In order to remain liquid in such cold conditions (the research team estimate between -10 and -30 Celsius where it meets the ice above), the water likely has a great many salts dissolved in it.
In Robert Zubrin's fictional (but hard-science) book "First Landing" on the first manned Mars mission they actually find a hot spot and drill into it resulting in a geyser (which obviously, the water instantly freezes creating a snow-like effect). All of Zubrin's books are worth reading if you are at all interested in manned missions to Mars. He has some non-fiction looks (that are a bit outdated now due to scientific discoveries, like finding the polar caps are lousy with water ice) but he goes into the math and gets you in the right mindset. The fiction book is similar in nature to The Martian (came out well before The Martian) and he also has a short little fiction book meant to be sold to future colonists called 'How to Live on Mars: A Trusty Guidebook to Surviving and Thriving on the Red Planet' which also takes a hard-science fiction approach to the subject.
They are? The north cap has more than 800,000 cubic kilometers of water in it. Not much compared to Earth, but in space terms that seems pretty abundant to me.
lousy
1. very poor or bad; disgusting. "the service is usually lousy"
2. infested with lice; teeming with (something regarded as bad or undesirable). "the town is lousy with tourists"
From memory the Northern cap alone would cover Mars with something like 14 inches (it's close to 15 I think) of water :)
From my reading of the article, the primary evidence they have is a bright radar reflection. That is the long-wavelength equivalent of saying "we found some white stuff, it's probably water ice!" There are other things besides water which reflect strongly. The fact that the conditions at the site are extremely non-conducive to liquid water tempers the idea that there's water there. AFAICT, there is nothing null-hypothesis rejecting here. In fact, it seems barely as good as "it's not impossible that it's water." I suspect the current hoopla is premature at best.
They mention Lake Vostok at the end of the article which is 4km deep but it's probably not directly comparable.
Now if I could get my hand on a phase diagram for water on which I could find that point (263K, 5.1MPa) I could tell you how close pure ice is to melting under these conditions.
https://upload.wikimedia.org/wikipedia/commons/0/08/Phase_di...
From looking at the phase change chart[0], you don't really get into the "interesting" stuff as far as pressure goes until pretty deep (something like 9-10km, from looking at some sample number of pressure at various depths [1]). Given that they indicated in the article that this is under 1.5km of ice, I don't think that's deep enough to have pressure significantly change it, but would be happy to learn that I'm not correct in my reading of this :)
0 - https://physics.stackexchange.com/questions/60170/freezing-p... 1 - https://www.earthlearningidea.com/PDF/189_Pressure_rock.pdf - table on page 2
Pressure != compression, and water is otherwise sensitive to pressure. The boiling point of water varies by ambient air pressure, and in fact, the ambient air pressure on the surface of Mars is too low for liquid water, AFAIK; it boils away.
Also, see gmueckl's helpful comment: https://news.ycombinator.com/item?id=17610197