This Symposium is dedicated to the memory of

mohammed el-sabh

SEA-LEVEL CHANGES AND VERTICAL GROUND MOVEMENTS

(IAG, IAPSO, IASPEI, IAHS, ILP)

Location: Law Building 115 LR2

Location of Posters: Old Gym

Wednesday 21 July AM

Presiding Chair: Phil Woodworth

JSG11-A3 Introduction 0830

Phil Woodworth

JSG11/E/01-A3 0850

SEA SURFACE HEIGHT VARIABILITY DERIVED FROM ERS-2 ALTIMETER DATA NEAR CHINA SEA

HANJIANG WEN, H.Suenkel, Institute of Theoretical Geodesy, Steyrergasse 30, TU-Graz A-8010 Graz, Austria

A small area near China sea was used to study the sea surface height variability using one-year ERS-2 altimeter data. A local cross-over adjustment was carried out to reduce the residual radial orbit error, and the influence of the earth's geopotential model was also studied. Collinear analysis was then used to derive the RMS sea surface height variability, and seasonal variation of the sea surface was also compared.

JSG11/E/04-A3 0910

GLOBAL SEA LEVEL CHANGES FROM SATELLITE ALTIMETRY AND TIDE GAUGES

A. Cazenave, J.F. CRETAUX, K. Dominh and L. Soudarin (LEGOS-GRGS/CNES, 18, Avenue Edouard Belin, 31401 Toulouse Cedex 4, France, e-mail: anny.cazenave@cnes.fr)

Since late 1992, the global mean sea level variations are precisely monitored from space by the Topex-Poseidon (T/P) altimeter satellite. A mean sea level rise of 1.8 + 0.2 mm/yr is recorded between January 1993 and October 1998.

Excluding data since early 1997 which evidence the large rise and fall of the sea level associated with the 1997 El Ni=F1o, the global mean sea level rise measured by T/P. Poseidon reduces to ~ 1.0 + mm/yr. There is indication however that T/P underestimates the mean sea level rise by ~ 1-1.5 mm/yr due to an instrumental drift of the onboard radiometer. By comparing sea level changes measured by T/P and tide gauges of the GLOSS network since early 1993, we confirm the presence of a sea level drift of ~ -1.8 mm/yr between T/P and tide gauge measurements in agreement with independent estimates of the T/P radiometer drift. As a consequence, the global mean sea level monitored from space shows a rate of change of ~ 2.5 mm/yr since early 1993. Attempt to understand this global sea level change in terms of thermal expansion and mass exchange with continental waters is also presented.

JSG11/E/09-A3 0930

FOURIER AND WAVELET ANALYSES OF TOPEX/POSEIDON-DERVIED SEA LEVEL ANOMALIES OVER THE SOUTH CHINA SEA - A CONTRIBUTION TO SCSMEX

Cheinway Hwang and Sung-An Chen (both at Department of Civil Engineering, National Chiao Tung University, 1001 Ta Hsueh Road, Hsinchu 300, Taiwan. email: hwang@geodesy.cv.nctu.edu.tw)

We process 5.6 years of TOPEX/POSEIDON altimeter data and obtain time series of sea level anomalies (SLA) over the South China Sea (SCS). Fourier analysis shows that sea level variability of SCS contain major components with periods larger than 180 days. Tidal aliasing create 30 to 180-day components which can be misinterpreted as wind-induced variabilities. Continuous and multiresolution wavelet analyses of SLA show that sea level variability of SCS has monthly to interannual components, but with time-varying amplitudes. The sea level slope is 8 mm/year, possibly caused by decadal climate variation. Coherences of sea level with wind and sea surface temperature are significant only at frequencies lower than 2 cycles/year, and wavelet coefficients show high coherences at the interannual scales. Zero-crossing of SLA is highly correlated with the onset of the summer monsoon. The interannual variabilities of SCS sea level are correlated with ENSO, and most important is that when the El Nino-like wavelet coefficients change curvatures, an El Nino starts to develop, suggesting that real-time SLA over SCS or the western Pacific warm pool can provide a warning to the occurrence of an EL Nino. This a contribution to SCSMEX.

JSG11/E/12-A3 0950

SEALEVEL VARIATION DATA EVALUATION IN INDONESIAN WATERS

Sobar SUTISNA (Center for Basicdata Survey, National Coordination Agency for Surveying and Mapping, P.O.Box 46/CBI, Cibinong 16911, INDONESIA, email: sobar@cbn.net.id) Mujiana and Bambang S. Pratomosunu (both at Center for Marine Mapping and Aerocharting, Bakosurtanal, P.O. Box 46/CBI, Cibinong 16911, INDONESIA)

Bakosurtanal (the National Coordination Agency for Surveying and Mapping of Indonesia) has today 28 tide stations in operation. The sealevel observation in the country are essential and are used for different purposes in both local and regional level as well as on a global scale. Tidal analysis were carried out using harmonic analysis method by treating the sum of a finite number of harmonic constituents whose angular speeds and phases are determined from the astronomical arguments.

This paper presents the results of analysis have been carried out at Bakosurtanal for all of the 28 tidal station data in order to determine mean sea level (MSL) position, to calculate low astronomical tide (LAT) and high astronomical tide (HAT) at each tide station locations. The possibility to define sealevel baseline data for sealevel rise monitoring in relation to climate change in Indonesia will also be discussed.

JSG11/W/03-A3 1010

A COMPREHENSIVE ANALYSIS OF THE TIDAL AND OCEANIC SIGNALS FROM THE WOCE SEA LEVEL OBSERVATIONS AND PRELIMINAR APPLICATIONS.

F. Ponchaut (email: ponchaut@cnes.fr), F. Lyard (email: Florent.Lyard@cnes.fr) and C. Le Provost (email: Christian.Le-Provost@cnes.fr) (all at LEGOS/GRGS, UMR5566 CNES- CNRS-UPS,

14 Avenue E. Belin, 31400 Toulouse, France)

A new sea level observation network initiated by the WOCE program is delivering hourly data from 150 ocean stations worldwide. A complete analysis of these data, the WSLA 98 data set (standing for WOCE Sea Level Analysis, 1998) was performed using a least-squares method allowing better accuracy over time series virtually unlimited in time (up to 13 years in this case). In addition to the classical tidal constants, the power density spectrum of the de-tided signal was computed and the 95% confidence intervals have been systematically estimated. Beyond the need for high accuracy de-tided sea level data, like in vertical ground movements applications, or for mean sea level monitoring purposes, the analysis outputs are a highly valuable material in the studies of the ocean waves, and in particular for the tidal modelling. The availability of confidence intervals allows to investigate more precisely the tidal waves of low amplitude, like long period or non-linear tides. For instance, the oceanic residual energy spectrum is used to select the tidal constant set needed to validate and compare the long period tide models. This selection is performed by eliminating stations where the oceanic continuum spectrum magnitude (around a given tidal frequency) is more than 25% of the corresponding tidal peak amplitude. It also gives the opportunity to prescribe realistic data error bars in the data assimilation models. This analysis intends to lead to a better exploitation of sea level observations, which contain highly valuable information in the tidal and non tidal domains. The authors emphasise the need for the WOCE Sea Level network to be maintained.

