A Secondary Source of Water Vapour in the Upper Atmosphere

Abstract

THE early pioneering work of Dobson, Brewer and Cwilong with frost point hygrometers indicated that over South England the water-vapour content of the stratosphere between 12 and 15 km is extremely small in all seasons, being about 2 mg/kg (frost point temperature being about − 80°). Subsequent measurements of humidity between the same levels over a geographical region extending from Norway to equatorial Africa revealed the interesting fact that in all seasons and in most of the locations the moisture content of the lower stratosphere is very small. On the basis of such observations, which led to the concept of a uniformly dry stratosphere, Dobson et al.¹ postulated the existence of a meridional circulation in the stratosphere involving the poleward movement of very cold saturated air near the equatorial tropopause and its descent over high latitudes. This theory has been amplified further by Ramanathan² and Murgatroyd and Singleton³. On the other hand, a large number of observations made over the North American continent and Greenland during 1959 and 1960 using automatically registering frost point hygrometers indicated a very wide range of frost point temperatures from − 67° to − 80° C (refs. 4, 5). Similar observations made over Japan indicate that on many occasions the water-vapour mixing ratio increases with height in the stratosphere⁶. Rohrbough⁷, using a balloon-borne sampling system, has recently reported high humidity mixing ratio of 320 mg/kg at a height of 30 km over the United States, this height being a record for the determination of humidity. These recent observational results, which indicate a rather complex distribution of water-vapour, suggest that the earlier concept of a uniformly dry stratosphere needs re-examination.