|
Absolute Gravity Case Study, Newlyn Tide Gauge, UK Simon D. P. Williams, Trevor F. Baker, Graham Jeffries May 30, 2001 1 Introduction Proudman Oceanographic Laboratory (POL 2 The Absolute Gravimeter The FG5 absolute gravimeter is the latest in a series of technical developments made over many years by the USA National Institute of Standards and Technology and the University of Colorado [Niebauer et al., 1995]. It is now produced commercially by Micro-g Solutions Inc., Colorado, USA. The principal of the absolute gravimeter is straightforward in that it measures the acceleration of a mass in free fall in a vacuum. However, in order to achieve an accuracy of 2 parts in a billion of gravity (i. e. 20 nms-2, 2 microgals) several technical advances have been made so that various possible sources of systematic errors are reduced to well below this level. The falling mass is a corner-cube retroreflector and its position is measured using a laser interferometer. The light path to the corner-cube forms one arm of the interferometer and a corner-cube mounted on an active long period seismometer (Super Spring), so as to act an inertial reference mass, forms the other arm of the interferometer.
Figure 1: Location map of the absolute gravity site and its relationship to the nearby tide gauge and CGPS site at Newlyn, Cornwall. The interference fringes between the 2 light
beams are counted and timed and the distance-time pairs collected during the 20cm
drop are used in the equation of motion to solve for the acceleration due to
gravity 'g'. The whole system is automatically controlled by computer and
typically 200 drops i.e. 200 independent determinations of 'g' (one set), are
made per hour. The key to the accuracy of the FG5 absolute gravimeter is that
the measurements are directly related to international standards of length
and time. The instrument uses a newly developed iodine-stabilised laser,
which is the international length standard. A rubidium frequency reference is
used for timing and this can be calibrated against a caesium standard. For
further details of the instrument and the technical developments see Niebauer
et al. [1995] (and http:// www. microgsolutions.com)
. Figure 2: The location of the absolute gravity measurements inside St Pol de Leon Church, Paul, Cornwall. 3 Site Selection Before making any absolute gravity measurements at Newlyn a reconnaissance visit was made in order to choose the best available site. The Newlyn absolute gravity site was established inside the church of St Pol de Leon, in Paul, 1.5 km from the tide gauge (Figure 1). The church is mainly 15th Century with Norman foundations and is constructed of rough granite. The foundations of the church are on solid granite bedrock which is important for stability and for being representative of the vertical crustal movements of the area. In addition, a non-porous bedrock with little or no soil cover should help to reduce any hydrological-induced signal. Being an historic church it is unlikely that it will be severely altered/ extended/ demolished in the near future. In addition, the risk of large scale construction, for instance hotels etc, nearby is very low. The church also provides easy access (it is open all day to the public), power, thermal stability and limited security (locked overnight). Some negative aspects of sitting the AG in the church include the unlimited access (visitors can walk right up to the machine) and the possible interruption due to religious services. We chose the largest slate flagstone in the northeast corner of the church (Figure 2) on which to place the gravimeter. Having the whole instrument on one flagstone would hopefully reduce any possible noise due to vibrations set up in the instrument-floor system when the mass is dropped (site dependent noise).
