A New Method of Significance Testing for Correlation-Coefficient Fields and Its Application

Correlation-coefficient fields are widely used in short-term climate prediction research. The most frequently used significance test method for the correlation-coefficient field was proposed by Livezey, in which the number of significant-correlation lattice (station) points on the correlation coherence map is used as the statistic. However, the method is based on two assumptions: (1) the spatial distribution of the lattice (station) points is uniform; and (2) there is no correlation between the physical quantities in the correlation-coefficient field. However, in reality, the above two assumptions are not valid. Therefore, we designed a more reasonable method for significance testing of the correlation-coefficient field. Specifically, a new statistic, the significant-correlation area, is introduced to eliminate the inhomogeneity of the grid (station)-point distribution, and an empirical Monte Carlo method is employed to eliminate the spatial correlation of the matrix. Subsequently, the new significance test was used for simultaneous correlation-coefficient fields between intensities of the atmospheric activity center in the Northern Hemisphere and temperature/precipitation in China. The results show that the new method is more reasonable than the Livezey method.