Coastally Trapped Tsunami Generated by the Ping-dong Earthquake • Guan-Yu Chen, Hsiao-Ching Chien • National Sun Yat-Sen University, Taiwan • Tso-Ren Wu • National Central University, Taiwan • Hsiang-Mao Tseng • Center of Harbor & Marine Technology, Taiwan South China Sea Tsunami Workshop 2008
台灣與海嘯之關係 Historical tsunami in Southwestern Taiwan is probably from the South China Sea South China Sea Tsunami Workshop 2008
Tsunami in history in the South China Sea （許明光、李啟彤） • 1781 年（清乾隆 46 年） 5 月 , 高雄 「忽海水暴吼如雷，巨浪排空，水漲數十丈，近村 人居被淹，皆攀援而上至尾，自分必死，不數刻， … 水暴退，人在竹上搖曳呼救 。 間有牧地甚廣及附 近田園句壑，悉是魚蝦，撥刺跳躍，十里內村民提 … 籃契筒，往爭取焉。 漁者乘筏從竹上過，遠望其 家已成巨浸，至水汐時，茅屋數椽，已無有矣。」 South China Sea Tsunami Workshop 2008
• Taiwan should be prepared for tsunamis from the SCS • Tsunamis have very high impact, but tsunami incidents are very rare. • Studies of small incidents, such as the tsunami induced by the 2006 Ping-dong earthquake, are very important because these studies can increase our understanding of tsunami features. South China Sea Tsunami Workshop 2008
Arrival time will lag dramatically if edge waves are generated • One of the most important features in tsunami propagation is the generation of edge waves. • The arrival time of a tsunami will lag dramatically if edge waves are generated • An edge wave might arrive unexpectedly several hours after the first tsunami waves have subsided South China Sea Tsunami Workshop 2008
Edge waves are trapped nearshore South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
Examples for Edge Wave Tsunami • The arrival time of the Kamchatka tsunami in 1952 were found to be twice longer for some places. • Similar phenomena had been observed in the Iturup tsunami in 1963 • More recently, the 25 April 1992 Cape Mendocino earthquake generated a Stokes edge wave and the largest tsunami wave arrived almost 3 hours after the first wave. • This indicates that an edge wave might arrive unexpectedly several hours after the first tsunami waves have subsided South China Sea Tsunami Workshop 2008
Why the generation of edge wave is important? (II) • Besides the propagation time, a tsunami in edge wave form can destroy a building with weak side walls even if the wall facing the sea is very solid and the same building can usually survive a direct tsunami of similar wave height. South China Sea Tsunami Workshop 2008
From the velocity-elevation relation, we can identify a Stokes edge wave South China Sea Tsunami Workshop 2008
The Mw 6.9 Ping-tong earthquake occurred at 20:26 local time, December 26, 2006, with epicenter (21.81°N 120.52°E) South China Sea Tsunami Workshop 2008
• As part of a monitoring system, an Acoustic Wave and Current Profile (AWCP) system is deployed near the Kaohsiung Harbor at a water depth of 18 meters • Both pressure and flow velocities are recorded twice a second for 1024 seconds (17 min) in each hour South China Sea Tsunami Workshop 2008
Traditionally, edge waves are measured along a straight coast with some 20 instruments to find the wavenumber in the dispersion relation South China Sea Tsunami Workshop 2008
If waves are generated far away along the coast, their direction is easy to determine Although reflection of the shoreline is possible, these waves are basically progressive. South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
Tsunami Simulation by COMCOT • Assuming a rigidity of 3.3×1011dyn cm-2, the fault length, width and slip are inferred from the seismic moment through scaling laws (Geller, 1976). • Together with parameters from the Harvard CMT Catalog such as centroid depth and fault geometry described by strike, dip, and rake South China Sea Tsunami Workshop 2008
The numerical simulation demonstrates that edge waves are not generated from the beginning. At first, the propagation speed is very quick because the deep water depths near the epicenter. South China Sea Tsunami Workshop 2008
Then the tsunami wave incident obliquely to the coast and some energy is trapped. According to Fig. 9(b), the edge wave is trapped at two places: the coast near (120 。 E, 22.4 。 N) and the coast near Kaohsiung. South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
Verification of the Tsunami Simulation Comparison of time-history elevation between the field measurement (blue line) and the numerical simulation (red line) at Syun-Guagg-Zuei (SGZ). Sampling rate is one point every six minutes South China Sea Tsunami Workshop 2008
Numerical simulation of the surface elevation at Kaohsiung. The red boxes are the internals when the field measurement is available South China Sea Tsunami Workshop 2008
Comparison of numerical simulation and the field measurement of Kaohsiung of the first two hours South China Sea Tsunami Workshop 2008
For waves of period 50-130 sec, amplitude power does not change after the tsunami except for 21:10-21:27 PM South China Sea Tsunami Workshop 2008
For waves of period 130-500 sec, amplitude power does not change after the tsunami except for 21:10-21:27 and 22:10-22:27 PM South China Sea Tsunami Workshop 2008
For waves of period 500-700 sec, amplitude power increases significantly after the tsunami South China Sea Tsunami Workshop 2008
HHT Analysis: Seven Intrinsic Mode Functions (IMF) for pressure during 21:10 － 21:27 in Kaohsiung South China Sea Tsunami Workshop 2008
Nine IMF’s for longshore flow during 22:10 － 22:27 in Kaohsiung South China Sea Tsunami Workshop 2008
Tsunami pressures and their marginal spectra at Kaohsiung during 21:10 － 21:27 local time South China Sea Tsunami Workshop 2008
Tsunami pressures and their marginal spectra at Kaohsiung during 22:10 － 22:27 local time South China Sea Tsunami Workshop 2008
Detailed simulation near the Kaohsiung shows the reflection of the local bathymetry and the breakwater is not serious South China Sea Tsunami Workshop 2008
If waves are generated far away along the coast, their direction is easy to determine Although reflection of the shoreline is possible, these waves are basically progressive. South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
Estimate the arrival time to find out where the edge wave is generated • The average bottom slope along the southwest coast of Taiwan is estimated • The average bottom slope from KH to DG is 1/140, from DG to FS is 1/150, from FS to SGZ is 1/70 • Hence the arrival time of edge waves can be estimated South China Sea Tsunami Workshop 2008
South China Sea Tsunami Workshop 2008
Estimated arrival time of mode 2 edge wave at Kaohsiung is 21:50 South China Sea Tsunami Workshop 2008
Future Works • Analyzing field measurement at other stations • Analyzing the simulated flow field South China Sea Tsunami Workshop 2008
Conclusion • Edge wave is a significant component of the tsunamis generated by the Ping-tong earthquake • The southwestern Taiwan should be prepared for both direct tsunami and edge wave tsunami components in a future tsunami South China Sea Tsunami Workshop 2008
More Discussions • The present flow-pressure approach needs only a single instrument and hence makes it much easier and cheaper to measure trapped tsunami waves. Extension of the present approach, therefore, will greatly improve the availability of information about trapped tsunamis. • To analyze a tsunami by HHT, we can choose the period of time when the tsunami dominates the infragravity wave field so that the analysis is not contaminated by infragravity waves not South China Sea Tsunami related to the tsunami. Workshop 2008
Limitation of the method • The flow-pressure approach assumes only one wave mode exists at one time. Therefore, if two or more modes coexist, analyzing the relation between pressure and longshore velocity cannot give useful results. However, this approach provides the best estimate of the oscillation mode, especially when the number of instrument is limited. South China Sea Tsunami Workshop 2008
Recommend
More recommend
