Abstract
In preference to the use of persistent height anomalies for determining the presence of blocking, new criteria were developed which take the observed flow conditions into account. These criteria were applied daily at 500 mb, using filtered Northern Hemisphere height maps retaining zonal waves 0–5, beginning in July 1981. Blocking statistics for the 3-yr period through June 1984 are in remarkably close agreement with the results from longer-period published data sets involving the use of a split-jet flow as a principal criterion. Thus this data set, albeit short, is deemed appropriate for investigating the long-wave properties of blocks.
During July 1981–June 1984, 47 blocks were observed; the long-wave composition of blocks lasting at least 10 days (24 cases) is examined in detail. Contrary to Austin's (1980) hypothesis that blocking should arise from the constructive interference of stationary long waves with enhanced amplitude but normal phase, it is found that traveling waves (predominantly retrogressive) played a significant role in 20 of the 24 cases. The long waves making up these blocks, in the zonal wavenumber domain 1–4, moreover had a phase behavior consistent with the wavenumber dependence in the Rossby wave dispersion equation.
A dramatic increase in Northern Hemisphere blocking activity within the sample, from 1982–83 to 1983–84, is shown to be due mainly to increases in the Pacific–North American (PNA) and North Soviet Union (NSU) sectors. A time–longitude diagram depicts the shifting of blocking activity from the main region of blocking, the Atlantic–west European area, to the PNA and NSU sectors. These regional shifts were substantially associated with the longitudinal phase pattern of travelling waves 1 in high latitudes.
