Microgal gravimetry's unique ability to sense mass anomalies and movements beneath the surface of the Earth makes it a promising technology for science and engineering, particularly when it is used with Global Positioning System geometrical measurements. It is rapidly becoming the method of choice for measuring postglacial rebound, monitoring active volcanoes, modeling subsurface cavities, detecting changes in aquifers, and studying changes in the Arctic and Antarctic ice masses. More than 50 scientists and students from 14 nations recently met at a Chapman Conference to discuss microgal gravimetry and its applications.
There are two distinct approaches to measuring the temporal variations of gravity at any given station: using an absolute gravimeter to make a series of measurements at selected epochs, and using a superconducting gravimeter installed and operated continuously at a station. Each approach has advantages and disadvantages. An absolute gravimeter can monitor gravity variations at many stations with a single instrument and determine the differences in gravity among stations. However, the resolution of an absolute meter is perhaps an order of magnitude worse than the superconducting gravimeter, and the sparse time series of measurements make it more difficult to identify the sources of observed gravity changes such as relocation of atmospheric mass, groundwater changes, and vertical crustal motion. Superconducting gravimeters do not measure absolute gravity, but they measure changes in gravity with high resolution and precision. They are limited to single station monitoring and have long-term drifts and tares that must be removed. The best measurements of microgal gravity changes are obtained by the complementary use of both absolute and superconducting gravimetry.
Researchers in several nations continue to develop one-of-a-kind absolute gravimeters, but the FG5 absolute gravimeter currently is the only instrument commercially manufactured and marketed internationally, and 18 FG5 absolute gravimeters are now operational. The FG5 has become the de facto "standard," which defines the current state-of-the-art for the measurement of absolute gravity, against which other instruments are compared.
Nearly all of the FG5 meters were tested at the NOAA Table Mountain Gravity Observatory (TMGO), near Boulder, Colo. Comparing gravity measurements from the different meters at TMGO is perhaps the best way to estimate the precision and accuracy of the current generation of instruments, although there are significant data from comparisons of two or more instruments at other locations. Analysis showed that the precision of the FG5 meter is better than 1 microgal. Results from recent experiments using an FG5 meter to measure the gravitational constant G were also consistent with an instrumental precision better than 1 microgal.
The absolute accuracy of the FG5 meter is still contentious within the community. Some instrument intercomparisons show agreement at the 1-microgal level, while others show disagreements of 5 microgals or larger. Two researchers reported troublesome results from simultaneous measurements with two FG5 meters at Wettzell, Germany. Both meters appeared to be operating properly, but the measured gravity values differed by 6 microgals. After the conference, both instruments were shipped to Colorado for evaluation by the manufacturer and testing at TMGO. Some mechanical wear associated with the extensive use of the instruments since they were last serviced was found. The tests also suggested that the gravity values may have been contaminated by the inclusion of data collected during the catch phase of observation. Participants agreed that frequent intercomparisons are needed to assess the accuracy of gravimeters and to provide an early warning system for instrument problems. They also agreed that intercomparison tests should include computer processing programs when possible. Finally, there was an appeal for researchers to use standard computer programs and keep their software current.
A. Lambert presented a definition of field-achievable, site-dependent accuracy. Field measurements of absolute gravity are degraded by uncertainties in the local vertical gravity gradient measurement, tidal modeling, groundwater variations, and site dependent path length perturbations. More pessimistic estimates of the errors resulting from these sources, which maybe encountered at certain sites, could degrade the measurement accuracy to 5 microgals or worse. Assessing the accuracy of absolute gravity measurements under these "real-world"conditions is crucial in determining accurate uncertainty limits on the rate of change of gravity at stations, especially those sets of measurements that incorporate data from different instruments.
The time-dependent stability of absolute gravity measurements at some sites is difficult to assess. Certain sites clearly have significant time variation, primarily due to groundwater changes. At these stations, superconducting instruments could play a major role in monitoring short-term time variations.