JSG11/W/04-A3 1030

SEPARATION OF THERMAL AND MASS SIGNALS IN SEA LEVEL VARIABILITY BY COMBINING SATELLITE ALTIMETRY AND XBT DATA

Don CHAMBERS, Jianli Chen, and Byron D. Tapley (University of Texas at Austin, Center for Space Research, R1000, Austin, TX, 78712, USA, Email: chambers@csr.utexas.edu)

Sea level variability can be measured quite accurately on regional and global scales with satellite altimetry. However, the altimeter measures the combination of changes caused by thermal and mass variations. In order to study the relative effect of mass or thermal variations, other measurements must be combined with altimetry. Expenable Bathythermagraphs (XBTs) dropped from ships measure temperature at depth, and can be integrated to compute the sea level variations due to thermal changes. However, the XBT casts are limited in both time and space. In this study, we interpolate XBT data for the period 1993 to 1997 to a uniform grid by using Empirical Orthogonal Functions computed from TOPEX/POSEIDON (T/P) altimeter data. The T/P and XBT sea level measurements are then compared on regional and global scales to determine the relative size of thermal to mass variability to the total signal. Interannual and annual variations in the global water mass balance are also studied by comparing the T/P and XBT measurements with output from various atmospheric and hydrological models.

JSG11/W/18-A3 1110

SEA-LEVEL CHANGES AND VERTICAL CRUSTAL MOVEMENTS IN THE SOUTHERN CARIBBEAN FROM SATELLITE ALTIMETRY, TIDE GAUGE RECORDS AND GPS HEIGHT DETERMINATIONS

Wolfgang BOSCH, Hermann Drewes, Klaus Kaniuth (all at Deutsches Geodaetisches Forschungsinstitut, Marstallplatz 8, D-80539 Muenchen, Germany,

email: bosch@dgfi.badw-muenchen.de) , Hans Kahle (Institut fuer Geodaesie und Photogrammetrie, ETH Hoenggerberg, HPV G 52, Zuerich, Switzerland, email: kahle@geod.ethz.ch) , Napoleon Hernandez (Servicio Autonomo de Geografia y Cartografia Nacional, Av, Este 6, C.S.B., Edif. Camejo, Piso 2, Caracas 1010, Venezuela, email: hernandez.sagecan@conicit.ve)

The sea surface variability derived from four years of Topex/Poseidon altimeter data shows a heterogeneous feature in the southern Caribbean. The variations are analyzed and compared with records of tide gauges along the northern South American coast of the corresponding epoch. Long-term tidal records present a non-uniform behaviour ranging from -2 mm/a decrease to +2 mm/a increase of sea-level. To separate in the tidal records the sea-level changes from the vertical crustal movements, the tide gauges with extreme observed variations (Carupano, La Guaira, and Amuay, all at the Venezuelan coast) have been precisely determined in height by GPS measurements in three campaigns from 1993 to 1999. The results are analyzed together with the satellite altimetry data.

JSG11/W/19-A3 1130

THE INTEGRATION OF SEA-LEVEL MEASUREMENTS FROM SATELLITE ALTIMETRY SPANNING TWENTY-FIVE YEARS

Timothy J. URBAN, John C. Ries, Byron D. Tapley (all at University of Texas at Austin, Center

for Space Research, 3925 W. Braker Lane, Suite 200, Austin, TX, 78759-5321, USA,

e-mail: urban@csr.utexas.edu, ries@csr.utexas.edu, tapley@csr.utexas.edu), C.K. Shum (Ohio State University, 2070 Neil Ave., Columbus, OH 43210, USA, e-mail: ckshum@osu.edu)

Satellite altimeter data from the two and a half decades of data collection has been integrated into one data set for the purpose of monitoring long-term sea level. The satellites included are GEOS-3 (1975-1978), SEASAT (1978), GEOSAT (1985-1989), ERS-1 (1991-1996), TOPEX/POSEIDON (1992-present), and ERS-2 (1995-present). The historic data are improved with the most modern satellite orbits, media and geophysical correction models, yielding data sets as homogenous as currently achievable. Crossover and other analyses are performed to validate the correction models and resulting sea-level data. An accuracy assessment of the combined data is given, including orbit error and long-term analysis of altimeter corrections. Relative sea-level biases between satellite missions are determined through tide gauge and other analyses. Including gaps in the time series, a twenty-five year estimate of satellite-measured sea-level rise is produced.

JSG11/W/22-A3 1150

A GENERALIZED APPROACH TO THE CROSSOVER ANALYSIS OF ALTIMETRY DATA

Q.W. LIU and Y.Q. Chen (both at Department of Land Surveying & Geo-Informatics, The Hong Kong Polytechnic University, Hunghom, Kowloon, Hong Kong, email: lsqwliu@polyu.edu.hk)

The crossover adjustment has been extensively employed to correct altimetry data for the radial orbit error. The traditional procedure for the crossover adjustment is to use the difference of the sea surface height at crossover points to form the system equation and solve it by least squares. It is shown that the method is not effective in small ocean basins due to the few crossovers per track and to the relatively short tracks. An alternative approach is to make direct use of altimetry measurements at crossover points without differencing and to estimate simultaneously orbit parameters and mean sea surface heights in a single, unified adjustment. The rank defect inherent in the crossover estimation problem is overcome by introducing additional constraints, as fixing certain number of parameters of non intersecting tracks (Schrama 1989), or adding an appropriate set of new constraints (van Gysen et al., 1997), or by combing the altimetry data with tide gauge station data (Fenoglio and Groten 1995). The purposes of our research are twofold. One is to theoretically generalize the existing crossover methods in order to provide a convenient theory for crossover analysis. It is shown that all the existing methods are special cases of our generalized approach developed by considering a weight for the estimated parameters of orbit error and mean sea surface height. Another is to practically compute a realistic estimate of the total radial orbit error and the mean sea surface. The detailed formula of the generalized approach is derived. Formulation and numerical example demonstrate its application.

JSG11/W/21-A3 1210

Local sea level can be monitored...

Kyra VAN ONSELEN, Delft University of Technology, Email: onselen@geo.tudelft.nl

Local sea level can be monitored using a tide gauge, which relates sea level height variations to a local bench mark. In order to combine sea level height series for a number of tide gauges, the heights of these local tide gauge bench marks have to be determined in a common reference system. The geodetic measuring techniques used to perform these height connections will introduce measuring and systematic errors in the collection of sea level height series. The purpose of this work is to investigate how the detectability of a common pattern in sea level height series is influenced by the limited accuracy of the geodetic measuring techniques used to connect the tide gauges in height. Time series used are a combination of simulated patterns with periodic fluctuations obtained from actual tide gauge data. The effect of connecting the tide gauge bench marks is simulated based both on knowledge of error characteristics of geodetic measuring techniques and (differences) between actual sea level height series. Sea level height variation patterns under consideration are one common variation curve for all tide gauges, and a linear variation in this pattern for tide gauges situated along a specific coast.

Wednesday 21 July PM

Presiding Chair: R. Sabadini

JSG11/E/10-A3 1400

DETERMINATION AND CHARACTERIZATION OF LONG-TERM MEAN SEA LEVEL CHANGE

C. K. SHUM (Department of Civil and Environmental Engineering and Geodetic Science, The Ohio State University, 470 Hitchcock Hall, 2070 Neil Avenue, Columbus, OH 43210-1275, USA,

Email: ckshum@osu.edu), Hong-Zeng Tseng (Ohio State University) Tim Urban and Byron Tapley (Both at University of Texas), Michael Anzenhofer GeoForschungs Zentrum Potsdam, Germany), and Philip Woodworth (Proudman Oceanographic Laboratory, Bidston, UK)

