Extreme value methods have commonly been used to predict and quantify uncertainty around environmental or climatological events that could have high impact on human ca- sualties or costs (e.g. earthquakes, hurricanes, flooding, wildfires). In this work, our focus is to study the number of casualties as the variable of interest, from the Global Terrorism Database (GTD) for a particular region and time frame and characterize events via finding extreme observations and fitting both a Generalized Extreme Value (GEV) and General- ized Pareto Distribution (GPD) to this data. We assess whether the goodness of fit of the GEV and GPD parameters are adequate for our framework. For the latter, we also provide graphical representations of predicted 95% and 99% quantiles based on our models and compare these to the actual data. The results of these analyses are a building block into the development of a representative Bayesian hierarchical model that fully characterizes the spatial-temporal relationships present in extreme events from the GTD.