Our goal is to predict environmental phenomena, such as the number of tropical storms in a given hurricane season, using both point and interval estimates. We use Bayesian linear regression and Bayesian negative binomial regression models, which can accommodate missing data. Our models incorporate variable selection uncertainty via a stochastic search variable selection prior. Using real data, we illustrate that our models can improve upon the results of some non Bayesian models in the literature. This is joint work with Xun Li and Gabriele Villarini.