Recent studies of global sea level change have concluded that the global eustatic rate of sea level rise during the last century has been 1-2 mm/yr [Warrick et al., 1993]. Although there is no firm evidence of an acceleration in the rate of sea level rise over this time period [Woodworth, 1990], the projected future sea level rise is 13 +/- 4 cm during the next 40 years (1987-2027), and 61 cm over the next 100 years (1987-2087) [Woodworth, 1995]. Sea level change represents consequences of complicated interactions of the solid Earth-atmosphere-hydrosphere-ocean- cryosphere system and in part forced by human-originated greenhouse effect. Current long-term (40-100 years) sea level change estimates (e.g., IPCC studies) are primarily provided by long-term tide gauges located near coastal regions and continental margins. The estimate has deficiencies from local land movements (e.g., due to postglacial rebound and subsidence) and vertical datum knowledge, and the fact that the data only covers less than 5% of the global ocean. Satellite radar altimetry missions, both historic and current (Geos-3, Seasat, Geosat, ERS-1, TOPEX/POSEIDON, ERS-2, GFO-1, and future (Envisat, Jason-1, NPOESS), would provide global coverage but is deficiency both in data span (less than 10 years of continuous data), and the difficult in the knowledge of instrument biases. In this study, we attempt to use the available, improved processing of multiple mission altimeter measurements [Urban et al., abstract submitted to this session] and long-term tide gauges, to provide a combined analysis to examine the current rate of mean sea level rise. Second part of the study is intended to provide a characterization of the sea level change by using modern observations, including global sea surface temperature (spaceborne and in situ), ice extents [M. Anzenhofer et al., abstract submitted to session JSH12], postglacial rebound models, glacier and ice sheet mass balance data. Specific examples such as the sea level change in the China Sea region during the last decade will be discussed in the presentation.

JSG11/P/01-A3 1420

INTERANNUAL AND ANNUAL SEA LEVELS IN THE SEISMIC AND TECTONIC ACTIVE ZONE IN THE NORTHWESTERN PACIFIC

Shigehisa NAKAMURA(Famille Villa-A104,Tanabe 646-0031,Japan)

lnterannual and annual sea levels in the seismic active zone is studied in order to realize what effects are seen just neoghbour the tectonic boundary. A special reference is that in the north- west Pacific. As for the problem of the tectonic effect to the apparent sea level at a tide station, a typical example of the 1995 seismic event can be introduced. The tide station of Kobe was strongly affected at the event, so that the marigram showed the patterns of the seismic motions and the successive upheaval as the vertical ground motion.This effect can be seen in the sea level variations of monthly mean or annual mean after processing the hourly read tide data. The interannual sea levels for about 100 years in the 20 century suggests the effect of under-ground waters and the effects of the crustal upheaval and the Kuroshio as an ocean curent intensified in the western boundary of the Pacific. Spectral analyses are undertaken to demonstrate what specific variations of the spe(--trat patterns at several tide stations located just south of the tide station Kobe neighbour the boundary of the Eurasian tectonic plate. At the three tide stations located on the coast facing the Pacific, the spectral patterns are suggesting a signifcant effect of the Kuroshio meandering just off the tide stations. Successive studies should be followed after the year of 1999.

JSG11/W/05-A3 1440

BIFROST PROJECT: FENNOSCANDIAN NETWORKS FOR CRUSTAL DEFORMATION AND SEA-LEVEL CHANGES

J.M. JOHANSSON, H-G, Scherneck, G. Elgered (all at Onsala Space Observatory, Chalmers University of Technology, SE-43992 Onsala, Sweden, email: jmj@oso.chalmers.se) J.L. Davis, R.A. Bennett

(both at Harvard-Smithsonian Center for Astrophysics, Cambridge, MA, USA,

email: jdavis@cfa.harvard.edu); H. Koivula, M. Poutanen (both at Finnish Geodetic

Institute, PL 15, FI-02431 Masala, Finland, email: hannu.koivula@fgi.fi) ; G. Hedling

and B. Jonsson (both at National Land Survey of Sweden, SE-80182 G‰vle, Sweden,

email: gunnar.hedling@lm.se)

Project BIFROST (Baseline Inferences for Fennoscandian Rebound Observations, Sea level, and Tectonics) uses the continuous GPS observations in northern Europe to determine the 3-D motion associated with glacial isostatic adjustment. The backbone in this project is the permanent networks in Finland (FinnRef) and Sweden (SWEPOS). These networks has also proven to be an excellent tool for studying various effects of the environment near to the permanent sites.

In total, GPS-data from about 2000 days, have been acquired and analyzed. We have used different software packages and strategies. The entire data set has been analyzed with the GIPSY software package, both in a network solution as well as with the Precise Point Position method. Data from 1996 an onwards have also been analyzed using the Bernese software, partly as a contribution to the EUREF permanent network and the IGS. Furthermore, periods of data have been analyzed with the GAMIT software package. The BIFROST project, the networks and their purposes are presented. We discuss the error budgets in an inter-comparison of analysis procedures. We discuss antenna and site related perturbations and methods for their reduction.

JSG11/W/09-A3 1500

BIFROST PROJECT: 3-D CRUSTAL RATES INFERRED FROM GPS IN APPLICATION TO THE POSTGLACIAL REBOUND AND SEA-LEVEL IN FENNOSCANDIA.

H-G. SCHERNECK, J.M. Johansson (both at Chalmers, Onsala Space Observatory, SE-439 92 ONSALA, Sweden, e-mail: hgs@oso.chalmers.se, jmj@oso.chalmers.se) M. Vermeer (Finnish Geodetic Institute, PL 15, FI-02431 Masala, Finland, e-mail: mv@fgi.fi) J.L. Davis, R.A. Bennett,

G.A. Milne (Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02133, USA,

e-mail: jdavis@cfa.harvard.edu, gmilne@cfa.harvard.edu, rbennett@cfa.harvard.edu) J.X. Mitrovica (Dept. Physics, University of Toronto, Toronto, Ontario, Canada M5S 1A7)

Since autumn 1993, the BIFROST project has provided daily GPS solutions of geodetic positions from a network of more than 40 stations covering a large area of the Baltic shield. This area is supposed to show large vertical motion due to glacial isostatic rebound after the deglaciation at the end of the pleistocene. This presentation will discuss the inference of three-dimensional rates of crustal motion at the GPS stations with respect to (1) a plate-fixed average as regards the horizontal components; (2) a geocentric reference in order to infer absolute changes of sea level from vertical crustal motion and models of geoidal rebound. We show that the horizontal strain rate pattern is largely dominated by unilateral extension and low in horizontal shear. As regards vertical motion a crucial issue is the stability of the geocentre in the GPS frame. We show results from Precise Point Positioning analysis which exhibits the problem more clearly than network solutions. We also show the importance to systematically reduce regional and local problems related to a number of surface loading problems.

JSG11/E/05-A3 1520

BIFROST PROJECT: APPLYING 3-D CRUSTAL DEFORMATION DATA TO CONSTRAIN MODELS OF GLACIAL ISOSTASY IN FENNOSCANDIA

Glenn MILNE, Jim Davis (Harvard-Smithsonian Center for Astrophysics, Cambridge, USA;

e-mail gmilne@cfa.harvard.edu), Jerry Mitrovica (University of Toronto, Toronto, Canada), Hans-Georg Scherneck, Jan Johansson (Chalmers Institute of Technology, Onsala Space Observatory, Onsala, Sweden) and Martin Vermeer (Finnish Geodetic Institute, Masala, Finland)

The most recent solutions of BIFROST (Baseline Inferences for Fennoscandian Rebound, Sea-Level and Tectonics) project data for 3-D crustal-motion rates show a coherent picture that is clearly related to the glacial isostatic adjustment process. In this contribution, we shall present results from a series of both forward and inverse modeling calculations that adopt a spherically symmetric earth model with a compressible Maxwell viscoelastic rheology, a suite of different ice models and a sea-load model that is calculated by solving the sea-level equation in a gravitationally self-consistent manner. Our forward modeling analysis shows, in particular, that the data provide good constraints on the location of the center of mass and the large-scale geometry of the ancient Fennoscandian ice sheet. Also, the data are highly sensitive to viscosity variations within the upper mantle and, given the results of a simple sensitivity analysis, should provide a relatively firm lower bound on this parameter. An inverse analysis, based on ICE-1 [Peltier and Andrews, Geophys. J. R. astr. Soc., vol. 46, 1976] and an earth model with a 120 km elastic lithosphere and upper/lower mantle viscosities of 1/2 x 10**21 Pa s, indicates that the data prefer a relatively `soft' upper mantle (< 10**21 Pa s) with a substantial increase in viscosity into the shallow lower mantle (viscosity within the deep mantle is not well constrained by the data). These general characteristics are consistent with a number of recently published viscosity profiles. We hope to present results from a new inverse analysis based on a viscosity profile that exhibits these general features and a recently published Fennoscandian ice model [Lambeck, Smither and Johnson, Geophys. J. Int., vol. 134, 1998].

