A series of numerical experiments using the Advanced Regional Prediction System (ARPS) is conducted with the objective of assessing the impact of data assimilation on the simulation of an idealized bow-echo storm.

The ability of such an organized convective system to maintain a coherent structure for several hours indicates the prospect for considerable predictability. Moreover, previous investigations suggest that the longevity of bow-echoes can be strongly associated with the interaction of processes on different spatial scales (e.g. synoptic-scale vertical wind shear, convectively-generated surface cold pool).

In this study, a high-resolution control run of an idealized bow-echo is developed in order to generate simulated data to be assimilated on somewhat coarser grid runs. The main goal is to evaluate the relative importance of different meteorological variables (mass, wind) upon the simulation of a bow-echo as well as the impact of data at different spatial scales. Is the assimilation of mesoscale (synoptic-scale) data sufficient to capture the evolution of a bow-echo, or are smaller-scale data required? In other words, can large-scale information adequately generate small-scale structures? Do the small-scale structures control the large-scale? Or is there some sort of balance among the two. We address these questions via the selective assimilation of different data on multiple scales and subsequent evaluation of the manner in which information is communicated among them.