Figure 3: The influence of loading and attraction corrections on the absolute gravity observations at Newlyn. (A) The set means (200 drops per set). The error bars represent the 1 Ã scatter of the drop-to-drop residuals about the set mean. (B) is the same data after correction for the body tide and (C) is the data if the ocean loading and attraction corrections are applied. Figure reproduced from Bos [2000]. A closer AG site, for example, in the tide gauge hut or lighthouse, could possibly have been chosen. However, these sites do not allow access to granite and have much higher microseismic noise. The Newlyn area experiences one of the largest ocean tide loading effects in the world (Figure 3) [Baker, 1980]. To obtain realistic values for the ocean loading and attraction (from global ocean tide models) for stations above sea-level and near to the coast a very high resolution coastline is required. Of course, the mean ocean loading effect is negligible given a sufficient observation period (several days) or the ocean loading effect could be estimated using the data. The Newlyn AG site is also part of the British Precise Gravity Network (BPGN), which consists of 65 gravity sites covering most of Britain, that have been measured by the University of Edinburgh using LaCoste and Romberg relative spring gravimeters [Charles and Hipkin, 1993]. In addition the relative spring gravimeters have been used to determine the gravity gradient at Newlyn [Hopewell, 1999]. 4 Methodology Since the absolute gravimeter produces typically 200 values of gravity per hour, it is common practice to measure gravity at a site by measuring for 1 day or even just for a few hours. The procedure developed for our work is to measure for at least 3 days, with the gravimeter being carefully set-up again at the start of each day. This not only produced significantly more data for a given field trip, but also allows a comparison of the standard deviations of the hourly means each day with the variability from day to day. This enables the assessment of the noise in the measurements at a particular site and epoch. Before and after each field visit, measurements are made at the gravity site in the sub-basement of Bidston Observatory (POL), in order to ensure that the absolute gravimeter gives consistent values. The long series of measurements at POL had already clearly shown that one of the main sources of noise in the absolute gravity measurements is due to microseisms, which typically have periods from 5 to 7 seconds. These are generated by ocean waves and are therefore important at coastal sites. The Super Spring has a period of about 60 seconds, in order to reduce the effects of high frequency noise. However, on very windy days the energy in the microseismic band increases significantly and oscillations of the reference mass give an increase in the noise of the gravity measurements. The extended visits to each site are therefore also valuable, since this allows the possibility of avoiding days with very high microseisms.
Figure 4: Absolute Gravity time series for Newlyn. Each point is a daily mean value. Error bars are one sigma (68%). Also shown is the weighted least-squares fit to the data. An important part of the methodology is to ensure that the absolute gravimeter continues to give results with an accuracy equivalent to the highest international standards. The POL absolute gravimeter has therefore been regularly intercompared with other absolute gravimeters. Whenever the instrument is returned to the manufacturer, Micro-g Solutions USA, for service, the opportunity is taken to make measurements at the USA fundamental gravity site at Table Mountain in Colorado and to compare the results with measurements made with other FG5 gravimeters. Inter-comparison experiments have also been made with other absolute gravimeters at European sites in Germany, Belgium, Italy and France. The latter measurements were made at the Bureau International des Poids et Mesures (BIPM) near Paris during the inter-comparison experiment between all available absolute gravimeters, which is organised every 4 years. In these various intercomparisons FG5-103 was verified to be in agreement with other FG5 absolute gravimeters at the 2 microgal (20 nms-2) level, which is the specified accuracy of the instrument. 5 Results The absolute gravity results at Newlyn are shown in figure 4. It can be seen that Newlyn is particularly stable and has an overall change of gravity of less than 1 microgal from 1995 to 2000. The results at Newlyn and two other sites, Lerwick and Aberdeen are discussed in Williams et al., [2001]. 6 References Baker, T. F., Tidal gravity in Britain : tidal loading and the spatial distribution of the marine tide, Geophys. J. R. astr. Soc., 62, 249-267, 1980. Bos, M. S., Ocean tide loading using improved ocean tide models, Ph. D. thesis, 203 pp., University of Liverpool, November 2000. Charles, K., and R. Hipkin, British precise gravity net 1993. Technical report, Department of Geology and Geophysics, University of Edinburgh, Edinburgh, 1993. Hopewell, H., Environmental and instrumental effects on high precision gravimetry - A case Study in Britain, Ph. D. thesis, 194 pp. University of Edinburgh, December 1999. Niebauer, T. M., G. S. Sasegawa, J. E. Faller, R. Hilt, and F. Klopping, A new generation of absolute gravimeters, Metrologia, 32, 159-180, 1995. Williams, S. D. P., T. F. Baker and G. Jeffries, Absolute gravity measurements at UK tide gauges, Geophys. Res. Lett., 28, 2317-2320, 2001.
|
