I have a Rotax 912S and my experience was a little different.
For one thing my plane is a pusher and the carbs are stuck out there in the cold part of the airstream just above the wing. The plane is just 25 hrs old and I have not yet decided whether to install carb air heating.
I went flying in December, it was cold, about 2 or 3 deg C and there was wispy low cloud in the valley, otherwise sunny.
I flew for about 40 mins and decided to to do a couple of circuits before packing in. Circuit and approach were normal. To minimise vibration on the central style mounting I usually keep revs up above 2,500 on approach and roundout and then just cut the throttle when she settles. This time when I did that she just spluttered and quit.
I restarted the engine easily enough, sat on the runway ( the place was otherwise deserted,) and tried a few throttle advances from idle. It seemed sudden advance would make her splutter and even quit but a medium steady advance would not. ( I have not had this before when testing, the engine always seemed to pick up ok now matter how I advanced the throttle.)
After a while the problem seemed to go away and after checking the throttle for quick advance seven or eight times I decided to do a close in circuit. The circuit went fine, absolutely no problem till I closed the throttle on roundout. Same thing, splutter, quit.
Again after about five minutes of ground running I was back almost to normal but decided to ask around before testing any further.
One forum I belong too suggested "Change to Winter jets." though I can see no mention of this in the installation or running handbooks. (I have however jacked up the idle to 1800 pro tem.)
Looking at the pictures on a UK agent for Rotax I see that if icing occurs it can occur right out there where the butterfly is and I am guessing, that if it did, when the butterfly was closed there would be no remaining air passage for the idle air, so the engine would quit when the throttle was closed. Without the cooling effect of Zero deg C airflow the carbs would absorb enough radiant heat from the engine to melt the ice and I'd be fine until we were back in the air when icing would start again.
I flew around both in sunshine and shadow for that 40 mins and there was no apparent effect on power or smoothness and that would have allowed plenty of time for a serious build up of carb ice so I'm wondering if there is something about the shape or operation of the carbs that inherently inhibits larger ice build up.
One concern, with a pusher, is that any parts not perfectly secured etc damage the prop, can even write it off, if they seperate so I tend to be a bit careful about what I put on that area of my plane.