On the definition and persistence of blocking

In this paper, we have investigated under what conditions positive geopotential height anomalies can be interpreted as blocking structures. Furthermore, we have studied the persistence of blocking. The data used here are the observed daily 500 hPa geopotential heights for 10 winter seasons from 1982/83 to 1991/92. The objective blocking criteria of Dole (1978) have been compared with the subjective blocking criteria established by Rex (1950a, b). Our study indicates that if we want to relate positive geopotential height anomalies to blocking as defined by Rex, not only the amplitude, but also the latitudinal position of the anomalies is important. Positive anomalies centered at about 45°N represent a northward extension of the subtropical anticyclone. Large positive anomalies centered at about 60°N are accompanied with negative anomalies at low latitudes and blocking anticyclones centered at about 50°N. We found that blocking and strong zonal flow have the same anomaly pattern but with opposite sign. The day-to-day changes in geopotential height associated with blocking are computed and the relevance of these day-to-day changes to the persistence of the blocking pattern is examined. We find that the day-to-day changes are smaller than normal over the blocking area, but larger than normal north of it. The anomalous day-to-day changes are mainly caused by the northward shift of the jet stream and the corresponding shift in the track of the transient weather disturbances. The strength of the day-to-day changes depends very much on the exact latitudinal location where the changes are computed. These results indicate that the day-to-day changes in geopotential height do not contain information about the persistence of the blocking pattern itself, but merely about the shift of the transient eddies caused by the blocking pattern. Hence, we can not simply interpret the persistence of blocking in terms of the day-to-day changes in positive anomalies. The persistence of blocking and strong zonal flow has been examined. As blocking and strong zonal flow are associated with approximately the same anomaly pattern but with opposite sign, the persistence of blocking and strong zonal flow can be expressed in terms of the duration of the corresponding anomaly patterns. We have shown that the average duration of the blocking and strong zonal flow patterns is approximately the same. This leads us to conclude that although the daily changes in geopotential height are small near the center of the blocking high, the blocking regime itself is not more persistent than the strong zonal flow regime. The persistence of blocking and strong zonal flow has also been compared with a first-order Markov process. Generally speaking, we find no significant differences in persistence between the blocking and strong zonal flow regimes and the first-order Markov process. This leads us to conclude that blocking is not more persistent than other atmospheric anomalies of the same geographical scale. This is in agreement with Lindzen (1986).