S U M M A R YGravity Green's functions for a column load of a model atmosphere on a spherical, elastic Earth are presented and they are used to evaluate the contribution of global atmospheric pressure variations to local gravity. The Green's functions are found to be relatively insensitive to the details of the model atmosphere, but they are dependent on the temperature at the base of the column, and on the relative height difference between the base of the column and the gravity station. The total signal that global pressure systems contribute to gravity is about 30 pgal, of which about 90 per cent is produced by the atmosphere within 50 km of the gravity station. A zone between 50 and 1000 km from the gravity station contributes a couple of pgal, as does the remainder of the globe. This pattern, the coherence scale of pressure fluctuations, the time and spatial scales appropriate to the hydrostatic approximation, and the distance of the gravity station from the oceans, suggest a division of the globe into local, regional, and global zones. Data requirements, processing details, and the reliability of the computed signal are different in each zone. The local zone is within about 50 km of the gravity station. Within this zone pressure changes rapidly in time, but is spatially coherent, so that hourly observations of pressure and temperature at the gravity site alone are sufficient to compute an accurate correction, except when a front is passing through. The regional zone extends from the edge of the local zone to between several hundred and a thousand kilometres. The signal from this zone is small and is only weakly coherent with the signal from the central zone, so that a rather sparse array of hourly samples of pressure and temperature are required. The gravity signal from the global zone can reach about a pgal. It varies on a time-scale of days, and is influenced by the response of the oceans to pressure variations. Previously reported observations that the admittance between local pressure and gravity residuals depends on epoch, frequency, or site, are most probably due to incorrect modelling. A proper local, regional, temperature, and global correction can adequately account for the gravity signal from the atmosphere to within a few tens of ngal in the diurnal band, and about 100ngal in the days to seasonal band, except during extreme weather conditions. The application of the local correction lowers the power spectral density of the gravity residuals in every band from seasonal to hourly. The regional, global, and temperature corrections lower the residual noise in the seasonal and synoptic bands, but are not consistently effective at periods less than about half a day.

A global network of superconducting gravimeters (SGs) is compiling significant data for a range of important studies spanning a number of disciplines concerned with the Earth's gravity, tides, environment, and geodetics. Among phenomena being looked at are seismic normal modes, the Slichter triplet, tidal gravity, ocean tidal loading, core nutations, and core modes. Hydrologists and volcanologists also may benefit from SG data. The network was set up by the Global Geodynamics Project (GGP),an international program of observations of temporal variations in the Earth's gravity field. Observations began in 1997 and will continue until 2003. Eighteen SGs currently are in operation in the network (see Figures 1 and 2).

No abstract

Induced-polarization (IP) responses arising from activation overvoltage are investigated, with the goal of relating established electrochemical properties and processes to IP time constants. In activation overvoltage, an electrical double layer with a well-defined areal capacitance forms at the mineral-electrolyte interface. This study shows that the known capacitance of the interface is compatible with observed initial time constants for the decay of electrode polarization. Furthermore, because the electrical double-layer capacitance is determined largely by the electrolyte and not the mineral, the conclusion is that, although the polarization is caused by the presence of the mineral, the time constants of the decay are determined primarily by the electrolyte and not the mineral. The dependence of the capacitance on the sign and magnitude of the potential implies that the effective decay time constant will be small at the start of depolarization and will progress to larger values as the decay proceeds.

S U M M A R YThe Earth's zonal response coefficient K is estimated from the tidal signals in the observed length-of-day (LOD) data. Its magnitude and phase are functionals of the Earth's internal structure and dynamics. In this paper, an analysis of 13 years of precise LOD data (1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992) reveals strong signals for nine zonal tidal groups ranging from 5 to 35 days in period. Numerical estimates of K for 27 major tides are thus obtained, 11 among which are considered sufficiently high in signal-to-noise ratio to provide meaningful geophysical constraints on the Earth's rotational dynamics. The results favour a K magnitude close to, but somewhat smaller than, 0.315, which is the theoretical value for an elastic mantle completely decoupled from the fluid core plus equilibrium oceans. A small amount of dispersion is also detectable, where shorter periods tend to have lower K magnitude and larger phase lag. Our K magnitude estimates are consistent with two recently published non-equilibrium ocean-tide models and an anelastic response in the mantle, although an equilibrium response in the ocean and a purely elastic response in the mantle is not disallowed. Phase lags of a few degrees are required by both ocean-tide models, and by our data.