A BRIEF CLUSTERING ANALYSIS OF THE MEXICO CITY LOCAL WIND STATES OCCURRED DURING THE MILAGRO CAMPAIGN

摘要

A clustering analysis of the Mexico City local wind events that occurred during the MILAGRO field campaign was carried out for identifying their main characteristics and patterns. This study was performed with the hourly wind data provided by the official atmospheric monitoring network for March 2006. The local winds of the period were described with a meso-P scale lattice wind model endowed with a wind state concept defined by the wind velocity horizontal components and the wind's divergence and vorticity. The wind states produced by the lattice wind model were analyzed through their occurrence frequencies and hierarchical cluster analysis. This procedure revealed wind patterns very similar to those already described by other authors. However, new and interesting features were also revealed such as a strongly stable cluster composed by wind states befallen from sunrise to midafternoon, and that the 4-group clustering revealed itself as the more stable wind states organization, while those with 6 and 8 clusters resulted the less stable ones. A Fourier transform analysis of the wind states time series evidenced wind patterns driven by the diurnal cycle of incoming solar radiation, and other patterns that may be associated with the urban heat island phenomenon.