JSG11/W/02-A3 1540

MODELLING OF THE MULTIPLE PROCESSES AFFECTING SEA LEVEL RISE

IN THE CENTRAL MEDITERRANEAN

G. Di Donato, A.M. Negredo, R. SABADINI, L.L.A. Vermeersen, J.X. Mitrovica (Dipartimento

Scienze Terra, Sezione di Geofisica, Universita di Milano, Via L. Cicognara 7, I-20129 Italy,

Email: stud1@sabadini.geofisica.unimi.it)

Geophysical modelling of tectonic mechanisms active in the central Mediterranean and of post-glacial rebound allows us to establish their relative contributions to sea-level changes. For a mantle stratified in viscosity, we obtain that the peak value of sea-level increase due to the two processes are comparable in magnitude, while its spatial variability portrays a different pattern. In the Adriatic sea, along the eastern part of the Italian peninsula, the effects of tectonic subsidence and post-glacial rebound sum up to a total sea-level increase that can be as high as 1 mm/yr. Both the amplitude of sea level rise and its spatial variability along the eastern coasts of Italy agree with secular sea-level changes inferred from archeological data, from Aquileia in the northern sector of the Adriatic sea to Egnatie in the southern part. Also in the historical cities of Venice and Ravenna model predictions agree with the data. The effects of a weak layer in the crust on the whole set of observables associated with glacial isostatic adjustment are also analyzed.

JSG11/W/10-A3 1620

MANTLE VISCOSITY INFERRED FROM FENNOSCANDIAN STRAND LINE DATA

Karin WIECZERKOWSKI (Institute of Planetology, University of Münster, Wilhelm-Klemm-Str. 10, D-48149 Münster, Germany) Jerry X. Mitrovica (Department of Physics, University of Toronto, 60 St. George Street, Toronto, Ontario M5S 1A7, Canada) Detlef Wolf (GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473 Potsdam, Germany)

The postglacial rebound of the Fennoscandian shield has raised Holocene strandlines up to 200 m above the present sea level. The space-time distribution of these strandlines was used previously (e.g. McConnell 1968, JGR 73, 7089) to calculate the relaxation spectrum of rebound, i.e. the decay time of the uplift versus the wavelength of the deformation. Since it is unbiased by assumptions regarding the space-time history of the load, this spectrum has played a significant role in the inference of mantle viscosity. Previous relaxation spectra were derived on the basis of highly contentious hypotheses on strandline evolution (see Wolf 1996, GJI 127, 801). In this study we use strandlines based on more reasonable assumptions. Accounting also for the results of synthetic strandline calculations, we derive a new relaxation spectrum. Following Mitrovica & Peltier (1993, GJI 114, 45), we apply formal inverse theory to the new relaxation spectrum. In particular, we present estimates of the resolving power of the data as well as revised constraints, including uncertainties, on mantle viscosity. We find (5 +- 1) x 10^20 Pa s in the depth range 100 to 500 km and (2 +- 1) x 10^21 Pa s in the depth range 300 to 800 km.

JSG11/E/08-A3 1640

MICROFOSSIL EVIDENCE OF LAND SUBSIDENCE OF THE 1964 EARTHQUAKE IN ALASKA

YONGQIANG ZONG and Ian Shennan Department of Geography, University of Durham, Durham, UK

Diatom and pollen assemblages from a number of outcrops and cores at Girdwood Flats and Kenai River, in the Cook Inlet, record four phases of relative land and sea-level movements. The first phase is the development of freshwater swamp above high marsh sediments during relative land uplift, caused by strain accumulation along the locked portion of the Alaska-Aleutian subduction zone. In phase two, diatom and pollen groups from the top few centimetres of the peat reveal relative sea-level rise (or relative land subsidence) prior the main shock of the earthquake of March 1964. The third phase is rapid land subsidence, 1.7m at Girwood Flats and 0.5m at Kenai River, during the main shock that initiated intertidal silt sedimentation above the peat. The final phase is the colonisation of mudflat by saltmarsh communities during the post-seismic land uplift.

The concentrations of radionuclide 137Cs were measured to determine the time span of the second phase. The results suggest that the small land subsidence occurred about twenty years before the main shock in 1964, and was possibly a result of a major pre-shock.

The microfossil data compare favourably with sequences from Washington, Oregon and British Columbia that record late Holocene submergence events caused by large earthquakes. The results from Alaska raise the question whether pre-seismic sea-level rise (or land subsidence) represents any kind of warning of large earthquakes.

Co-workers include Dr. R. Combellick, Alaska Division of Geological & Geophysical Surveys, Fairbanks, Alaska, USA; Miss S. Hamilton and Ms M. Rutherford, Department of Geography, University of Durham, Durham, UK.

JSG11/E/17-A3 1700

TECTONIC-HYDRO-CLIMATIC MECHANISM OF THE LEVEL FLUCTUATIONS IN THE INTRACONTINENTAL SEAS: THE CASPIAN

Dmitri A. LILIENBERG (Institute of Geography, Russian Academy of Science, 29, Staromonetny per., 109017 Moscow, Russia, e-mail: geography@glas.apc.org)

A climatic concept has failed to predict a sudden and disastrous rise of the Caspian Sea level in 1978-95. As our recent studies show, the major impact to the Caspian Sea bath volume is caused by the lateral crustal motions in the Caucasian-Caspian nappe-overthrust domain. The lateral motions are parented by compression and tension produced by the Arabian Plate collision and are in complicated relations with hydro-climatic variations. The concurrent action of the tectonic and hydro-climatic mechanisms has also resulted in disagreement between Caspian transgressions and glaciation events. In the second half of the 20th century the Circum-Caspian was covered by repeated geodetic, sea-level, and seismic surveys with involvement of 10 geodynamic test areas. The resultant set of maps and profiles show the high mobility of the Circum-Caspian area, the reciprocal and variable contemporary motions, their dependence on the general morphostructure. Phases of the crustal extension and subsidence of the Caspian bath correlate with the sea-level drops in 1905-15 and 1929-40; the compression and uplifting correlate with the sea-level rise in 1978-95; but all level changes are independent of the actual sea water volume. The fluctuations of the sea level are in connection with the large-scale geodynamic processes, such as seismic activation in the Carpathians, Caucasia, and Asia Minor. The vertical motion rates make up few millimetres per tens of years, but can attain 5-7 cm/y. As recorded by trilateration and GPS surveys the lateral displacements are in the range of centimetres per year. The crustal motions, seismicity, mud volcanism, dynamics of subsurface fluids have a quasi-recurrence of 100-120, 80, 50-60, 35-40, 25-30, 10-20, 5-7, 2-3, and 1 year, which correlate with the hydro-climatic variations. Thus the Caspian Sea level fluctuations are caused by the integral tectonic-hydro-climatic mechanisms governed by the global processes.

