Severe Weather Probability Question

I am trying to get a useful probability of various environmental
disasters for my area (severe storm, severe winter storm, etc.) I have
an exhaustive list of the occurances for the last 100+ years. I can
think of three ways to compute the probability of an event in a given
year:

1) Simple mean number of events for the last n years (31 in 100 years =
31% chance). I don't really like this because it doesn't account for
the wide variance in time periods between events (2 weeks - 6 years)

2) Use a poisson distribution based on the time between events and then
find the likelihood that one will occur during a 1 year period.

3) Determine mean and standard deviation of events per year and then
find the area under 1 or more events in a year.

Which of these seems the best? What other methods could I use (please
no calculus or differential equations, I'm a math wimp).

Alternatively, where can I find stats like this online for a relatively
small geographical area (a single county or zip code)?

Thanks,

-Dave

.