Abstract
Trend‐curves of thirty‐year means are discussed (a) in general terms and then applied specifically to (b) temperature, (c) pressure, (d) rainfall, (e) wind, (f) temperature gradient and (g) temperature amplitude.
Correlations between the departures of the different elements suggest that we can consider the various anomalies as depending on those of pressure and wind. This is notably true of rainfall departures which can be regarded as the sum of two terms, the “area” term depending on the pressure anomaly and the “local” term depending on the vector wind anomaly. High pressure is associated not only with consistently dry years but also with cold years. Variations of wind or pressure gradient are particularly felt as changes of frequency of the prevailing wind. The southwesterly component is of course linked with the “zonal index” and the circulation of the wind systems of the earth. The increase of this component gives more maritime conditions and, except to the lee of hills or mountains, wetter weather. The recent increase explains the main differences between nineteenth and twentieth century climates but not the phases of the transition. These phases necessitate first the concept of a ring of high pressure which, by some south‐steering process, left the Arctic Circle about 1880 and reached the Tropic of Cancer about 1915. Preceding and following this ring are rings of high rainfall, low pressure and high temperature‐gradient that can be loosely regarded as frontal belts. At any one place therefore, the following sequence of five phases should occur : the first front (Frontal‐A stage), the easterly or cold period (T‐min), the pressure maximum (P‐max), the westerly period culminating in the second front (Frontal‐B stage).
In the British Isles the five phases explain the climatic fluctuations from 1875 to 1925 and much of the same sequence can be followed in Scandinavia from 1865 to 1915, Italy from 1885 to 1935 and in the United States from 1895 to 1945. Temperature data suggest that this equatorward movement with time may be of more general application, and that the preceding West European climatic fluctuations can also be interpreted as anomaly patterns which spread from the north to the south.
Pressure anomalies are normally greater over oceans than continents. The south‐steering of anomaly patterns necessitates, therefore, a backing of the prevailing wind in NW. Europe after the pressure maximum (P‐max) of about 1895. Such a change is confirmed by the wind records themselves.
An effect of persistent wind on ocean temperatures and therefore pressures might well account for the south‐steering, but there is as yet no evidence that delayed action effects of this kind occur.