JSG11/W/06-A3 1720

HOLOCENE SEA-LEVEL CHANGES ALONG THE MEDITERRANEAN COAST OF ISRAEL - ARCHAEOLOGICAL OBSERVATIONS VERSUS ISOSTATIC MODEL

DORIT SIVAN and Avner Raban (both at the Center for Marine Studies, .University of Haifa, Haifa, 31905 Israel, Email: Dsivan@research.Haifa.ac.il). Shimon Wdowinski (Department of Geophysics and Planetary Sciences, Tel Aviv University, Ramat Aviv, 69978, Israel, Email: Shimon@geol.tau.ac.il). Kurt Lambeck (Research School of Earth Sciences, The Australian National University,Canberra, ACT 0200, Australia, Email: Kurt.Lambeck@anu.edu.au) Ehud Galili (Marine Archaeology branch, Israel Antiquities Authority, P.O.B. box 180, Atlit 30350, Israel, email: Udi@israntique.org.il)

Underwater and coastal archaeological research conducted along the Mediterranean coast of Israel reveals significant sea-level changes during the Holocene. We consider the archaeological observations as upper and lower constraints of paleo sea-level. High resolution numerical models of sea-level change during the Holocene have been developed for the Mediterranean coast of Israel, and the predicted curves have been compared with field observations. The archaeological observations and the model curve are generally consistent indicating that sea-level rising rates along the Israeli coast were high and reached values of 7.5-9.0 mm/yr between 8,000-6,000 BP. The rate decreased to less than 1 mm/yr from 6000 BP. The close agreement between the modeled and the inferred curves also suggest that the local vertical crustal movements were low: less that 0.25 mm/yr, for the past 8 000 years.

JSG11/W/12-A3 1740

SEA LEVEL CHANGES AND GEODYNAMICS

Nils-Axel Morner Paleogeophysics & Geodynamics, S-10691 Stockholm, Sweden,

email: morner@pog.su.se

Sea level can only change within the limits set by the physical processes involved. Sea level changes is a multi-facetted problem, however, that calls for deep knowledge in a number of different subjects and interacting processes. Sea level changes as a function of several different processes and by that it is, at the same time a remarkable measure of a number of different geodynamic processes. On the Ma-time scale, we are primarily dealing with long-term deformations of the ocean basin volume, the mass distribution and the rate of rotation. On the Ka-time scale, we are dealing with glacial eustasy and related effects on isostasy, rotation and geoid shape. On the Cy-time scale, we seem - though much debated - to be dealing primarily with changes in the distribution of ocean masses via changes in the oceanic circulation in a feed-back coupling with the Earth's rate of rotation. On the Yr-time scale, we are dealing with local dynamic factors, and in some places ENSO-effects, too. For different reasons and aspects, both geophysesists and geologists tend to oversimplify the picture, however. This is especially true when it concerns the changes during the last centuries and the expected future changes. Glacial eustatic changes virtually finished in Mid-Holocene time. The most powerful Late Holocene process of rapid changes in regional sea level refers to the variability in the main oceanic circulation system as a function of interchange of angular momentum between the solid Earth and the hydrosphere. The steric expansion of the water column is limited to individual oceanic layers with quite different recycling times. Only the surface layer can experience rapid changes in the order of are less than 10 cm in a century. If, on the whole, there was a global rise in sea level in the last 150 years, it cannot have exceeded 1.1 mm/yr. Evaluating the past and present changes in sea level, the expected changes in the near-future are estimated at "some 10 cm - at the most 20 cm - per century".

Thursday 22 July AM

Presiding Chair: H. Drewes

JSG11/E/11-A4 0950

SCHEME OF THE CIRCUM-PACIFIC HIGH PRECISION SEA LEVEL GPS MONITORING NETWORK

Hu Jianguo, Chinese Academy of Surveying and Mapping=20

The circum-pacific high precision sea level GPS monitoring network(i.e. a net work combing GPS points with tide gauging)is proposed to establish according to this scheme. And at the same time, the satellite altimeter data is used to monitor the absolute changes of sea level in this area, to study the el niÒo, the movements of geological plates and other geodynamical matters, and to unify the vertical datum in the circum-pacific area. The network is composed of 41 points which lie in Asia, Australia, South America, North America, Antarctica and some islands around the Pacific Ocean. After all those points are Surveyed during the same period through GPS. This network will be put into the ITRF96 Reference Frame. It is supposed the horizontal precision of those points will be 5-8mm, the vertical precision 10-16mm.

JSG11/E/15-A4 1010

GPS MONITORING OF VERTICAL GROUND MOVEMENTS AT UK TIDE GAUGE SITES SINCE 1990

Richard BINGLEY, Nigel Penna, Alan Dodson and Vidal Ashkenazi (IESSG, University of Nottingham, Nottingham, NG7 2RD, UK, Email: richard.bingley@nottingham.ac.uk), Trevor Baker (Proudman Oceanographic Laboratory (POL), Bidston Observatory, Birkenhead, Merseyside L43 7RA, UK,

Email: tfb@pol.ac.uk)

Long term changes in mean sea level recorded by tide gauges are corrupted by vertical ground movements, which can be of a similar order of magnitude. Hence, to properly monitor changes in mean sea level, the rates of any vertical ground movements at tide gauge sites must be determined. The application of GPS to monitoring vertical ground movements at selected sites of the UK National Tide Gauge Network has been on-going at the IESSG and POL since 1990.

This work has closely followed the recommendations of the "Carter Reports" of 1989 and 1994, plus the more recent IGS/PSMSL workshop in 1997. Nine episodic GPS campaigns were carried out between 1991 and 1996, incorporating sixteen UK tide gauge sites. Since 1997, five of these sites have now been equipped with continuously operating GPS receivers (COGRs), which were some of the first to become operational at tide gauges in Europe.

This presentation charts the developments in using GPS at tide gauge sites, presents the results obtained in the UK from episodic GPS campaigns and COGRs, and discusses the contribution of these COGRs to a future European Sea Level Observing System (EOSS).

JSG11/E/16-A4 1030

GROUND MOVEMENTS OF THE DORIS STATIONS DUE TO THE SEASONAL LOADING

LAURENT SOUDARIN, Anny Cazenave, and Jean-Francois Cretaux (Laboratoire d'Etudes en Geophysique et Oceanographie Spatiales, 18 avenue Edouard Belin, 31401 Toulouse cedex 4, France, email: Laurent.Soudarin@cnes.fr)

The ground positioning performances of the DORIS space system are at present at the centimeter level for the horizontal and vertical components (even subcentimetric for high-latitude stations), thanks to the combination of the data collected by the satellites carrying a DORIS receiver. Spot2, Spot3, and Topex/Poseidon data have been analysed over 1993-1997, and monthly positions computed for the about 50 stations of the permanent tracking network. Periodic variations are detected in the time series of the coordinate stations. In particular, a few mm annual signal is observed for a quarter of the sites. We investigate the atmospheric pressure loading effects which could cause the earth deformation signals observed. From the gridded surface pressure of the NCEP/NCAR reanalysis project, and the load Love numbers, we compute the crustal deformation by the atmosphere. We also investigate the loading deformation of snow and Continental hydrology.

