Porosity profile within the Taiwan Chelungpu Fault, reconstructed from X-ray computed tomography images

Transcription

1 JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, Report Porosity profile within the Taiwan Chelungpu Fault, reconstructed from X-ray computed tomography images Tetsuro Hirono, Weiren Lin 2, En-Chao Yeh 3, Wonn Soh 2, Masafumi Murayama 4 Chien-Ying Wang 5 and Sheng-Rong Song To evaluate the porosity profile within the Chelungpu Fault, which slipped during the 1999 Taiwan Chi-Chi Earthquake, we performed X-ray computed tomography imaging of core samples from Hole B of the Taiwan Chelungpu Fault Drilling Project. We established the relationship between the degree of X-ray attenuation and wet bulk density and obtained a porosity profile through the major fault zone at 1,136 m depth. The shear zones within the fault zone generally showed high porosity, and the porosity of the shear in the black gouge zone, which previous research suggests was the slip plane for the 1999 earthquake, was deduced to be 33.0%. This high porosity might indicate dilative shear deformation during the earthquake. Keywords : X-ray CT, CT number, porosity, Chelungpu Fault, Chi-Chi Earthquake Received 15 December 2008 ; accepted 12 May Department of Earth and Space Science, Graduate School of Science, Osaka University, Toyonaka, Japan 2 Kochi Institute for Core Sample Research, Japan Agency for Marine-Earth Science and Technology, Nankoku, Japan 3 Department of Geosciences, National Taiwan University, Taiwan 4 Center for Advanced Marine Core Research, Kochi University, Japan 5 Institute of Geophysics, National Central University, Taiwan Corresponding author: Tetsuro Hirono Department of Earth and Space Science, Graduate School of Science, Osaka Universit. Toyonaka , Japan hirono@ess.sci.osaka-u.ac.jp Copyright by Japan Agency for Marine-Earth Science and Technology

2 Porosity profile reconstruction from XCT number 1. Introduction The 1999 Taiwan Chi-Chi Earthquake (Mw 7.6) occurred on 21 September 1999, with its epicenter at lat N, long E, and a focal depth of 8 km (Ma et al., 1999) (Fig. 1). The earthquake initiated from a hypocenter in the southern part of the Chelungpu Fault and ruptured both up-dip and laterally northward (Chen et al., 2001; Kikuchi et al., 2000; Ma et al., 2000). The faulting produced surface ruptures over a distance of about 100 km along the Chelungpu Fault, with the largest net slip in the north, where it was up to 11.5 m (Lee et al., 2003). The fault motion was thrusting with a left-lateral component. The Chelungpu Fault is one of the western deep thrusts in the Taiwan mountain belt (Fig. 1). The Taiwan Chelungpu Fault Drilling Project (TCDP) was undertaken in 2002 to gain an understanding of the physics and faulting mechanism of the Chi-Chi Earthquake. TCDP drilled two cored holes, Hole A (total depth 2, m) and Hole B (total depth 1, m). In Hole B, cores were recovered only from between and 1, m. Nondestructive continuous physical property measurements and mesoscopic observations of the cores were performed at the Kochi Core Center in Japan. Hirono et al. (2006, 2007) described three main fault zones within the Chinshui Shale in the Hole B core samples: FZB1136 (fault zone around 1,136 m depth in Hole B), FZB1194, and FZB1243 (Fig. 2), and interpreted these as segments of the Chelungpu Fault. Different mechanisms of fault lubrication during the 1999 Chi-Chi Earthquake have been suggested, such as frictional melting, thermal pressurization, and elasto-hydrodynamic lubrication (e.g. Ma et al., 2003). All of these are greatly dependent on the quantity of pore water in the slip zone, on the thermal properties of the faulted rocks, and on fluid transport properties in the fault zone. Therefore, to clarify which of these mechanisms contributed to the 1999 earthquake, it is important to determine quantitatively the porosity in the fault zone. Hirono et al. (2006) calculated porosity and wet bulk density profiles from wet mass, dry mass, and volume (Fig. 2), but their sampling interval of approximately 1 m was too large to correlate porosity and density with features of the fault zone architecture such as fault gouge and fault breccia. Lin et al. (2008) measured the water content (which corresponded to porosity because the cores tested were completely water-saturated) through the three fault zones by time domain reflectometry (TDR) to develop a water-content profile. However, the TDR spatial resolution was 8 cm, which was not fine enough to correlate the profile with detailed structures such as shear zones and fractures. Fig. 1 Location map showing the Chelungpu Fault, regional geological map, an E-W cross section through the drill location for Hole B, and a photo of the drill site. CF: Chelungpu Fault. Modified from Hirono et al. (2006) JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

