Abstract: The tornado formation is largely influenced by the synoptic-scale and sub-synoptic scale anomalous systems. In this paper, the “anomalous weather chart”, which removes the temporal climatology from the conventional weather chart, is applied to analyze the tornado-related anomalous systems and evaluate the global forecasting models’ (ECMWF model and NCEP GFS) performances in predicting such anomalous systems. In this paper, two cases were studied: 1) the recent tornado event occurring in Funing County, China on 23 June 2016 and 2) the strong tornado outbreak in Southern US during 27-28 April 2011. Results showed that both cases are associated with the lower-level anomalous low pressure, upper-level anomalous high pressure systems and the contrast of anomalous warm/cool air masses in the troposphere. The ECMWF model and the NCEP GFS can predict the tornado-related anomalous system for the case 1 in advance of 54 hours and 18 hours, respectively. Furthermore, the ECMWF model is able to predict the spatial structures of strong geopotential height and temperature anomalies, and their evolution for the case 2 in advance of 3, 5, 7, 9 days.