What if you're an Italian seismologist, and you want to produce a prediction which is in some way useful, but still expresses an appropriate level of doubt to a

What if you're an Italian seismologist, and you want to produce a prediction which is in some way useful, but still expresses an appropriate level of doubt to a lay audience?