Using a validated general circulation model, we determine where and for how
long the surface pressure and surface temperature on Mars meet the minimum
requirements for the existence of liquid water in the present climate syst
em: pressures and temperatures above the triple point of water but below th
e boiling point. We find that for pure liquid water, there are five "favora
ble" regions where these requirements are satisfied: between 0 degrees and
30 degreesN in the plains of Amazonis, Arabia, and Elysium; and in the Sout
hern Hemisphere impact basins of Hellas and Argyre. The combined area of th
ese regions represents 29% of the planet's surface area. In the Amazonis re
gion these requirements are satisfied for a total integrated time of 37 sol
s each Martian year. In the Hellas basin the number of degree days above ze
ro is 70, which is well above those experienced in the dry valley lake regi
on of Antarctica. These regions are remarkably well correlated with the loc
ation of Amazonian paleolakes mapped by Cabrol and Grin [2000] but are poor
ly correlated with the seepage gullies found by Malin and Edgett [2000]. In
both instances, obliquity variations may play a role. For brine solutions
the favorable regions expand and could potentially include most of the plan
et for highly concentrated solutions. Whether liquid water ever forms in th
ese regions depends on the availability of ice and heat and on the evaporat
ion rate. The latter is poorly understood for low-pressure CO2 environments
but is likely to be so high that melting occurs rarely, if at all. However
, even rare events of liquid water formation would be significant since the
y would dominate the chemistry of the soil and would have biological implic
ations as well. It is therefore worth reassessing the potential for liquid
water formation on present day Mars, particularly in light of recent Mars G
lobal Surveyor observations.