ChrisN
This is not an indicator of the risk severity.
That is calculated by Probability x Impact (no pun intended.)
So, if the prob is 0.01 (ie 1/100th of a percent) x 100 (on a scale of 100), the severity will be 1, i.e very low indeed.
However, all we have proven is the limitation of of quantitative analysis, since qualitative risk analysis (as demonstrated by ShyTorque) illustrates the potential risk scenarios and how they can occur.
It then becomes a matter of taking a view.
Presently, it is legal to cloud climb in gliders, which suggests that the lawmakers take a view that the severity is very low.
But, and it is a big but IMHO, the laws were drafted years ago in an era where there was less traffic movements and some might argue that this should be considered.
As we tend to plan to fight the last war, we end up with a status quo and this is usually broken by an incident that compells us to take notice.
One argument is that cloud climbing is justifiable, whereas a contrary point of view would be that stopping cloud climbing would break a link in a particular accident chain.
However, glider pilots are entitled to climb in clouds and the4 rest of us must hope that they do so with skill and discretion.