Report of GPS-99 Session 3
A new view of the tectonic deformation in the Pacific and Asia using space geodetic techniques (joint with APSG)

Session 3 was a well-attended and very exciting session, with 19 oral presentations and 13 poster presentations.

G. Michel (GFZ Potsdam) presented the latest velocity field from the GEODYSSEA campaign in southeast Asia. The network was surveyed in 1994, 1996 and 1998. The present solution, based on a combination of solutions by GFZ and BKG, shows the Sundaland/South China block moving SE relative to Eurasia at a rate of 13.5 mm/yr.

E. Reinhart (BKG Frankfurt) focused on the geodetic aspects of GEODYSSEA rather than the crustal deformation. BKG and GFZ solutions agree very well (1 mm horizontal, 7 mm vertical rms) after application of a similarity transformation. The velocity field has a precision of 2 mm/yr horizontal, 6 mm/yr vertical. The Euler pole for Sundaland/South China has the same location as the Euler pole for Eurasia, but an angular speed about 30% faster (all these refer to the NNR frame).

Hu Xiaogong (Shanghai Astronomical Observatory) presented results from the 1997 APRGP campaign using GIPSY Precise Point Positioning. He then presented an extensive statistical analysis of the post-fit residuals to test the hypothesis that they were white. The lower moments of the distribution matched the expectation of white noise fairly well, but there were some discrepancies.

P. Tregoning (ANU Canberra) presented recent GPS results from the Papua New Guinea region. This is a very complex area with relative plate rates in excess of 100 mm/yr. Sites on the north coast of Papua New Guinea were found to move slower than expected for the Australian plate. The Weitin fault in southern New Ireland was studied using a combination of GPS and 1975 EDM data. Although the GPS sessions were very short and the data thus imprecise, the very large deformation rates demonstrate that this fault and several parallel faults are strike-slip faults.

W. J. C. Simons (DEOS, TU Delft) showed results from GPS campaigns in northern Sulawesi, Indonesia. The data clearly showed about 50 mm/yr of strike-slip motion on the Palu-Koro fault. Some sites showed coseismic and postseismic displacements due to the Minahassa earthquake sequence.

Simons also presented a paper led by C. Vigny (Ecole Normale Superieure Paris) discussing the eastward propagation of seismicity in the Minahassa earthquake sequence. The progression of earthquakes in the sequence was consistent with stress triggering as each successive event was in an area where the Coulomb failure function was increased by the previous events. An earthquake on the Palu-Koro fault was interpreted as having been triggered by gfluid migration into the fault plane enabled by unclamping of the fault.

Takao Tabei (Kochi University) presented GPS results from the arc between Minda紡o and Sulawesi, in a gap between sites of the GEODYSSEA project. Observations at four sites show 50 mm/yr east-west shortening across the Davao trench, and a total of 100 mm/yr convergence between the Philippine Sea plate and the Sunda block.

M. Aurelio (Mines and Geosciences Bureau, Philippines) showed results from the GEODYSSEA network within the Philippines, and interpreted the probable slip rates on several faults. Estimated slip rates included 5 cm/yr on the Manila trench, 2.5 cm/yr on the Philippine fault, and 2 cm/yr on the Luzon trench.

T. Okhura (Kyoto University) presented new GPS results from the Macolod Corridor, southern Luzon, Philippines. Velocities relative to Eurasia were derived using the velocity of Manila (Yu et al., 1999). Westward velocities in Luzon decrease systematically from south to north, with an abrupt decrease in the Macolod Corridor region. A GPS site on the east side of the Philippine fault suggests a slip rate of at least 25 mm/yr.

J. Beavan (GNS New Zealand) presented an estimation of coseismic and postseismic deformation following the 1990 Ms 7.8 Luzon earthquake on the Philippine fault. Coseismic slip reached 5-6 meters over the mapped rupture area, and the rupture at depth extended >40 km north of the mapped surface rupture with 4ア2 m of slip. GPS data from 1993-1998 seem to indicate a slip rate of 40 mm/yr for the Philippine fault, much faster than geological estimates. Using the Savage and Prescott's (1978) elastic-viscoelastic model, the same data can be explained as a postseismic transient with a far-field velocity of 22 mm/yr and a viscosity of 3 * 1019 Pa-s. Beavan also showed that NE Luzon is part of the stable Philippine Sea plate, which implies that the Philippine Sea plate is subducting beneath itself at the Luzon trench, which has propagated northward over time.

