Scientists (and science fiction writers) have speculated about the presence of water on Mars ever since Percival Lowell speculated about canals on the Red Planet. In the last decade the exploration of Mars by robotic probes has revealed a wet past for the planet, with dried up oceans, lakes and rivers and (more interesting to geologists) evaporite deposits (sediments formed when mineral laden water evaporates). But the general consensus is that modern Mars is a dry, waterless place, all its water having either been frozen solid or evaporated away into space.
Layered sedimentary rocks in Victoria Crater, Mars.
Imaged by NASA's Opportunity Rover.
In the 5 August edition of the journal Science, a team lead by Alfred McEwen, of the Lunar and Planetary Laboratory at the University of Arizona, published a paper in which they speculate that dark streaks seen on north-facing slopes in Mars' southern hemisphere by the Mars Reconnaissance Orbiter might be caused by outflows of briny water in the Martian summer. These streaks were noticed by student Lujendra Ojha, who observed that they were seasonal, appearing in the southern hemisphere summer, then disappearing in the winter, indicating that these were not ancient features, but had to be explained in correlation with modern Martian seasonal variation.
Seasonal dark streaks on the Martian surface.
This is not a completely new finding, it was first announced at the 42nd Lunar and Planetary Science Conference in March, and there have been numerous prior claims of running water on the modern Martian surface, all of which have evaporated away, so McEwen et al. are rightly cautious, stressing that at this time this is only a theory.
Salt water has a lower freezing point than fresh water, and the more salty water gets, the colder it can get whilst still remaining a liquid. In 2008 NASA's Phoenix Rover discovered the presence of salt crystals in the soil in the Martian north polar region, so it is not unreasonable to expect that any remaining water trapped in the Martian soil might be extremely salty, and therefore remain a liquid and flow at temperatures significantly bellow 0°C.
Salt deposits unearthed (unmarsed?) by the Phoenix Rover.
However there are still a couple of problems with this theory, and McEwen et al. do not claim to have an explanation as to how these salt water flows could occur.
Firstly there is no such thing as frozen salt water; when water freezes the water segregates from the salt. In polar regions on Earth it is possible to obtain fresh water by melting frozen sea ice. Thus, while water will freeze at lower temperatures when it is salty, once it is frozen it must be raised above 0°C to melt again. So any salty water on Mars would have had to remain a liquid since ancient times, and not be forming seasonally. Liquid water, no matter how salty, flows all year round; it is not seasonal, and will head downwards all the time it is a liquid. It is difficult to imagine how any liquid water could remain high enough to flow downhill suddenly after billions of years without replenishment, and there is no known mechanism which would cause this to happen on a seasonal basis.
Secondly it is unclear how water could remain a liquid on Mars' surface at any temperature, due to the low atmospheric pressure. Mars has only about 1/170 the air pressure of Earth (an average of 600 pascals on Mars, as opposed to 101 300 pascals on Earth), which should make liquid water impossible. On Mars water should sublimate, passing from a solid to a gas and back with no liquid phase as carbon dioxide (dry ice) does on Earth. Salt does not influence this process one way or another.
See also NASA sets its sights on Gale Crater and Mars on Sciency Thoughts YouTube.