Short-range dynamics and prediction of mesoscale flow patterns in the MISTRAL field experiment

Abstract

In a limited area of about 50 km by 50 km with complex topography, wind measurements on a dense network were performed during the MISTRAL field experiment in 1991/92. From these data, the characteristic wind fields were identified by an automated classification method. The dynamics of the resulting 12 typical regional flow patterns is studied in the present paper. It is discussed how transitions between the flow patterns take place and how well the transition probabilities can be described in the framework of a Markov model. Guided by this discussion, a variety of prediction models are described which allow a short-term forecast of the flow-pattern type. It is found that a prediction model which uses forecast information from the synoptic scale has the best forecast skill.