This paper presents initial results from the Weather Prediction Uncertainty Management And Representation (PUMAR) project sponsored by the Army Research Laboratory, White Sands Missile Range in New Mexico. The goal of the PUMAR project was to cognitively engineer a user-friendly software tool to effectively communicate weather forecast uncertainty to the operator. The development of the PUMAR software began with the development of an operational scenario incorporating an April 2004 weather event of interest over New Mexico. Then, a detailed cognitive task analysis was performed to determine critical parameters that influence operational decisions. This analysis led to the identification of key information requirements, including the need to interact with and understand the uncertainty inherent in the weather forecast. One of the greatest challenges in meteorology is quantifying and communicating the uncertainty in weather forecasts, especially those forecasts based on numerical weather prediction (NWP) information. Weather decision aids, widely used in military operations, are often exploited to "translate" the NWP information into actionable weather intelligence. However, in several places in the NWP process unrepresentative or erroneous environmental observations, errors in the data assimilation process, and inaccurate or imprecise microphysics or numerical analysis techniques can lead to uncertainty in the output data. Extensive, on-going research on each segment of the NWP process is leading to the development of better analytical techniques, but the issue of quantifying the uncertainty remains a challenge. Ensemble techniques applied to NWP have proven to be valuable forecast tools but interpreting the output is often difficult, and they have demonstrated sensitivities to initial conditions. Bayes' Theorem and Bayesian Hierarchical Model (BHM) techniques have recently been applied to the NWP process with documented success at quantifying weather forecast uncertainty. The PUMAR software incorporates the BHM approach using the commercially available BNet:Builder software package, which allows the user to graphically interact with the NWP information to represent and thus understand the types and sources of uncertainty in the forecast process. In this manner, the PUMAR software aids in the understanding and management of uncertainty in weather forecast data. The PUMAR software, including the BNet:Builder package, could be easily extended to other hydrology and climatology applications requiring an assessment of information uncertainty.