Climatological analysis of tornado reports is complicated by reporting errors, non-gaussian nature of count data and rare events, and the presence of non-stationary spatial and temporal correlation. Analyses to date have estimated sample statistics such as sample mean frequency or sample correlation between tornado counts and climate indices. Hypothesis tests are often applied to these estimates of sample statistics. An underlying assumption of this approach is statistical indepence of observations. Although some studies have accounted for the complicating factors mentioned above (e.g.,preprocessing the data to remove spatial and temporal correlation),typically such factors have not been formally included in the underlying statistical model used for inferential decisions.

We have taken an alternative approach in which a statistical model is constructed so that the aforementioned characteristics of tornado reports are explicitly accommodated in the model structure. In particular, a hierarchical Bayesian framework is considered, in which the various complicated structures are considered in a series of conditional models, formally linked by basic probability rules. This formalism allows one to evaluate characteristics of the spatio-temporal variability of an underlying (unobservable) tornado count process given the noisy observations. In addition, one can consider the effects of exogenous processes on this tornado count process. For example, preliminary analysis shows that the North Atlantic Oscillation (NAO) is potentially more important for modeling variability in tornado counts than ENSO.