Yesterday our local news was all about a major river event just a few miles from our cabin. A major landslide sent a "bore" of mud, boulders and debris downriver, scouring it from bank to bank. This seemed like a great opportunity to use the Phantom and my Drift action cam for some dramatic footage. (I'm still a noobie but gaining confidence in flying this thing.) I had to launch from a plastic box set among the rocks but it went smoothly. There was a brisk upriver breeze so I decided to fly in GPS mode, as I was nervous about keeping the Phantom well under control. I also taped a small 100 lumen LED headlamp above the battery door, to help me with orientation. Based on previous flights with the Drift camera, I set my kitchen timer for 7 minutes. Long story short... Somewhere just past 6 minutes I was bringing the Phantom in close and low when it began to descend steadily toward the rocks. I tried to bring it to my feet but didn't quite make it. It crashed into the stream very close to me, bashing into a rock (gouged the camera's lens protector). I fished it out very quickly -- it was still running -- disconnected the battery and drained the water. One motor turns reluctantly, with a distinct grinding sensation. Otherwise, no apparent damage either externally or internally. Looking back on this I think I underestimated the battery drain while flying in wind, in GPS mode. Also, although the LED headlamp is very light, it did add some weight and drag -- and I forgot to compensate for that. This leads me to a question: I checked the battery voltage and it shows 11.3v, which I believe is "normal" at the end of a flight. Did a low battery actually cause the Phantom to descend or is it more likely pilot error? At what point does a low voltage actually cause a forced landing? I'm not clear on that.... Awaiting a new motor now as I don't trust the gritty one.