J.-H.Hung et al. (National Central University, Taiwan) combined structural geology and GPS results to try to connect earthquakes with specific active faults. Their work integrated wide-angle seismic reflection data, seismic reflection data, structural analysis and well log data. They include a very dense network of 60 sites surveyed annually, as well as data from 15 permanent sites around Taiwan. They found the eastern part of their network to be behaving as a rigid block, with many thrust faults active in the western part. The slate belt is uplifting rapidly along the western Foothills, suggesting that the accretionary wedge is subcritical.

S. Fujihara (Kyoto University) presented modeling results from the Okinawa trough, trying to answer the question of whether rifting there was a passive or active process. They can explain the observations using a soft crust model with passive spreading and a low-viscosity zone beneath the backarc spreading center.

K. Heki (National Astronomical Observatory, Japan) presented models for the deformation of SW Japan based on GEONET data. Heki pointed out the importance of understanding all of the components that affect the deformation of SW Japan: (1) the motion of the Amurian plate, (2) elastic deformation associated with subduction at the Nankai trough, and (3) the collision between the NE Japan and SW Japan arcs. It is difficult to separate all of the components because they do not act orthogonally. GEONET data can resolve slip on the Median Tectonic Line but are too sparse to determine locking depth.

Y. Nakagawa (Toyama University) reported on modeling work done to understand the GEONET observations from the Chubu district, central Japan, where the Hida Mountains (highest in Japan) are located. He modified the block model of Hashimoto and Jackson (1993) to account for elastic deformation, and compared models with varying fault dip angles. He suggested that the main fault in the area dips to the north, and may represent the boundary between the North American (NE Japan) and Eurasian or Amurian plates.

J. Freymueller (University of Alaska) presented results from repeated GPS surveys of a network across the Tibetan plateau. The GPS results show significant east-west extension across the southern Tibetan plateau, and north-south shortening across the entire plateau. No strike-slip motion is observed across the Karakoram-Jiali fault zone, or any of the other major strike-slip fault zones that extend into the plateau, except for the Kunlun fault on its northern edge. Freymueller also showed results from a profile across the Altyn Tagh fault that shows a low slip rate, approximately 10 mm/yr.

Z. Shen (UCLA) reported on GPS surveys of sites around the Altyn Tagh fault, in the Qilian Shan and Tarim Basin. The slip rate on the Altyn Tagh fault is 9ア3 mm/yr, estimated independently in three segments. Tarim is a non-deforming block, and the pole of rotation of Tarim relative to Eurasia is located at about 35。N, 102。E. Slip rates for several other fault systems were estimated as well.

J. Liu (Wuhan Technical University of Surveying and Mapping) presented velocities of sites across all of China based on GPS surveys from a variety of sources. The solutions were all reduced to ITRF97 using the WTUSM PowerADJ software. Some sites have multiple measurement points, and a weighted average of their velocities were used. The velocity field was compared to velocities estimated from geologic slip rates and a crustal block model for China, with generally good agreement.

A. Sharav (Research Center for Astronomy and Geophysics, Mongolian Academy of Sciences) reported on the status of the permanent GPS site in Ulaanbaatar. The site was installed in 1995 in cooperation with GFZ-Postdam, with a Trimble 4000SSE with a Trimble Compact antenna. The site is downloaded manually and data are sent via mail to Postdam. The site moves to the southwest at 6 mm/yr relative to Eurasia, consistent with motion due to the opening of the Baikal rift.

Eric Calais (Geosciences Azur, France) presented GPS results from the Baikal rift and western Mongolia. The Baikal network has 14 sites, 11 with ウ3 occupations, and shows 4-5 mm/yr of extension. This is much greater than expected, and numerical models show that it is possible for passive rifting to be the cause of the deformation. A network in western Mongolia was installed in 1998 but has not yet been resurveyed. The observed motion of the Amurian block inferred from the Baikal data and Ulaanbaatar data is not reproduced in the numerical models of England and Molnar nor Peltzer and Saucier.

H. Takahashi (Hokkaido University) presented results from a network of 12 permanent GPS sites in NE Asia installed beginning in 1995. Velocities relative to Eurasia were determined using the ITRF velocities relative to Tsukuba and Heki's (1996) velocity of Tsukuba relative to Eurasia from VLBI. Eastward velocities of NE Asia indicate counter-clockwise rotation of the Amurian plate relative to Eurasia. The site BKI in Bering Island moves 7.5 cm/yr WNW relative to Eurasia, more than 80% of the motion of the Pacific plate. The 1995 Okha earthquake in northern Sakhalin cannot be explained by the relative motion of the Okhotsk plate and Eurasia since it lies very near the pole.

Jeff Freumueller

Session 1
Session 2
Session 3
Session 4
Session 5
Session 6
Session 7
Session 8
Session 9
Session 10
Session 11
Session 12
Session 13
Session 14