C. Cai

Peking University, Beijing, Beijing Shi, China

Are you C. Cai?

Claim your profile

Publications (7)0 Total impact

  • Source
    Yuan Tian · Jieyuan Ning · Chunquan Yu · Chen Cai · Kai Tao
    [Show abstract] [Hide abstract]
    ABSTRACT: The 2008 Wenchuan earthquake, a major intraplate earthquake with M w 7.9, occurred on the slowly deforming Longmenshan fault. To better understand the causes of this devastating earthquake, we need knowledge of the regional stress field and the underlying geodynamic processes. Here, we determine focal mechanism solutions (FMSs) of the 2008 Wenchuan earthquake sequence (WES) using both P-wave first-motion polarity data and SH/P amplitude ratio (AR) data. As P-wave polarities are more reliable information, they are given priority over SH/P AR, the latter of which are used only when the former has loose constraint on the FMSs. We collect data from three categories: (1) permanent stations deployed by the China Earthquake Administration (CEA); (2) the Western Sichuan Passive Seismic Array (WSPSA) deployed by Institute of Geology, CEA; (3) global stations from Incorporated Research Institutions for Seismology. Finally, 129 events with magnitude over M s 4.0 in the 2008 WES are identified to have well-constrained FMSs. Among them, 83 are well constrained by P-wave polarities only as shown by Cai et al. (Earthq Sci 24(1):115–125, 2011), and the rest of which are newly constrained by incorporating SH/P AR. Based on the spatial distribution and FMSs of the WES, we draw following conclusions: (1) the principle compressional directions of most FMSs of the WES are subhorizontal, generally in agreement with the conclusion given by Cai et al. (2011) but with a few modifications that the compressional directions are WNW–ESE around Wenchuan and ENE–WSW around Qingchuan, respectively. The subhorizontal compressional direction along the Longmenshan fault from SW to NE seems to have a left-lateral rotation, which agrees well with regional stress field inverted by former researchers (e.g., Xu et al., Acta Seismol Sin 30(5), 1987; Acta Geophys Sin 32(6), 1989; Cui et al., Seismol Geol 27(2):234–242, 2005); (2) the FMSs of the events not only reflected the regional stress state of the Longmenshan region, but also were obviously controlled by the faults to some extent, which was pointed out by Cai et al. (2011) and Yi et al. (Chin J Geophys 55(4):1213–1227, 2012); (3) while the 2008 Wenchuan earthquake and some of its strong aftershocks released most of the elastic energy accumulated on the Longmenshan fault, some other aftershocks seem to occur just for releasing the elastic energy promptly created by the 2008 Wenchuan earthquake and some of its strong aftershocks. (4) Our results further suggest that the Longmenshan fault from Wenchuan to Beichuan was nearly fully destroyed by the 2008 Wenchuan earthquake and accordingly propose that there is less probability for great earthquakes in the middle part of the Longmenshan fault in the near future, although there might be a barrier to the southwest of Wenchuan and it is needed to pay some attention on it in the near future.
    Earthquake Science 12/2014; 26(6):357-372. DOI:10.1007/s11589-014-0067-y
  • S. Ning · F. Niu · K. Tao · C. Cai · J. Ning
    [Show abstract] [Hide abstract]
    ABSTRACT: The present-day tectonic setting of China is featured by Indian-Eurasian collision in the west and Pacific subduction in the east. The interaction of the Eurasian, Indian, and Pacific Plates has resulted in a unique topographic contrast between eastern and western China. The two regions are bounded by a ~100 km wide, NNE trending lineament known as the North South Gravity Lineament (NSGL). Motion style of the upper mantle around this region is essential for understanding the evolution of China continent. Seismic waves in an anisotropic media travel at different speeds depending on their propagation and polarization directions. In the upper mantle, it is generally believed that seismic anisotropy is caused by a preferred orientation of olivine crystal. The fast direction is parallel to the maximum shear and maximum extension directions for simple shear and pure shear, respectively, and accordingly relates with the motion direction of the upper mantle. We employed a multi-event stacking method to measure the fast direction (φ) and splitting time (δt) beneath NSGL with SKS waveform data recorded at epicentral distances of 90°-130°. A total of 219 regional seismic stations operated by the China Earthquake Administration (CEA) were used in this study. These stations covered the continental scarp that extends from the Great Xing'an Range in the northeast to the eastern edge of the Yunnan-Guizhou (Yungui) plateau in the south. From a total of 100 events, we chose 40 earthquakes with high SNR for analyzing. The measured seismic anisotropy exhibited large variations in both φ and δt along the strike. The fast direction starts from nearly NS at the NE end of the Great Xing'an Range, changes to almost EW along the Yanshanian Orogenic Belt. It then rotates to NE through the Taihang mountain range, and changes back to EW within the Qinling mountain range. Along the eastern border of the Sichuan basin, the fast direction matches well with the NNE oriented mountain ranges in the region. At the eastern and southeastern corner of the Yungui plateau, we observed a nearly EW fast direction. Although the large variation in seismic anisotropy is difficult to be explained by a single deformation process, our results imply that there should be convection in the big mantle wedge beneath eastern China continent.
