| BACK | |
The IESSG and the Proudman Oceanographic Laboratory (POL) have been working on the problem of monitoring crustal movements at tide gauges in the UK since 1990. Initial projects were based on the use of episodic GPS campaigns. Since 1997, research has concentrated on the use of continuous GPS (CGPS) stations.
A CGPS station has been operational at Sheerness tide gauge since March 1997. This case study provides details on the technical issues that were considered during the establishment of the CGPS station and provides some preliminary results.
2.
Mean sea level measurements at Sheerness tide gauge
Sheerness is located on the Isle of Sheppey in the South-East of England, where the River Medway joins the River Thames estuary about 50 km East of Central London. The tide gauge is currently located on a small pier (or jetty) adjacent to Garrison Point Fort in the Sheerness docks.
The Permanent Service for Mean Sea Level (PSMSL) database has mean sea level (MSL) records for Sheerness dating back to 1832, following the installation of the first self-recording tide gauge (Pugh, 1997; Cartwright, 1999). Operating details are obscure until a Legé gauge was installed in August 1955. The Legé gauge was replaced by a Munro gauge in November 1962, which was moved to a new location in April 1963, before being moved to the current location in May 1973. The Munro gauge was replaced by an Ott gauge in 1980, which was subsequently replaced by an Aanderaa pressure system gauge in July 1983. Finally, an "A Class" bubbler gauge was installed and has been operational since October 1986 (PSMSL, 1994; Woodworth et al, 1999).
Woodworth et al (1999) computed trends in MSL for a selection of UK tide gauges, with more than 15 complete years of MSL and TGBM information in the PSMSL data set. Based on 51 complete years of data over the period from 1901-96, a rise in MSL of 2.14 ± 0.15 mm per year was computed for Sheerness. This is slightly higher than the rise in global sea level of 10 to 20 cm over the past century (IPCC, 1995), which may be due in part to ground subsidence.
3.
Historical geodetic measurements at Sheerness tide
gauge
Historical geodetic measurements at Sheerness tide gauge can be broadly separated into precise levelling surveys carried out by the Ordnance Survey and episodic GPS measurements made by the IESSG and POL since 1990.
The primary tide gauge benchmark (PTGBM) at Sheerness is an Ordnance Survey flush bracket which is set in the stone wall of the Garrison Point Fort about 100 m from the tide gauge. This was first connected to the primary levelling network in 1973, when the tide gauge was moved to its current location, and was last verified by precise levelling (line TG70) in 1984.
The TGBM network (see Figure 1) is effectively formed from six benchmarks, comprising the PTGBM, three benchmarks in the docks (AUX1, AUX2 and AUX3) and two benchmarks inland. AUX2 is located about 400 m, AUX1 about 600 m and AUX 3 about 750 m from the PTGBM. The two inland benchmarks are located about 1.4 km and 4 km to the South-East of the PGTBM on old buildings in the High Street in Sheerness town and on Minster Road outside of the town. All of the additional benchmarks are either Ordnance Survey flush brackets or bolts set into walls. The TGBM network was first connected to the PTGBM in 1973 and last verified by precise levelling (line TG70) in 1984. However, AUX1 had been connected to the two inland benchmarks over the period from 1959 to 1974 (line G152).
The results of repeated precise levelling surveys showed no changes in the heights of AUX1 relative to the inland benchmarks over the period from 1959 to 1974. Following on from this, the results of repeated precise levelling surveys showed a slight change (ie an apparent rise of +1.0 mm over the period from 1976 to 1984) in the heights of the PTGBM and the High St BM 1.4 km away. There were, however, significant changes of +5.7 mm (AUX1), +3.1 mm (AUX2) and +2.3 mm (AUX3) at the benchmarks in the docks over the same period. These results could suggest that the PTGBM and the High St BM had actually subsided relative to the benchmarks in the docks or that the benchmarks in the docks had uplifted relative to the PTGBM and the High St BM.
In terms of decoupling land movements from the MSL measurements made by the tide gauge, these benchmarks are of little use as none of them are actually located on the jetty that supports the tide gauge. However, they do given an indication that localised ground movements may have taken place within the docks and town.
A new benchmark (NBM) was, therefore, added to the TGBM network in 1985. This is located on the jetty a few metres from the stilling well. However, it was only connected to the PTGBM using precise levelling over the period from 1985 to 1987, with no change in height observed.
IESSG/POL episodic GPS and precise levelling
surveys
From 1991 to 1996, a total of nine episodic GPS campaigns were carried out, with two to eight epochal measurements at 16 tide gauges in the UK (Ashkenazi et al, 1998). For the episodic GPS campaigns, a new benchmark (SHE1) was installed at Sheerness in 1990, about 1.4 km from the PTGBM (see Figure 1).
