The present invention relates to a computational fluid dynamics (CFD) and multi-data sources-based urban real-time global environment estimation method. The method is based on historical data of urban environment monitor stations, global mesoscale meteorological forecast results, national meteorological data, urban major pollutant source data, a urban geographical three-dimensional model and road surface real-time monitor data of a motor vehicle exhaust detection system, CFD is used as a calculation engine, an environment quality mode is adaptively switched according to meteorological information, a multiscale grid discrete urban model is used, a multi-component pollution model is introduced, and a meteorological dry and wet settlement process is considered, so as to establish an urban global air and environment quality real-time estimation method. The urban environment quality estimation method is much better than existing methods in comprehensiveness, accuracy and real-time performance.