Observing, analyzing, and modeling mesoscale weather phenomena

Recent progress and current research in this field are placed in perspective, along with key problems to be addressed in the coming years. Modeling has progressed further than analysis because mesoscale analysis requires assimilation of information from diverse observing systems. Physical equations are essential tools in data analysis; four‐dimensional dynamic assimilation is a powerful but complex tool. Dynamic model predictions permit observations at different times to be interpreted in a consistent manner. Key observing systems and field programs are discussed briefly. Analysis techniques are discussed in relation to dynamic models with which they interact. Different classes of dynamic models are discussed, including prediction models, simulation models, and parameterization schemes. The relation of research on different classes of models is described. The emphasis throughout is on systems of models that use real data and can be verified against real data. Subsections are included dealing with communications among researchers and analysis of precipitation. The conclusion is that even more rapid progress is likely in the coming years, but the key problem will be four‐dimensional data assimilation. The challenge is to efficiently combine diverse types of data and provide for effective collaboration of diverse scientific specialties. When the challenge is met, this field will have advanced to the point that operational applications will be feasible and productive. An extended bibliography is included and sources of important technical reports are identified.