Abstract: It is shown from real data that warm (or cold) surface temperature anomaly usually corresponds with 500hPa height anomalous high (or low)for Jaunary and December. It is reasonable to develop an adjustment model with a proportional coefficient between them. The method calculating this proportional coefficient of surface temperature anomaly and 500hPa height anomaly is proposed in this paper, This coefficient is a function of space,or is a matrix in spherical harmonics. The "future" weather can be specified by the adjustment model in which the matrix is obtained from the "historical" data. The results show whether the specification of 500hPa heights from surface temperature or vice versa has high quality. The correlation coefficient between the observation and specification for monthly mean fields is about 0.8.Furthermore, the longer the time averages and the more the "historical" data, the better the results.