3 T. Hirono et al., In this study, we reconstructed the porosity profile of the Taiwan Chelungpu Fault by using X-ray computed tomography images. We established the relationship between the degree of X-ray attenuation and rock density and used it to determine bulk density and porosity. These data allow us to present the porosity profile of the Chelungpu Fault and discuss the deformation during the 1999 Chi-Chi Earthquake. 2. X-ray computed tomographic imaging 2.1. Principles X-ray computed tomographic (XCT) imaging is a radiological imaging technique first developed by Hounsfield (1973). The attenuation of two-dimensional X-ray fan beams penetrating a sample is measured by an array of detectors. X-ray projection data from various directions are obtained by rotating the X-ray source Fig. 2 Porosity and wet bulk density from discrete sample logs at approximately 1-m intervals in Hole B (Hirono et al., 2007). LC: lithological column. JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

4 Porosity profile reconstruction from XCT number through 360 (Fig. 3a). A two-dimensional image representing the linear distribution of X-ray attenuation is reconstructed using Fourier transformation of the projection data. The degree of X-ray attenuation depends on the density and atomic number of the materials in the samples. Materials with higher density and higher atomic number generally cause higher attenuation of X-rays. The amount of attenuation in the XCT image is expressed as the CT number (N ct ), which is defined as N ct = ( μ - μ w ) / μ w 1,000, (1) where μ is the linear X-ray absorption coefficient of the sample and μ w is the linear absorption coefficient of water, used as a standard reference. The CT number of water is defined as zero, and the CT number of air as -1,000. The CT number is a function of the density and chemical composition of the material Capture of XCT images An X-ray CT scanner (Pratico, Hitachi Medical Co., Tokyo) at the Kochi Core Center was used for this study (Fig. 3a). X-rays were produced by electrons striking a Mo-W alloy target in an X-ray tube. To determine the correlation between the degree of X-ray attenuation and density, we captured at least one slice image for every 1 m of core (Fig. 3b). The electron current was 100 ma, the accelerating voltage was 120 kv, and the scan time was 4 s. Each image was 1 mm thick, and the resolution was mm ( pixels). The output XCT images were digitized as DICOM-formatted 16-bit grayscale image files. Because X-ray equipment with a wide energy range was used in this study, the XCT images include beamhardening artifacts: X-rays with lower energy (longer wavelength) are attenuated more and penetrate a shorter distance into the sample than higher energy X-rays. However, the detector counts only the number of photons without discriminating between different energy levels (wavelengths). As a result, the outer part of the sample has an apparently higher CT number. The line profile of CT numbers across a homogeneous intact siltstone core from 1, m in Hole B (Fig. 4) shows relatively high CT numbers at the edges of the core. To exclude the effect of this artifact, only the central parts of each XCT image were used for the CT number analyses. These were calculated as the average of the CT numbers within the cm ( pixels) square shown in Fig Correlation of CT number with density 3.1. Bulk density measurements For bulk density and porosity measurements, we took discrete subsamples at intervals of 1 m from intact rocks. The depths sampled were the same as those used for the capture of XCT images. First, the wet mass, M wet, of each of the fully water-saturated subsamples was measured on an electronic balance with a precision of ± g. Then the subsamples were oven-dried at 105 ± 5 C for 24 h and allowed to cool in a desiccator before the dry mass, M dry, and dry volume, V dry, were measured. The dry volume of each subsample was determined at least three times with a helium-displacement pycnometer (Quantachrome Penta- Pycnometer) with a nominal precision of ± 0.01 cm 3, and then averaged. Pore volume, V pore, was calculated by subtracting dry mass from wet mass, assuming a constant Fig. 3 (a) Medical X-ray computed tomography scanner. (b) Geometry of X-ray CT image slices from a core sample. JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