  • H. Yu · K. TAO · C. YU · C. Cai · X. Zheng · J. Ning
    [Show abstract] [Hide abstract]
    ABSTRACT: We mainly employ centroid-moment tensor solutions (CMTs) for the March 11, 2011 Tohoku Earthquake and its aftershocks reported by Harvard University in studying deformation and stress state around Japan Sea area. After confirming the validity of CMTs by P-wave first motion polarity data from IRIS, we do cluster analysis and find that the solutions can be divided into three groups. The first group events are the kind of the main shock which is a low-angle thrust event, mainly occurred on the interface between the Pacific Plate and the North American Plate. The second group events are normal-fault earthquakes with principal extensional directions roughly pointing East-West direction. Most of these earthquakes located in the forearc uplift region of the Pacific plate. Some others occurred in the forearc accretionary wedge. The third group events are also normal-fault earthquakes though their principal extensional directions are roughly along the direction of the trench. These events located to the west of the Japan trench. We employ finite element method to simulate the stress state and deformation revealed by the focal mechanism solutions as well as GPS observations. Results show that the March 11, 2011 Tohoku Earthquake and some of its aftershocks released much of the East-West directional compressional stress accumulated in hundreds of years and resulted in at least a temporary East-West tensile state in the Japan Sea region which extends to China mainland. The normal-fault aftershocks located to the west of the Japan trench with East-West principal extensional direction were produced by this stress condition, while the existence of the normal-fault aftershocks with North-South principal extensional direction expressed the values of the tension along North-South direction and along East-West direction are roughly equal.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We relocate the spatial distribution of the devastating 12 May 2008 Wenchuan earthquake and its aftershocks. The relocation database is obtained from 89 stations deployed by the China Earthquake Administration, including 54 525 seismograms from 1 376 local earthquakes over M S 3.5 between 12 May 2008 and 3 August 2008. The cross-correlation technique used in this paper has greatly improved the relocation precision by giving much more accurate P-wave differential travel-time measurements than those obtained from routinely picked phase onsets. At the same time, we pick P-wave polarity observations of the Wenchuan earthquake series (hereafter referred to as WES) from 1 023 stations in China and 59 IRIS (incorporated Research Institutions of Seismology) stations. Then, employing a newly developed program CHNYTX, we obtain 83 well-determined focal mechanism solutions (hereafter referred to as FMSs). Based on spatial distribution and FMSs of the WES, we draw following conclusions: (1) The region near the main shock exhibits a buried low-angle northwest-dipping seismic zone with the main shock at its upper end and two conjugated seismic zones dipping southeast with roughly equal dip-angle; (2) The compressional directions of all kinds of FMSs of the WES are subhorizontal, which reflects the dominant stress in this area is compressional; (3) The principal compressional direction of the regional stress around Wenchuan is roughly perpendicular to the strike of Beichuan-Yingxiu fault, while around Qingchuan it is roughly parallel to the strike of Qingchuan fault. In intermediate part of the Longmenshan area, the principal compressional direction of the stress should be in-between; (4) The possibly existed molten materials in the lower crust of Songpan-Garze terrain have small contribution to the local stress state in Longmenshan area. The listric geometries of the Longmenshan faults most probably resulted from subhorizontal compression along NW-SE direction in history. Key wordsseismotectonics–stress state–relocation–focal mechanism solution–Wenchuan
    Earthquake Science 02/2011; 24(1):115-125. DOI:10.1007/s11589-011-0775-5
  • [Show abstract] [Hide abstract]
    ABSTRACT: We employed a double-difference algorithm (hypoDD) to relocate earthquakes within the region bounded by 66°E–78°E and 32°N–42°N in the period of 1964–2003 reported by the International Seismological Center (ISC). The improved hypocentral locations delineate a double-layered Wadati-Benioff zone in the eastern Hindu Kush intermediate seismic belt. Based on this feature and other evidences, we propose that the intermediate-depth earthquakes beneath the Pamir-Hindu Kush region may occur in two collided subduction zones with opposite dip directions.
    Earthquake Science 12/2009; 22(6):659-665. DOI:10.1007/s11589-009-0659-0
  • C. Yu · C. Cai · K. Tao · J. Liu · J. Ning
    [Show abstract] [Hide abstract]