The first episodic GPS measurements were made at SHE1 in September 1991, August 1992 and August 1993, as part of the UKGAUGE I project. Further episodic GPS measurements were then made at SHE1 in April 1995 and September 1996, as part of the UKGAUGE II project. In all campaigns, GPS data were recorded for between 8 and 10 hours per day, for 5 consecutive days. Precise levelling surveys were also carried out by the Ordnance Survey in 1991 and 1993, to connect SHE1 to the PTGBM.
Using a series of episodic GPS campaigns over such a relatively short, 5 year period, it is only really possible to detect vertical land movements of more than a few centimetres. In the case of SHE1, such movements were not observed. Instead, assuming that no vertical land movements had taken place at SHE1 over the period from 1991 to 1996, the standard deviation of height from a time averaged-mean was computed to be ± 10 mm (Ashkenazi et al, 1998).
In terms of decoupling land movements from the MSL measurements made by the tide gauge, in 1991 the Environment Agency installed a new benchmark (EABM) on the jetty at Sheerness (see Figure 1). This is located a few metres from the stilling well and was connected to the PTGBM using precise levelling in 1991 and 1996. Over this period, a change in height of +7 mm was observed, implying that the EABM had uplifted with respect to the PTGBM or that the PTGBM had subsided with respect to the EABM. This change in height was re-observed as + 8 mm by further precise levelling in 1997.
The 1997 precise levelling survey included a connection from the PTGBM to both the EABM and the NBM installed by the Ordnance Survey in 1987. Over the period from 1987 to 1997, a change in height of + 6 mm was observed at the NBM, implying that the NBM had uplifted with respect to the PTGBM or that the PTGBM had subsided with respect to the NBM.
The most recent precise levelling survey was carried out in December 2000, with a connection from the PTGBM to both the EABM and the NBM. These latest results suggest a total change in height of +7 mm for the NBM - PTGBM connection over the 13 year period from 1987 to 2000, and a total change in height of +10 mm for the EABM - PTGBM connection over the 9 year period from 1991 to 2000. Perhaps more importantly, considering the period over which the CGPS station has been operational, the two latest precise levelling surveys suggest that the jetty has uplifted with respect to the PTGBM or that the PTGBM has subsided with respect to the jetty, by 1 to 2 mm over the 3 year period from 1997 to 2000.
These results provide further indications that localised ground movements may have taken place and still be taking place within the docks, even over the short range between the Garrison Point Fort and the jetty that supports the tide gauge.
4.
Siting of the CGPS station at Sheerness tide gauge
As stated in the technical recommendations (www.soest.hawaii.edu/cgps_tg), to obtain the best possible vertical accuracies for a CGPS station it is crucial that the GPS antenna has a clear view of the sky in all directions for elevation angles above 15 degrees. Apart from this, the main technical issue centres around whether it is better to build a CGPS station immediately adjacent to the tide gauge or close to one or more TGBMs which are known to be in ‘stable’ ground.
The historical geodetic measurements discussed in the previous section suggest that localised ground movements may have taken place between the PTGBM on the Garrison Point Fort and the other benchmarks, both in the docks and on the jetty that supports the tide gauge. It is also evident that detailed precise levelling surveys have not been carried out over any significant distance since 1984. In fact, the most recent precise levelling surveys have focused on the connection between the PTGBM and the jetty, which is less than 100 m apart.
For Sheerness tide gauge, therefore, it was a relatively straightforward choice between establishing the CGPS station immediately adjacent to the tide gauge or close to the PTGBM. Bearing in mind the evidence for differential ground movements between the PTGBM and the jetty, locating the CGPS station in either location would necessitate a regular program of precise levelling surveys, but only over a distance of less than 100 m.
The PTGBM is actually an Ordnance Survey flush bracket in the stone wall of the Garrison Point Fort, which is a fairly substantial structure, about 20 m high. To obtain a clear view of the sky and be close to the PTGBM, the GPS antenna would have to be located on top of the Fort, ie 20 m directly above the PTGBM. In this case, further geodetic measurements, such as trigonometric heighting, would then be required in order to establish and monitor the height difference between the CGPS station and the PTGBM.
Instead, the decision was made to establish the CGPS station immediately adjacent to the tide gauge, on the jetty. The tide gauge is actually housed inside a rigid, brick building about 3 m high, which has a single, 150 mm thick, concrete slab for a roof. In this case, the GPS antenna could be located on the roof of the tide gauge building, affording a reasonably clear view of the sky, although not perfect, as things seldom are in docks and harbour environments.
