditions in the relatively recent past."
http://www.lanl.gov/worldview/news/releases/archive/02-134.shtml

09:30h
AN: U61B-04 INVITED
TI: Implications of the Near-Equatorial Distribution of Hydrogen for Past Near-Surface Water Deposits on Mars.
W C Feldman, W V Boynton, T H Prettyman, R L Tokar.
"Maps of thermal, epithermal, and fast neutron fluxes equatorward of 40 degrees latitude were studied to determine the abundance of hydrogen and its near-surface stratigraphy. Large areas centered on Arabia Terra and 180 degrees east longitude near the equator are found to contain enhanced hydrogen abundances that amount to between 3.5 and 7.5 percent water-equivalent hydrogen. The thermal and fast fluxes indicate that these deposits are buried below about 30 g/cm2 of hydrogen-poor soil. This spatial distribution of hydrogen deposits cannot be explained by the general north-south latitude gradient of water vapor that is in the present Martian atmosphere, thereby requiring different climatic conditions in the relatively recent past. A possibility is the diffusion of water vapor from a relatively water-rich atmosphere that is predicted to develop during periods of high obliquity encountered during the past several million years."
http://www.agu.org/cgi-bin/wais?w=U61B-04

The theory that equatorial regions on Mars would be desiccated near surface is based on the *assumption* that this water would not be replenished as it sublimated or evaporated over geological time scales.
However, periods of high obliquity on Mars occur as often as every 100,000 years to a few million years for the highest angles of tilt.
If the water at these low latitudes could be replenished as Feldman et.al. suggest, then it is possible that the water seen by GRS is in fact in free form as liquid water or ice - and at the 15 to 20 cm (6 to 8 in) depth implied by GRS.

Bob Clark