The D and O18 contents of rain water are known to vary considerably with time and location. Dansgaard found a good correlation of the mean O18 content of rain with mean air temperature, which causes a latitudinal as well as a seasonal variation of O18 and a similar one for D. Under dry climatic conditions, as in South Africa, the D content appears to be strongly influenced by re‐evaporation of the falling raindrops. This has the effect of obscuring the normal seasonal variations. The evaporation obviously takes place under kinetic conditions, thus increasing the separation factors especially for the oxygen isotopes. A detailed study of a thunderstorm in Heidelberg shows that here too evaporation causes an enrichment in D.

Water evaporating into air not saturated with water vapour shows a stronger enrichment of D and O‐18 than could be ascribed to the simple equilibrium isotopic fractionation. Kinetic effects are responsible for the increase in isotopic fractionation between liquid and gaseous phase. The so‐called kinetic fractionation factor is calculated by means of a simple model. Comparison with the experimentally obtained fractionation factors proves the assumption that the kinetic fractionation is produced by the different diffusion velocities of H2O‐16, H2O‐18 and HDO‐16 in the surrounding turbulent air. The consequences of the kinetic separation for the evaporation of the oceans are briefly discussed. Those for the evaporation of a limited water reservoir in isotopic exchange with the water vapour of the air are more extensively investigated. In the latter case calculation and experiments show that during evaporation the D‐ and O‐18‐contents approach stationary values as the water mass decreases.

This paper presents and discusses electric field and plasma density data obtained by the GEOS 2 double probe experiment on March 22, 1979. The observation period encompasses a sudden storm commencement and two consecutive substorms. The E field data are compared with magnetic field recordings both at GEOS and on the ground. The influence of the solar wind on magnetospheric dynamics is studied with the help of additional data from IMP 8 and ISEE 3. The E field data, which were obtained when GEOS was in the dayside (front) of the magnetosphere, are grouped in transients, pulsations and underlying dc electric fields. We find that solar wind discontinuities cause strong electric field transients in the dayside magnetosphere. These transients are followed by Pc 4 oscillations of considerable amplitude. The data also show the existence of dc electric fields during buildup, onset and expansion phase of the two substorms. The combined observations by GEOS 2, ISEE 3, IMP 8 and ground‐based magnetometers provide information on propagation of transients in the magnetosphere, on the nature of large scale oscillations and on the overall convection pattern during periods when the interplanetary magnetic field has a negative Bz component.

Data from the spherical double probe electric field experiment on ISEE 1 have been used to study a number of plasma sheet/lobe boundary crossings during intervals selected for the Coordinated Data Analysis Workshop 6 on March 22 and 31, 1979. These crossings took place during periods of substorm occurrence. They could be identified by keV plasma measurements and by using the electric field probes as a reference for measurements of the spacecraft potential. Typical spacecraft potentials in the plasma sheet are +10 V to +20 V. Because of decreased plasma density in the lobes the satellite potential increases here to +20 V to +30 V. Strong electric fields, with a dominant dawn‐to‐dusk component, are observed throughout the boundary layer outside the plasma sheet both for contracting and expanding motions of the plasma sheet and for different magnetic field directions. Characteristic amplitudes and durations are 5–10 mV m−1 and 5–15 min. The corresponding E × B vectors are always toward the plasma sheet. The spacecraft were situated both north and south of the plasma sheet edge, in such boundary layers, and in two or three cases the spacecraft were most likely on the earthward side of an X line. These latter cases coincided with the maximum of the expansion phase of substorms. Prior to these substorms, during plasma sheet thinning, onsets of sunward convection were observed by an electric field experiment on GEOS 2 in the geostationary orbit.

No abstract