JSG11/W/01-A4 1050

MONITORING THE SEA-LEVEL AND VERTICAL CRUSTAL MOVEMENTSIN THE SOUTH ATLANTIC ALONG THE COAST OF ARGENTINA

Hermann DREWES, Wolfgang Bosch, Klaus Kaniuth (all at Deutsches Geodaetisches Forschungsinstitut, Marstallplatz 8, D-80539 Muenchen, Germany,

email: drewes@dgfi.badw-muenchen.de); Claudio Brunini, Juan Moirano, Francisco Azpilicueta (all at Facultad de Ciencias Astronomicas y Geofisicas, Universidad La Plata, La Plata, Argentina,

email: claudio@fcaglp.fcaglp.edu.ar)

Long-term tide-gauge records along the South Atlantic coast of Argentina show aquite heterogeneous behaviour of sea-level variations ranging from -3.5 mm/a decrease to +5 mm/a increase. These signals are analyzed and compared with time series derived from six years Topex/Poseidon satellite altimetry data. Since December 1998, a selected number of representative tide gauges is monitored by repeated precise height determinations using the Global Positioning System (GPS). The ellipsoidal height of the sea surface at the tide gauges is compared with the sea surface height from satellite altimetry in order to detect systematic effects. The repeated height determinations of the tide gauges will provide information about vertical crustal movements.

JSG11/W/08-A4 1130

TECTONIC MOTION OF NY-ALESUND VLBI OBSERVATORY

Oleg TITOV (Astronomical Institute of Saint-Petersburg University, Bibliotechnaya sq., 2, Petrodvorets, Saint-Petersburg, 198904, Russia, email: oleg_titov@usa.net)

Ny-Alesund VLBI observatory has been operated since October, 1994. It is located on Svalbard archipelago which is an active tectonic area near edge of Eurasian plate. Therefore motion of the Ny-Alesund observatory is very complex. Natural deformation processes at the region could cause an unexpected shift of the site position. The previous estimates by GPS and VLBI technique sometimes show a disagreement between each other. About 150 single VLBI experiments has been processed by special software OCCAM to provide a detailed picture of the Ny-Alesund velocity drift in relation to conventional NNR-NUVEL-1A model. The five-years time series for vertical and horizontal components are presented. Generally, Ny-Alesund observatory moves in a good accordance with considered model. But it is suspected that seasonal effects in vertical component are able to distort the estimates of linear velocity. Origination of the seasonal effects is not clear up to now. Slow variations of longitude component are observed after removing of linear trend.

JSG11/W/17-A4 1150

SEA LEVEL FROM GPS BUOYS

Authors: M. PARKE, C.K. Shum, K. Snow, K. Cheng (all at The Ohio State Univ., Columbus, OH) G. Mader and D. Martin (both at NOAA/NOS, Silver Spring, MD) F. Kelly and N. Guinasso (both at Texas A&M, College Station, Tx); G. Jeffress (Texas A&M, Corpus Christi, Tx) R. Gutierrez and B. Schutz (both at Univ. of Texas, Austin, TX) and J. Blaha (NRL, Stennis Space Center, MS)

Sea level measured from buoys can satisfy a wide range of oceanographic and geodetic requirements. In this talk we will address their roles in calibrating altimetric satellites, as "GPS tide guages", spatial mapping of the sea surface for oceanographic and geodetic purposes, and in coastline mapping (in conjunction with other remote sensing techniques). Two experiments will be used for examples.

Between 19 April 1998 and 3 May 1998 a gps equipped buoy was taken on a Texas A&M hydrographic cruise. Data were taken using a gps buoy at 53 stations in conjunction with CTD measurements. All but one station were taken along altimeter ground tracks either for Topex/Poseidon or ERS-2. Two stations encompassed T/P overflights while one station encompassed an ERS-2 overflight. Other measurements along ground tracks were never more than about two days separated in time from the corresponding altimeter measurements. Data were taken in the El Dorado Eddy and the Fourchon Eddy with dynamic sea level signatures expected to be up to 20 cm and 60 cm respectively relative to the position of the mean sea surface. One station was taken near the center of the Fourchon Eddy (away from altimeter tracks) with an expected dynamic signal of about 80 cm.

A GPS buoy is being developed for deployment near a triple crossover point (a location where the three major altimetric satellite orbits cross: ERS, T/P, and GFO) as part of a plan for long term altimeter calibration and cross calibration.

JSG11/W/23-A4 1210

Local Tide and Geoid determination from Airborne GPS Kinematic Positioning and LADS Survey

Shaowei Han, Chris Rizos and A. H. W. Kearsley (at School of Geomatic Engineering, The University of New South Wales, Sydney NSW 2052, Australia, email: s.han@unsw.edu.au, c.rizos@unsw.edu.au, w.kearsley@unsw.edu.au)

Airborne GPS kinematic positioning provides the geodetic height of the aircraft, and the Laser Airborne Depth Sounder (LADS) system can provide the distance between the aircraft and the sea surface. The height of the sea surface above the ellipsoidal model can then be determined, and this includes the local tide, local geoid, sea swell and noise. The local tide and local geoid can then be identified based the specific spectral features of the different components. In this paper, an experiment to test the utility of this concept using the Australian LADS system, together with state-of-the-art, geodetic GPS receivers, was carried out. The combination of an ambiguity recovery technique and a linear bias correction procedure has been used for this application to overcome the technical challenges of ambiguity resolution and cycle slip repair over distances many hundreds of kilometres. Finite Impulse Response filters are used to eliminate the measurement noise and sea swell distortion so that the local tide and geoid can be extracted. From a data analysis of this experiment, the accuracy of the local tide and geoid has been confirmed to be at the centimetre level from the comparison of four different sets of data in which three kinds of GPS receivers were involved. The comparison between the experimental results and theoretical computations of the local tide and geoid is attempted and commented upon.

Thursday 22 July PM

JSG11/E/02-A4 Poster 1400-01

THE SEA LEVEL CHANGES AND VERTICAL GROUND MOVEMENTS IN THE SEA OF OKHOTSK

Olga Poezzhalova and George SHEVCHENKO (both at Institute of Marine Geology & Geophysics, Russian Academy of Sciences, 693002, Nauki street, Yuzhno-Sakhalinsk, Russia,

E-mail: tsunami@sakhmail.sakhalin.ru)

We examined the tendencies of changes of annual mean levels for different stations of the Okhotsk Sea. Parameters of linear trend were determined by least square method. The most expressed trends occurred at the stations of Kuril Islands. Annual mean levels in Severo-Kurilsk (Paramushir Is.), Matua (Matua Is.) and Malokurilsk (Shikotan Is.) decreased quickly (velocities is about 5 - 6 mm/year). That it speaks about uplifting of islands. However, Shikotan subsided at 50 cm after strong earthquake in October, 1994.The land subsidence looks in area of Kurilsk (Iturup Is.) andYuzhno-Kurilsk (Kunashir Is.), particularly intensively this process proceed from 1955 to 1975. Analogous tendency observed in south-eastern coast of Sakhalin Island (Poronaysk, Vzmorye, Starodubskoe, Korsakov). That it speaks about subsiding of southern part of the Okhotsk Sea. The similar tendencies are not expressed at continental stations (Nagaeva bay, Okhotsk, Litke cape, Baydukov Island).

JSG11/E/03-A4 Poster 1400-02

INTERDISCIPLINARY APPLICATIONS OF THE GPS ARRAY "CRETE"

Erricos C PAVLIS (JCET/UMBC, NASA Goddard, Code 926, Greenbelt, MD 20771, USA,

email: epavlis@Helmert.gsfc.nasa.gov) and Stelios Mertikas (Tech. Univ. of Crete, Chania, Greece)

A regional GPS array (CRETE) is being deployed over the past two years on the western side of the island of Crete, Greece. When complete, the array will consist of a continuously operating permanent site at the Technical University of Crete (TUC), Chania, Greece, and a number of other sites some of which will be permanent, continuously tracking while others will be visited periodically in "campaign style". We present here results from the expanded utility of this multipurpose GPS array. The Eastern Mediterranean area is one of great interest for its intense tectonic activity as well as for its regional oceanography. The contribution of the array to tectonics is the deformation field that can be inferred from the daily position time series. Another component of the project is the occupation the two tide-gauge sites at Souda Bay and Heraklion. We show here the so-far available position time series from Souda Bay. Tide-gauge monitoring has gained importance lately since tectonics contaminate the inferred sea level variations, and a global network of tide-gauges with long historical records can be used as satellite altimeter calibration sites for current and future missions (e.g. TOPEX/POSEIDON, GFO, JASON-1), a common IOC-GLOSS-IGS effort, already underway. Crete hosts these two tide-gauges, the oldest in the area, and our plans are to further expand the array to the south of the island with a new site on the isle of Gavdos. Gavdos is situated under a groundtrack crossing point of the T/P and JASON-1 orbits. It can be an ideal calibration site if the tectonic motions are monitored precisely. Our future plans include the deployment of additional instrumentation at this site (transponders, etc.) to ensure the best possible and most reliable results.