5.
Monumentation and instrumentation at Sheerness tide
gauge
As described in the previous section, the decision was made to establish the CGPS station immediately adjacent to the tide gauge. This would enable the GPS receiver to be housed in the tide gauge building, with mains power, telecommunications and security. Meanwhile, the GPS antenna could be located on a monument fixed to the concrete slab roof of the tide gauge building.
The instrumentation installed at the Sheerness tide gauge followed the guidelines of the IGS in terms of using a dual-frequency code and phase measuring GPS receiver and an IGS standard Dorne Margolin choke ring antenna. The actual instrumentation used comprises a Trimble 4000 SSI GPS receiver and a Trimble Dorne Margolin choke ring antenna (p/n TRM29659.00).
The receiver is connected to a standard telephone line using a US Robotics Sportster modem. This enables remote operation and downloading from the IESSG, using the Trimble R Utilities software, running on a PC connected to a standard telephone line using a US Robotics Sportster modem.
The antenna monument installed at the Sheerness tide gauge is a very simple stainless steel bracket, with a mounting surface that is smaller than the diameter of the base of the antenna, as shown in Figure 2.
The square plate is bolted to the concrete slab roof, close to the South-East corner of the tide gauge building. The monument enables the antenna to be screwed on to the 5/8” BSW thread and oriented to North, before the knurled ring is raised in order to clamp the antenna in position.
The reference mark on the monument is defined as the top of the 5/8” BSW thread. The antenna height to the Antenna Reference Point (ARP) can be easily determined by using a steel ruler to measure the offset from the square plate to the ARP, and then subtracting the known offset from the square plate to the top of the 5/8” BSW thread (ie 158 mm). For example, the measured offset from the square plate is 151 mm, hence, the antenna height to the ARP is 0.151 – 0.158 = -0.007 m.
The vertical ties between the GPS antenna, the tide gauge and the TGBMs can be carried out using precise levelling. There is a lower section of roof (about 2 m high), on which a tripod can be set up, enabling a precise levelling connection to be made between a staff located on the top of the choke ring (TCR) of the antenna and a staff located on the EABM on the jetty. Conventional precise levelling can then be used to connect the EABM on the jetty to the tide gauge contact point inside the building and the PTGBM on the Garrison Point Fort.
Such a precise levelling survey covers a distance of about 100 m and can be easily accomplished within 1 day using modern equipment, such as the Leica NA2000 digital level. In parallel with the CGPS measurements, it is proposed to repeat the precise levelling surveys at annual intervals.
Photographs of the antenna and monument at the Sheerness tide gauge are given as Figure 3 below.
The CGPS station at the Sheerness tide gauge became operational on 26 March 1997. Data are downloaded on a daily basis to the IESSG, in Trimble (.r00) binary format. The 24 hour files are then converted to RINEX format and archived at the IESSG as part of the British Isles GPS Facility (Dodson et al, 2000).
The same equipment (receiver, antenna, cables, etc) have been operational at the Sheerness tide gauge since installation. Firmware upgrades have been made to the receiver, but no hardware changes have been made and the antenna cable has remained connected.
Signal-to-noise ratio (SNR) values are not routinely output in the RINEX format data files. However, at six month intervals, samples of binary data are re-run through the RINEX conversion program and SNR values are output. These are then checked for any changes in signal quality, which may indicate a change in the performance of the CGPS station. To date, no degradation has been found.
6.
Geodetic data processing and preliminary results for Sheerness tide
gauge
The data from the CGPS station at the Sheerness tide gauge are processed by the IESSG on a daily basis using their in-house developed GPS Analysis Software (Stewart et al, 1997). In this processing, the data from several CGPS stations in the UK are combined with data from IGS stations in Europe, as a series of ‘loosely constrained daily GPS network solutions’. In these solutions, the IGS precise ephemeris is held fixed and the coordinates for Kootwijk are tightly constrained to the same ITRF realisation and observation epoch as the IGS precise ephemeris.
For all of the loosely constrained daily GPS network solutions, the GPS data are processed using the L1/L2 ‘ionospherically free’ double difference observable, with integer ambiguities free, and systematic error models for solid Earth tides, ocean tide loading and antenna phase centre variations. The ‘dry’ zenith tropospheric delay is modelled using Saastamoinen and the ‘wet’ zenith tropospheric delay is estimated as a random walk process.
Following the GPS data processing, coordinate time series are formed for each of the CGPS stations. In order to form coordinate time series with no discontinuities, the vectors and associated covariance information output from each loosely constrained daily GPS network solution are transformed to a common reference frame. This is currently carried out by tightly constraining the European IGS stations of Onsala, Kootwijk, Wettzell and Villafranca to their ITRF97 coordinates, motioned to the observation epoch (Boucher et al, 1999).