    ABSTRACT: We have relocated the 2008 Wenchuan earthquake and its strong aftershocks using a double-difference algorithm (HypoDD). The relocation database is obtained from 89 earthquake stations, including 54525 seismograms from 1376 local earthquakes over Ms3.5 between 12/5/2008 and 3/8/2008. To improve phase pickup accuracy, we employ cross-correlation technique in our computation. Totally we get 80823 P-wave differential times with cross-correlation coefficients greater than 0.7. The P-wave differential times are mostly more than ten times more accurate than those obtained from routinely picked phase onsets and supply high-quality input for our relocation computation. To reduce the errors caused by homogenous velocity model, we adopt more reasonable different velocity models for the two sides of Longmenshan fault zone. Our results are obviously better than former results in presenting well coherence and showing clear fault image, which provide additional insight regarding the relationship between the Longmenshan Fault Series and their nearby seismicity. P-wave first motion, an unambiguous physical quantity, is a kind of steady seismic information. When enough accurate data were available, it can be used to ideally determine focal mechanism solutions. Yu et al. (2009) improved former grid search methods. Firstly, possible solutions are selected by weighted inconsistency ratio. The weighting scheme retains the data-quality weighting factor so as to consider the influence of data quality, omits the weighting factor reflecting the distance between data points and nodal planes for avoiding over-reducing the weights of the observations near nodal planes, and adds another weighting factor which gives more power to sparse data points on the focal sphere for partly canceling out the influence of uneven distribution of the employed data points. Meanwhile, they divided the focal sphere into several roughly equal-square areas, which prevent deviation from real mean solution by artificial concentration of test solutions. They improved the inversion quality by employing jackknife technique, which enlarges the solution set by adding those possible solutions for one observation being ignored. This technique, along with clustering analysis, not only increases the possibility of finding out true solutions, but also makes us comprehend the quality of focal mechanism solutions unambiguously. Yu et al. (2009) also provided a new scheme for evaluating the quality of focal mechanism solutions based on the dispersion of selected solutions as well as the minimum weighted inconsistency ratio. Our relocation results provide us with more accurate taking-off angles of P arrivals and the opportunity to get more accurate P to S amplitude ratios, which greatly improve the quality of focal mechanism solutions. We will show a group of new fault plane solutions of the great Wenchuan earthquake and its strong aftershocks and will discuss their implications in this meeting. Yu C Q, Tao K, Cui X F, et al. P-wave first-motion focal mechanism solutions and their quality evaluation. Chinese J. Geophys. (in Chinese), 2009, 52(5): 1402~1411.
  • Source
    J. Ning · X. Lou · C. Cai · C. Yu
    [Show abstract] [Hide abstract]
    ABSTRACT: We employed a double-difference algorithm (hypoDD) to relocate the Earthquakes reported by the International Seismological Center within the region bounded by 66~78°E and 32~42°N between 1964 and 2003. Among the listed 10224 events in the catalog, 7655 events have at least six P-wave arrival times recorded by 279 stations in the region within 60~90°E and 20~50°N. Totally we have about 135,000 P wave arrival picks and 42,000 S wave arrival picks. 269,365 P-phase pairs and 212,354 S-phase pairs are selected. The average offset between linked events is 10.74 km. The double-difference travel time match in the hypoDD program retains 6018 out of the 7655 events. Then 4751 events are grouped into 182 clusters recorded by 80 stations. The other 1267 events are outliers. Finally 2134 events are successfully relocated and 1479 of them have depth greater than 70 km. There is a distinct feature beneath Hindu Kush region, a double-layered Wadati-Benioff zone which has never been revealed before. Both layers are composed of two parts: the upper part and the lower part. However, the Wadati-Benioff zone in the Pamir region is totally different: it does not have double-layered structure. The Wadati-Benioff zone beneath the Hindu Kush region and the one beneath the Pamir region meet with each other at depth of about 130 km and form back-to-back bow shapes at the boundary region. This explicit feature not only gainsays the statement that a gap exists between the Wadati-Benioff zones beneath the Hindu Kush and the Pamir, but also gainsays the statement that the two Wadati-Benioff zones have geometrical coherence. Based upon above results, along with Harvard CMT solutions, seismological tomography results and geochemical evidences, we propose that beneath the Pamir-Hindu Kush region there exist two oppositely subducted slabs which are colliding with each other at depth of about 130 km. This new model is different from the tear model. It can reasonably explain the Harvard CMT solutions. It naturally relates the intermediate-depth earthquakes under the Pamir with both the intense shallow earthquakes along the northern boundary of the Pamir in Tajikistan and the Ophiolite belt facing south in northern Pamir zone. It also helps us find a way out of the difficulty that Punjab Wedge could not supply enough material diversely subducting into the deep Earth's interior of Pamir region although it can exert strong extrusion to Pamir zone. When the plate coming from south does not have enough material deeply subducting into the Pamir region, south-concaved northern Pamir arc is indeed the right place for supplying enough material. This new model is also different from the traditional opposing subduction model. It is not based on chanciness. It supplies a reasonable solution to overcome the difficulty that we will face when we find that the seismic gap between the two Wadati-Benioff zones does not exist. When there are no strong evidences for the traditional opposing subduction model, ``collision'' is the key to reconcile all observations.