Actually, the risk of disasters is not negligible at all, in particular if there is a sufficient number of power plants (431 worldwide currently). In the last

Actually, the risk of disasters is not negligible at all, in particular if there is a sufficient number of power plants (431 worldwide currently). In the last 30 years, we have already seen two incidents that were above the "maximum credible accident", that is the probability of them happening was considered to be negligible. With that number of plants running, I would expect a major incident _somewhere_ every 20 years and historically we have seen more than that. Here [0] is a quite recent study by probably the most important german research institute that predicts one incident every 10 to 20 years, based on the number of plants running and data of historical incidents. There is also a nice summary in German [1]. [0] http://www.atmos-chem-phys.net/12/4245/2012/acp-12-4245-2012.html http://www.atmos-chem-phys.net/12/4245/2012/acp-12-4245-2012... [1] http://www.mpic.de/index.php?id=34298&type=0 http://www.mpic.de/index.php?id=34298&type=0