You and I gave essentially the same answer zmx. Basically, insufficient IR at that distance to light up the smaller snowflakes at the exposure speed of the camera. You just neglected to mention (don't know?) about the disappearing effect that happens to blowing snow/dust when you take a picture of it. I don't know if you live in a snowy area, but try driving down a highway in whiteout conditions where the driver can barely see to stay on the road. If a passenger is videoing the conditions to show how bad it was, when you look at the recording later you'll wonder why everybody was blinded at the time. And yes, I've done this and the difference is quite dramatic. The same thing happens when taking videos of blinding dust conditions. This is because of the relatively long exposure of the camera compared to what the human eye can see and the results I mentioned. You'll see maybe 1/3 of the blinding effect of flying snow without IR even in daytime with its much shorter exposures than a night scene with even a fairly high-powered illuminator. Same reasoning behind, say, a long multi-minute exposure when taking a night picture under starlight. If someone walks quickly through the shot they won't even show up in the final picture except as possibly a softening or blurring.
Again though, we're both basically saying the same thing. Other factors such as aggressive noise reduction may also be removing some/much of the snow in Larebear's second picture as night-time "noise", essentially because the snowflakes are farther from the sensor and IR and therefore smaller and much more dimly lit. The big streaks left in the first shot because of the brighter and relatively larger (because they're much closer) flakes can't be removed as noise because they're huge.