The atmospheric state is variable, depending on complex physical processes that cannot be known exactly. Thus contemporary numerical weather prediction (NWP) approaches use ensembles of model output to diagnose the most probable atmospheric state. Ensemble deviations from this state are related to the NWP uncertainty. Atmospheric transport and dispersion (ATD) models often derive their transporting wind fields from NWP sources. The predicted contaminant plumes also inherit the ensemble variability, expressed as a dispersion uncertainty. Parameterization of NWP uncertainty for dispersion models, however, is only recently being addressed. This work discusses a Taylor-dispersion-based approach that relates NWP uncertainty to dispersion uncertainty modeling metrics. The parameterized effects of NWP uncertainty on dispersion are illustrated using NWP ensemble data and the second-order closure integrated puff (SCIPUFF) model as the ATD tool.