Let's get something clear right away, Mars has lots of water on and close to its surface. Over the years we've seen an increasing number of measurements and discoveries that indicate very significant reservoirs of martian water. The catch, and there is a big one, is that this water is frozen. It's not surprising; Mars has an average surface temperature of -55 Celsius (-67 F), and with average atmospheric pressure only 0.7% that on Earth, water hovers on the solid side of its triple-point - the magical spot where ice, liquid, and gas can all co-exist. Even though local conditions can sometimes warm up to above zero Celsius, pure water is more likely to sublimate directly into vapor rather than ever be liquid. On Mars solid or gaseous water is the norm.
Take for example the Mars Reconnaissance Orbiter (MRO) radar mapping of the northern polar regions of Mars. Analysis of this data in 2008 indicated a 95% pure water-ice cap totaling approximately two to three million cubic kilometers in volume. That's roughly 100 times the size of all of North America's Great Lakes. This conclusion is backed up by earlier measurements with the gamma-ray spectrometer of Mars Odyssey made in 2001, and the in-situ observations of the Mars Phoenix polar lander. At the southern polar regions the Mars Express mission has used its radar instrument (Marsis) to detect a 90% pure water ice region extending as far as 4 kilometers below the surface and covering an area comparable to the state of Texas. If that water melted to liquid it would swamp all of Mars in a layer about 11 meters deep.
Other evidence is scattered across the martian geography. For example, in the mid latitudes of the northern hemisphere MRO has spied the fresh craters formed by incoming meteorites (yes, just like the Earth, Mars is under continual siege by the rocky detritus of the Solar System). Remarkably some of these craters have spewed up fresh, white, frozen water of almost 99% purity by volume. How do we know it's water? In this case there was enough material to allow for a spectrum of reflected light to be made, confirming a water composition. Also at these latitudes the temperature and atmospheric pressure is such that almost anything else (like frozen carbon dioxide) would vanish immediately. The material dug up by the meteors faded as it sublimated into the martian air exactly as you'd expect water to.
Even the sticky patch of soil that snagged the Spirit rover's wheel in 2010 revealed evidence of possible snow-melt seepage into the ground in recent geological history - carrying soluble salts into different soil layers.
These examples, together with extensive evidence for ancient water shaped topography and water influenced chemistry that is seen across many areas of Mars, all indicate that the red planet harbors a significant amount of frozen water close to its surface. This includes substantial glaciers buried beneath the surface at mid-latitudes. Although a more minor component we also know that Mars has water-ice clouds. The Phoenix lander monitored high altitude water-ice cirrus that was even precipitating crystals to the ground - so Mars has a hydrological cycle, albeit on a vastly reduced scale compared to that on Earth.
There are certainly catches. For example, frozen carbon dioxide can confuse our sightings. At high latitudes it sticks around much like water ice. We end up having to rely on its distinguishing volatility - easily sublimating into gas in the thin martian atmosphere - and rapidly appearing and disappearing.
The really big issue is whether or not there are locations anywhere on Mars where water can, even if only temporarily, exist in a liquid phase. Liquid water is considered to be one of the key requisites for life. The dipolar water molecule is a unique biochemical solvent, a unique mediator of chemistry, and a unique component in terrestrial bio-geo-chemical cycles. Put simply, we just don't know of any way for life to exist without the availability of mobile water molecules. It is so central that in astrobiology there is a rule of thumb that says if you want to look for life - on Mars, on icy moons, or on rocky worlds orbiting other stars - you "follow the water".
This idea has driven the search for signs of places on Mars where liquid water may exist today, even if only temporarily. Which brings us to the most recent flurry of interest in the subject. In a recent paper in Science, McEwen et al. present evidence for a new phenomenon seen with the HiRISE imaging camera onboard MRO. They call it "recurring slope lineae", the appearance and disappearance of dark finger-like features on the steep slopes of a crater called Newton in the Terra Sirenum in the martian southern highlands at mid-latitudes. These streaky features are narrow (roughly a few meters in width) and long (up to hundreds of meters) and most critically they appear during the relatively warm summer months on Mars, disappear when it gets cold, and reappear next spring (for an animation click here). They are also numerous, probably thousands have been captured by the HiRISE instrument.
So what's going on? The placement, form, and seasonal (temperature) relationship of these features strongly suggests that they are produced by the flow of material - most likely a liquid. Local conditions make carbon dioxide sublimation unlikely - since solid CO2 would have long since vanished. Salt rich water, or brine, could meet the requirements at the near-surface environment of the crater walls. Chlorides or sulfates in water can drop its freezing point by as much as 70 Celsius and this region of Mars is known to be rich in ancient chlorides.
This liquid, if present, is not cascading down the slopes like a mountain stream. Rather it is likely to be a subsurface flow that somehow alters the visible texture and/or reflectivity of the dry martian regolith on the immediate surface - it's like a tumbler of liquid spilled into a shag-pile carpet, squishy underfoot but you don't see the liquid directly. And herein lies one of the tricky aspects of interpreting these extraordinary streaks, there's still a possibility that they are actually dry.
Back in 2006 it was announced that NASA's Mars Global Surveyor mission had spied a set of appearing and disappearing streak-like features on the surface, again in the Terra Sirenum region. In this case gullies in a steeply sloping terrain appeared to brighten with time. The immediate thought was that these could be signs of flowing water, bursting out from an unseen aquifer. However in this case there was no clear seasonal variation or correlation, and the shape or morphology of the streaks was a little different. In fact later analysis by Pelletier et al. came to a rather disappointing conclusion (for those interested in liquid water), the shape of these gullies was better matched by the tumbling flow of completely dry, granular material down a slope.
As material tumbled it would freshen up the surface, causing the observed brightening. A landslide, not liquid, seems to be a better explanation.
The new results of McEwen et al. are however quite different. These features darken, and they also visibly grow down the slopes with seasonal time. By contrast, landslide features could presumably form quite rapidly. The new dark streaks are also in warmer conditions, further suggesting a different origin.
Water about as salty as Earth ocean's would melt at the temperatures expected in the soil of the Newton crater walls, and on the basis of the current evidence this looks to be the best explanation for what we're seeing. Salty, muddy flow is a good bet - future laboratory work may help confirm or dispute this.
In the search to see whether Mars could harbor life this is incredibly exciting news. There are many terrestrial microbial organisms that thrive in salt-rich environments. In fact these so-called "halo-philes" (salt lovers) tend to be among the hardiest of Earthly life, dealing extremely well with dessication (as might happen during the chill martian winters) and nasty ionizing radiation (as exists on the martian surface). It's not inconceivable that organisms with similar traits could lurk in the martian subsurface in places like Newton crater, flowering during the comparatively temperate and moist spring and summer before returning to dormancy over the winter. Of course we'd have to go look up close to know whether this is true or not.
Mars is certainly not devoid of water, it is also rich in salts, and conditions hover close to the triple point of pure water. Different locations and the variation of seasons provide conditions of temperature and pressure that should allow for salty, briny water to be temporarily liquid at or near to the surface. It may be that the thousands of dark streaks seen in Terra Sirenum really are the signs of Old Briny stirring from its winter slumber.