JSG11/E/06-A4 Poster 1400-03

VERTICAL SITE MOTIONS FROM EuROPEAN VLBI CAMPAIGNS

James CAMPBELL, Axel Nothnagel, Klaus Boerger (Geodetic Institute, University of Bonn, D-53115 Bonn, Germany, email: campbell@picasso.geod.uni-bonn.de) Ruediger Haas (Onsala Space Observatory, Chalmers technical University, S-43900 Onsala, Sweden, email: haas@oso.chalmers.se)

Paolo Tomasi, Istituto di Radioastronomia, CNR, I-40129 Bologna, Italy,

email: tomasi@boira6.ira.bo.cnr.it) Dirk Behrend, Institut d'Estudis Espacials de Catalunya, Edif. Nexus, Gran Capita, 2-4, E-08034 Barcelona, Spain, email: behrend@ieec.fcr.es)

The European geodetic VLBI observations started in the late eighties and have been continuing through 1997 and 1998 at a rate of six experiments per year with six to nine stations in slightly different station configurations. The core network is formed by the stations of Onsala, Wettzell, Medicina, Matera, Noto and Madrid. More recently, the network has been consid-erably extended nothward with the station of Ny Alesund on Spitsbergen and eastward with Simeis on Crimea, Ukraine. For those stations with a long enough observing record we present the results for the vertical site motions from the most recent analyses. Due to the high sensitivity of the vertical component to mismodelling of the atmospheric path delays, the level of significance of the vertical rates has been marginal up to now. Special attention has been given to the local effects at the sites such as telecope deformation due to temperature variations. At three stations repair works on the wheels and tracks have lead to defined height changes which can be seen in the time evolution of the vertical components. Local geodetic measurements have been carried out to establish these discrete vertical changes with high accuracy. The size of the vertical site velocities has been shrinking over the years to become more realistic with increasing length of the data set. The most apparent vertical change with respect to Wettzell, a site in Central Europe, happens at Medicina where we see significant and sustained subsidence of several mm/year. It is sure that the results derived from the VLBI campaigns will constitute a much needed basis for the geophysical interpretation of sea level changes and other vertical motion applications.

JSG11/E/07-A4 Poster 1400-04

VERTICAL GROUND MOVEMENTS IN THE LOWER RHENISH EMBAYEMENT

Barbara GOERRES, James Campbell, Holger Kotthoff (all at Geodetic Institute, University of Bonn, D-53115 Bonn, Germany, email: goerres@picasso.geod.uni-bonn.de)

The Lower Renish Embayment is known for both its present-day seismo-tectonic activity and the extensive brown coal mining. The ground water withdrawal associated with the mining activities has produced widespread subsidence of several meters over a time span of 40 years. In parallel to the regular levellings in this area, a series of annual GPS-campaigns covering 13 simultaneously observed points has been started in 1993 in order to obtain vertical change by an an independent technique. The size of the GPS network is about 100 km x 100 km. The initial goal of the project was to develop strategies to improve the accuracy (repeatability) of the vertical component measured by GPS. Apart from taking into account troposperic refraction the observation sessions were extended to cover three entire days (3*24h) and to go down to low elevations. In addition, all antennas were calibrated for phase center variations. Extensive tests revealed that the best repeatabilities are obtained with an elevation cutoff of 10 degrees. The data from six annual campaigns show significant height changes of up to 22 mm/y at three points. For all GPS points connected to the levelling network, the vertical velocities from GPS were in excellent agreement with the levelling results. In terms of wrms repeatability the vertical GPS results are at the level of 3 to 4 mm for each epoch and the velocities are at 1 to 2 mm/y. This confirms the validity of the chosen strategy and demonstrates the potential of GPS as a high precision tool for monitoring vertical change.

JSG11/E/13-A4 Poster 1400-05

MODaLISATION STOCHASTIQUE DES VARIATIONS DU NIVEAU D'EAU DANS L'ESTUAIRE DU SAINT-LAURENT, CANADA

MOHAMMED EL-SABH (email: mohammed_el-sabh@uqar.uquebec.ca), Karim Hilmi et J.P. Chanut (all at Département d'océanographie, Université du Québec á Rimouski, Rimouski, Québec, G5L 3A1, Canada)

Les enregistrements du niveau d'eau horaire á la station Québec Lauzon située á la tête de l'estuaire du Saint-Laurent, Canada, sont analysés á la fois dans le domaine fréquentiel (de 1970 á 1979) et dans le domaine temporel ( durant l'année 1973). Les variations tidales expliquent 90 á 95% de la variabilité initiale de niveau d'eau observé entre 1970 et 1979. Les variations résiduelles (non tidales), de nature stochastique, peuvent contenir des valeurs irrégulières responsable des seiches et ondes de tempête. Bien que limitées á moins de 10% des variations initiales du niveau d'eau, ces variations résiduelles sont analysées et modélisées selon la méthodologie de Box et Jenkins afin d'identifier les facteurs environnementaux qui agissent de façon complémentaire sur le milieu marin. La réponse du niveau d'eau résiduel au forçage météorologique (pression atmosphérique et vents) se situe á une échelle de 2 á 28 jours; elle est de quelques heures á un jour pour les variations attribuables aux seiches longitudinales, aux cycles semi-diurne et diurne de la marée atmosphérique et aux périodes inertielles. Le débit d'eau douce contribue pour 29% á la variabilité mensuelle du niveau d'eau résiduel, tandis que les vents et la pression atmosphérique y contribuent pour 8,9% et 8,1% respectivement. La composante U du vent, parallèle á la cÙte, agit davantage sur les variations du niveau d'eau résiduel que la composante V, perpendiculaire au rivage, avec une contribution approximative de 7%.

JSG11/E/14-A4 Poster 1400-06

SUBSIDENCE PHENOMENON IN THE MONITORING OF SEA LEVEL RISE ALONG THE NIGERIAN COAST

DR. P.C. NWILO, Department of Surveying, University of Lagos, Yaba - Lagos, Nigeria,

Tel + 234 1 821114 ext. 1865, fax + 234 1 2691315, Email: Library@rcl.nig.com

The Nigerian coast is low lying; the rocks are of sedimentary origin and recent. The coast is still undergoing natural compaction due to the weight of the sediments. This zone is the home to oil and gas exploitation in Nigeria. A number of highly populated cities such as Lagos, Port Harcourt, Bonny and Calabar obtain a large portion of water for industrial and domestic use from the bore holes that dot the coastal environment. There is, therefore, subsidence due to fluid extraction such as water, gas and crude oil.

Analysis of tidal data from Bonny for 19 years shows that the sea level is rising at about 1mm per annum. This rate of sea level rise does not take into account the subsidence phenomenon. Also, using over 40 years of tidal data for Takoradi, Ghana, it was obtained that the sea level was rising at the rate of 3.9mm per annum. This may represent the regional situation for the Gulf of Guinea coastal environment. The impact of sea level would be more devastating than currently envisaged due to subsidence.