For the most part, barometric pressure signals are easily dealt with using point pressure measurements. On the other hand, the ocean loading signal is still problematic, but satellite altimetry data show promise in improving the models. Given the current instrument performance, one must account for all of these effects, and it is likely that the temporal changes in these effects are exactly what is of interest to other researchers, such as hydrologists.
Conference participants discussed the results of comparisons between superconducting gravity time series and Global Positioning System geometrical measurements of vertical crustal motion. Gravity measurements, converted to height equivalent signals, actually display superior resolution to the GPS measurements. However, group discussions focused on the complementary nature of the microgal gravity and GPS measurements. Participants concluded that every effort should be made to combine microgal gravimetry with other geodetic measurements, especially GPS measurements of crustal motion, both vertical and horizontal.
A series of absolute gravity measurements in the area of maximum post glacial rebound in Canada now shows a convincing trend line, which agrees closely with theoretical models. In fact, the scatter of the observations about the trend line is extraordinarily small, so much so that A. Lambert commented that the results probably display a certain degree of good "luck." Sea level control measurements near tide gauges show good agreement over year-long time scales, which indicates that modeling and instrument performance are adequate for providing vertical control.
Absolute gravity measurements in Mammoth Lakes, Calif., are detecting vertical uplift due to resurgence of the magma chamber, while in Italy they are being used to model subsurface cavities. Microgal gravity measurements are also used to monitor several active volcanoes around the world, including volcanoes on Reunion Island, and in Indonesia, Nicaragua, and Guadeloupe. An ongoing project is using absolute gravity, in conjunction with GPS, to measure variations in the mass of the Greenland ice sheet. The combination of the two methods can, in principle, be used to separate the effects of the viscous flow of mass within the mantle associated with long-term postglacial rebound and the elastic deformation associated with contemporary changes in the ice sheet.
Significant improvements in gravimeter instrumentation are already being tested and are expected to be incorporated into operational instruments in the near future. The availability of fast, inexpensive digital recording systems (both time interval and analog/digital conversion) is already beginning to have an impact. Faster sampling rates allow better noise rejection through averaging. Initial tests show improved precision in single measurements by a factor of 2 or more. However, long period seismic noise is likely the limiting factor for the statistical variance between individual measurements. New methods to measure and correct for seismic accelerations are also being explored by a number of researchers. Also discussed were early results from tests of a seismic isolation platform, which is expected to reduce noise from natural and anthropogenic sources, noise that can be troublesome in active seismic zones and at coastal sites.
Other approaches are being tried to improve the performance of absolute gravimeters. Early results of tests with a frequency doubled YAG laser may provide a more stable light source for the interferometer, and more powerful processing techniques are available for extracting the gravity signal from noise and estimating the accuracy. Smaller absolute systems are also under development. These systems are designed to be portable, but with reduced accuracy and precision. Such devices would be useful for exploring and monitoring oil fields.
Efforts to improve superconducting gravimeters through the use of dual sensors were also discussed. The two sensors allow better detection of tares, and thus will improve the reliability of the gravity time series. Certain setup techniques also seem to point to the way to further reduce instrumental drift, but this is still under investigation.
The microgal gravity community still needs to resolve several problems before the full potential of gravity measurements can be realized. More robust methods for modeling and removing the effects of perturbing signals when computing the trajectory of the mass are needed; the effects of nonlinear gravity gradients need to be accounted for; better procedures are needed for monitoring and validating instrumental performance under field conditions; and data processing standards need to be developed.
More than 20 oral presentations and 10 poster papers were presented during the conference, and the conveners plan to publish proceedings by the end of 1997. Questions concerning the proceedings should be directed to Bernd Richter, e-mail: richter@no3.ifag.de
The Chapman Conference on Microgal Gravimetry: Instruments, Observations and Applications" was held from March 3 to 6, 1997, in St. Augustine, Fla.