Figure 4 shows the daily height time series obtained for the CGPS station at the Sheerness tide gauge. This is represented in the ITRF97, based on daily coordinate estimates computed in the ITRF94, ITRF96 and ITRF97.
The estimation of vertical station velocities and their uncertainties, and the effects of periodic variations on CGPS height time series are current topics of research at the IESSG (Bingley et al, 2000) and many other research centres. At this stage, we can only remark about the possible long-term trends in the height time series, which may suggest that the CGPS station has experienced some uplift from March 1997 to December 1999 followed by subsidence.
Emerging estimates of vertical station velocities will be used in conjunction with the precise levelling links between the CGPS station and the PTGBM to identify if and where uplift and subsidence are occurring. Reliable estimates of vertical station velocities will then be compared with modelled estimates of crustal movements due to glacio-isostacy (Lambeck,1993a; 1993b; Lambeck and Johnston, 1995) and geological estimates of Holocene crustal movements (Shennan, 1989; Woodworth et al, 1999), which both suggest that the South-East of England is subsiding by about 0.5 to 1.5 mm per year.
Acknowledgements. Research carried out on the use of CGPS at tide gauges in the UK has been funded by the Ministry of Agriculture Fisheries and Food (MAFF), through the long term commission with POL, the Environment Agency and the Natural Environment Research Council. I would like to thank Simon Booth, Alan Dodson, Nigel Penna, Norman Teferle and Samantha Waugh from the IESSG, who have all contributed to the work described in this case study. I would also like to thank Trevor Baker and Philip Woodworth from POL for their comments and suggestions.
Ashkenazi, V, Bingley, R M, Dodson, A H, Penna, N T, and Baker, T F, 1998, Separating crustal movements and sea level changes using GPS at tide gauges in the UK, Book of Extended Abstracts of the WEGENER 98 Conference (Second, Revised Edition), Krokkleiva, Norway, July 1998, pp 102-105.
Bingley, R M, Dodson, A H, Penna, N T, Teferle, F N, Booth, S J and Baker, T F, 2000, Using a combination of continuous and episodic GPS data to separate crustal movements and sea level changes at tide gauges in the UK, Book of Extended Abstracts of the WEGENER 2000 Conference, San Fernando, Spain, September 2000, 4 pp.
Boucher, C, Altamimi, Z, and Sillard, P, 1999, The 1997 International Terrestrial Reference Frame (ITRF97), International Earth Rotation Service (IERS), Technical Note 27, Observatoire de Paris.
Cartwright, D E, 1999, Tides – a scientific history, Cambridge University Press, Cambridge.
Dodson, A H, Bingley, R M, Penna, N T, and Aquino, M H O, 2000, .A national network of continuously operating GPS receivers for the UK, Geodesy Beyond 2000, The Challenges of the First Decade, Edited by Schwarz, International Association of Geodesy Symposia, Vol 121, Springer-Verlag, 2000, ISSN 0939-9585, ISBN 3-540-67002-5, pp 367-372.
Lambeck, K, 1993a, Glacial rebound of the British Isles – I. Preliminary model results, Geophysical Journal International, 115(3), pp 941-959.
Lambeck, K, 1993b, Glacial rebound of the British Isles – II. A high-resolution, high-precision model, Geophysical Journal International, 115(3), pp 960-990.
Lambeck, K, and Johnston, P, 1995, Land subsidence and sea-level change: contributions from the melting of the last great ice sheets and the isostatic adjustment of the Earth, Proceedings of the Fifth International Symposium on Land Subsidence, The Hague, The Netherlands, October 1995, pp 3-18.
PSMSL, 1994, Documentation added to file pub/prints.rlr on 20 September 1994, http://www.pol.ac.uk/psmsl.
Pugh, D T, 1987, Tides, surges and mean sea-level: a handbook for engineers and scientists, Wiley, Chichester.
Shennan, I, 1989, Holocene crustal movements and sea-level changes in
Great Britain, Journal of Quaternary
Science, 4, pp 77-89.
Stewart, M P, Ffoulkes-Jones, G H, Chen, W, Ochieng, W Y, Shardlow, P J, and Penna, N T, 1997, GPS Analysis Software (GAS) Version 2.4 User Manual, IESSG Publication, UK.
Woodworth, P L, Tsimplis, M N, Flather, R A, and Shennan, I, 1999, A review of the trends observed in British Isles mean sea level data measured by tide gauges, Geophysical Journal International, Vol. 136, pp 651-670.