Efforts to monitor subsidence along the Nigerian Coast using geodetic levelling methods have not been successful due to the difficult nature of the environment and lack of funding. There is, therefore, an urgent need for Federal Government, coastal states, the oil companies prospecting in the environment and the international community to show interest in subsidence monitoring along the Nigerian Coast. This can be done through the provision of funds for subsidence monitoring. Subsidence along the Nigerian Coast can be accomplished using the Global Positioning Methods. A clear indication of the rate of subsidence will help in the design of a national policy on managing the impacts of climate change along the coast of Nigeria.

JSG11/W/07-A4 Poster 1400-07

TIDAL CHANGE IN THE YELLOW/EAST CHINA DURING THE LATE HOLOCENE

Katsuto Uehara (Research Institute for Applied Mechanics, Kyushu University, 6-1 Kasuga Koen, Kasuga, Fukuoka 816-8580 Japan, email: uehara@riam.kyushu-u.ac.jp)

One of the prominent features found in the present Yellow/East China Sea (YECS) is its large tidal influence. Maximum range exceeds 10m at the Korean coast and the vertical mixing effect of the tide is considered to be a significant factor to determine the circulation in the YECS (Lee, 1997). However, the tides in the past are expected to have different natures from today because the sea-level change is estimated to be more than 100m during the last 20000years. In particular, zonal asymmetry of the bottom topography become less evident in the lower sea- level stages because the present continental shelf extended in the western part of the YECS is mostly less than 60m deep.

In this study, two-dimensional tidal simulations are carried out for the ages of 0, 6, 9, 12 and 20 ka BP, in order to investigate the effect of the topographic change since the last glacial maximum. Several changes, including the reduction of the tidal range and the shift of the amphidrome points, are observed. Analyzing the results will not only provide a first scope to reproduce the paleotides of the Yellow/East China Sea, but also help us to understand the effect of the topography on the present tidal environment.

JSG11/W/11-A4 Poster 1400-08

LONG-TERM SEA-LEVEL VARIABILITY IN THE BALTIC SEA AND ITS RELATION TO CLIMATE PHENOMENA

Dietrich, R. and LIEBSCH, G. Institut fuer Planetare Geodaesie Technische Universitaet Dresden

D-01062 Dresden Germany

Investigations concerning long-term sea-level variations require sea-level records of a high quality and reliability. Over the past years records with a homogenous local height reference were generated at several stations in the southern Baltic Sea. The measurements at these stations were initiated in the middle of the last century and are among the longest records in the Baltic Sea. Beside monthly mean sea-level records, which are in common use, records of monthly extreme sea-level values were created. On the basis of these records we will outline different aspects of the sea-level variability. This implies the analysis of - the mean sea-level records in the time domain as well as in the spectral domain and - potential changes of the extreme sea level with respect to the mean sea level. Furthermore, we will discuss possible relations between the sea-level heights in the Baltic Sea and the North Atlantic Oscillation (NAO), which is a large-scale alternation of atmospheric masses in the North Atlantic Sector.

JSG11/W/13-A4 Poster 1400-09

CONTINUOUS GPS MONITORING AT BREST AND MARSEILLE GLOSS TIDE GAUGES

Serge Allain (*: SHOM, 13, rue du Chatellier BP 426, 29275 Brest, France), Claude Boucher

(#: IGN/LAREG, 6-8, Avenue Blaise Pascal, 77455 Marne-la-Vallee, France), Serge Lannuzel

(*), Didier Maillard (#), Bernard Simon (*) and Guy WOPPELMANN (*)

Brest and Marseille are among the longest tide gauge time series available in the world, beginning in the early (1807) and late (1885) 19th century respectively. Time series spanning over at least 40 years are actually required to filter out the interannual and interdecadal sea level variability and to derive sea level trends which might be related to recent climatic changes. However, the long term signal inferred from tide gauge records is ambiguous : it includes eustatic sea level change as well as land movements. Space geodetic techniques like GPS can measure these components and have already been recommended by several international group of experts to monitor crustal vertical movements at tide gauge sites in a well defined and maintained geocentric reference system.

Brest and Marseille tide gauges are committed to the Global Sea-level Observing System (GLOSS). The French agencies, SHOM (Service Hydrographique et Ocèanographique de la Marine) and IGN (Institut Gèographique National), are undertaking a close cooperation work to upgrade and fulfill GLOSS quality station requirements, especially the geodetic monitoring of the tide gauge zero. Both tide gauge stations have been equipped with continuous GPS receivers, since july 1998 in Marseille and since november 1998 in Brest. The hundred-year-old tide gauge in Marseille was repaired mid-1998 after the failure due to vandalism occurred in 1997. An acoustic one was purchased and installed in june 1998. This new tide gauge can provide numerical high frequency data.

The presentation will give a brief description of both coastal sea level observing stations : the local environmental conditions and the instrumentation. GPS and tide gauge data analyses will be presented. A more comprehensive sea level picture will be addressed at both sites by the synergetic tide gauge GPS approach.

JSG11/W/14-A4 Poster 1400-10

SPECTRAL ANALYSIS AND FILTERING BY TWO DIMENSIONAL FOURIER TRANSFORM OF SEA SURFACE TOPOGRAPHY OBSERVED BY TOPEX/POSEIDON ALTIMETRY.

Waldemar Popinski Department of Standards, Central Statistical Office, Al. Niepodleglosci 208, 00-925 Warsaw, Poland, Email: w.popinski@stat.gov.pl Wieslaw KOSEK Space Research Centre, Polish Academy of Sciences, Bartycka 18A, 00-716 Warsaw, Poland, Email: kosek@cbk.waw.pl

A method of spectral analysis and filtering of sea level anomalies obtained from Topex/Poseidon altimetric measurements is presented. Such sea level anomalies at a particular time moment create sea surface topography which can be analysed using the two dimentional discrete Fourier transform (DFT).

The original data called 'cycles' are defined on a rectangular grid in latitude-longitude domain and are transformed using the two dimensional DFT to corresponding wave-number domain. Filtering of the sea surface topography is achieved by multiplying the transformed data by a bivariate filter transfer function and next inverting the DFT.

These analyses enable detection of the most energetic waves in sea surface topography and estimating how much the sea level anomalies corresoponding to different time moments are correlated along the meridian and parallel directions. It has been found that the correlation and coherence of the sea surface topography along the parallel direction is higher than along the meridian direction for the whole wave-number range and independently of time.

JSG11/W/15-A4 Poster 1400-11

CHANGES IN SEA LEVEL, SEA SURFACE TEMPERATURE AND ATMOSPHERIC PRESSURE FROM SATELLITES.

Per Knudsen and Ole Baltazar Andersen (both at National Survey and Cadastre, Copenhagen NV, Denmark, email: pk@kms.dk, oa@kms.dk)

In studies of Global Change, sea surface temperature data may provide valuable information. Global sea surface temperature data may indicate changes in the heat budget of the oceans. Five years of sea surface temperature data and sea level height for the same period is analysed. Altimetry from the TOPEX/POSEIDON satellite will be used along with pressure observations provided in the T/P records and low resolution averaged sea surface temperature data from the ATSR 1 and 2 sensor onboard the ERS satellites and AVHRR data from the NOAA satellites.

The global and regional characteristics of the sea level trends and the trends in the sea surface temperature as well as trends in the atmospheric pressure during 1992-1997 are investigated. The changes in the sea level are compared with changes in sea surface temperature to decide whether the changes in sea level are related to changes in the heat content of the ocean. Spatial and temporal correlation between the signals are investigated, and a bivariate coherency analysis of the sea level together with the sea surface temperature is carried out at different spatial scales.