There are a few probable causes. I'd say that it's usually a combination of two or more factors, as is the case in most times (RC or otherwise).
Flying in extreme cold can cause a LiPO battery to shut down if pushed too hard. LiPOs work best around 30 to 40 degree Celsius, too much heat for too long can damage the battery but during operation the cold can also prevent the battery to reach optimum usage temperature (internal chemistry, IR, that stuff...) and it won't be able to provide the discharge that motors demand at certain situations.
LiPOs have a DISCHARGE RATING that works for both "continuous load" (normal use/drain) and "burst load" (high drain during short - like, 10sec. - periods). Quads in general don't drain that much power unless you're doing 3D or pushing really hard on takeoffs, hard accelerations, etc. Then you can top the burst capacity and go kaput if it's already too low on voltage.
Also, cells can get out of balance and thus provide wrong readings, and management systems will usually go for the lowest reading (i.e. one cell reaching 3.0V would make it shut down for instance) and so on. I don't know much about the P3 management system but I'd guess it's like that to protect the battery and aircraft. Anyway, voltage variations are not linear, you can have 3.65V during a sharp takeoff then after a couple of seconds it would go back to 3.8V when you start cruising slowly and it can mislead the system management.
So, if you're flying in the cold and at the same time doing something that could push the battery too hard to its limits, it can lose capacity to deliver enough power at a point during flight. Depending on the config of the management system or even the battery chemistry itself, or if it has some unbalance between cells, it would "think" that the battery has reached critical level or even suffer a sudden drop down and go bellow the aircraft operation minimum and it would cause the AC to fall.
In my experience, having just one factor at a time isn't enough to cause a problem serious enough to crash the AC. Most of the time anyway. Say, just flying in the cold but keeping it nice and easy until the battery warms up (or if it's not too too cold... Or one cell out of balance (if it's not too much off)... Or just pushing hard for short periods a few times then backing off a bit... If everything is OK with the AC and the battery, hardly one of those alone would cause it to malfunction. But if you add up 2 or more then it can overload the system somehow.
But then, in RC we usually "time" flights and take a close tab on batteries charges, monitoring and measuring it and keeping it safe. It's different from the P3 where we have it all measured in real time. It can be more or less accurate depending on a lot of things... and then there's also the firmware that controls it all. About that we may not know much, I'm afraid. So it's better to keep and eye on what we know about LiPOs, that's my opinion anyway!