In atmospheric transport and dispersion both Data Assimilation and Multisensor Data Fusion provide frameworks for combining observations with forecasts to better estimate contaminant dispersion. In Multisensor Data Fusion the state of an entity is identified, while in Data Assimilation the state of a field variable is estimated. Both methods are applicable to Dispersion modeling because a contaminant filled puff is an entity and concentration data is a field variable. These two disparate frameworks give differing formulations for the problem. Data Assimilation directly incorporates concentration data into the governing equations while Multisensor Data Fusion uses concentration data for state estimation. We compare these frameworks both analytically and numerically to determine which is more robust when concentration values are contaminated by noise. The comparison evaluates similar Assimilation techniques and Data Fusion techniques. In order to compare the two analytically, we initially consider a contaminant release in stationary, homogeneous turbulence. This analysis illustrates the similarities and differences between Data Assimilation and Multisensor Data Fusion and helps to bridge the gap between the two.