5 T. Hirono et al., 1.0 g/cm 3 density for the pore fluid evaporated during drying. Then porosity, ø, and wet bulk density, bulk, were calculated as ø = { V pore / ( V pore + V dry ) } 100, and (2) bulk = M wet / ( V pore + V dry ). (3) Porosity and density were accurate to within 0.1% and 0.01 g/cm 3, respectively. The resultant porosity and bulk density profiles with depth are shown in Fig Lithology of samples In TCDP Hole B (Fig. 2), the depth interval from to 1,040 m represents the Pliocene to Pleistocene Cholan Formation, which is composed predominantly of sandstone with some sandstonesiltstone alternations that are weakly to heavily bioturbated. The Pliocene Chinshui Shale extends from 1,040 to 1,280 m depth and consists predominantly of siltstone with weak bioturbation. The interval from 1,280 to 1, m depth is the late Miocene to early Pliocene Kueichulin Formation, which is composed predominantly of massive sandstone with minor amounts of siltstone. To correlate CT number with bulk density, we determined two lithological categories: sandstone, in which we included sandstone-dominant sandstonesiltstone alternations; and siltstone, in which we included siltstone-dominant siltstone-sandstone alternations Correlation We determined the averaged CT numbers for 423 subsamples from Hole B core samples, and classified each sample as either sandstone or siltstone (as categorized above). Then we plotted the CT numbers with wet bulk density (Fig. 5). For the rocks categorized as sandstone, we determined the relationship between CT number and bulk density to be N ct = 263 bulk + 2,362. (4) However, the correlation coefficient was 0.29, which might be attributable to the inclusion of a variety of mineral grains of different densities. On the other hand, we determined the relationship between CT number and density in the rocks categorized as siltstone to be N ct = 937 bulk (5) The correlation coefficient here was 0.73, which was better than that for the sandstone category. Fig. 4 (a) X-ray CT image of an intact homogeneous siltstone sample from 1, m in Hole B. The black square indicates the area used to calculate the averaged CT number. (b) Profile of CT numbers across the diameter of the core sample. Fig. 5 Cross plots of wet bulk density and porosity on discrete samples at approximately 1-m intervals from Hole B core samples. JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

6 Porosity profile reconstruction from XCT number Because the fault zones of the Chelungpu Fault are developed within the Chinshui Shale (Hirono et al., 2007), we adopted the relationship of the siltstone category rocks (equation [5]) for our construction of the density profile. 4. Porosity profile within the Taiwan Chelungpu Fault 4.1. Association of FZB1136 with the 1999 Chi-Chi Earthquake Although Hirono et al. (2006, 2007) found three major fault zones in TCDP Hole B core samples, FZB1136 is the fault zone most likely related to the 1999 Chi-Chi Earthquake. Kano et al. (2006) measured borehole temperatures in Hole A and observed a low-amplitude heat signal around FZA1111 (fault zone around 1,111 m depth in Hole A, which correlates to FZB1136 in Hole B). They suggested that the heat signal was produced by frictional heating during the 1999 Chi-Chi Earthquake. Wu et al. (2007) found low seismic velocity, low electrical resistivity, and a major stress orientation anomaly around FZA1111 in geophysical logs from Hole A. Therefore, we focused only on FZB1136 in this study Bulk density profile We captured continuous sequential XCT images of core samples from 1, to 1, m in FZB1136. The XCT settings were 100-mA electron current, 120-kV accelerating voltage, and 4-s scan time. The image slices were captured at 1-mm intervals, and each image was of a 1-mm thickness of core. The resolution of the images was mm. Stacked images were produced with VGStudio MAX software (Volume Graphics GmbH, Heidelberg, Germany). The XCT image in Fig. 6 is a cross section in a plane parallel to the direction of dip of the shear plane. We counted the CT number at the center area of the XCT image for each depth at 1-mm intervals and calculated bulk density by using equation (5). The reconstructed bulk density profile is shown in Fig Porosity profile The relationship of porosity to wet bulk density and grain density, grain, is defined as follows: ø = ( bulk - grain ) / ( pore fluid - grain ) 100, (6) where pore fluid is the density of pore fluid, which we assumed to be 1.0 g/cm 3. In the Chinshui Shale, the average grain was calculated to be 2.72 g/cm 3, which we used for the porosity calculation. The reconstructed porosity profile is shown in Fig Discussion and conclusions The general trends of the porosity profile revealed by XCT and the TDR porosity profile reported by Lin et al. (2008) compare well. Porosities within the black gouge zone are higher than those in the breccia zone and fracture-damaged zone for both the XCT and TDR porosities. However, the TDR porosities in the black gouge zone are approximately 30%, whereas the porosities reconstructed from XCT images are around 15-25%. One possible cause of these differences is the difference in vertical resolution of the two profiles. As previously discussed, the vertical resolution of the TDR profile is 8 cm, whereas that of the XCT profile is 1 mm. Another possibility is that the correlations between CT number and wet bulk density, and between density and porosity, were not high enough. The correlation coefficient between the CT numbers and densities was calculated to be 0.73, and the grain density was assumed to be 2.72 g/cm 3. The chemical composition of the samples likely affected the correlation between CT numbers and bulk densities. The CT number depends not only on the density of a sample but also on its chemical composition. High atomic number is generally associated with high CT number. However, Ishikawa (personal communication) used X-ray fluorescence spectroscopy for chemical analyses of core samples from the major fault zones in Hole B to show that there were not large differences in the major element contents of the gouge zone, breccia zone, fracture-damaged zone, and host rock (e.g. in the black gouge zone, SiO %, TiO 2 0.9%, Al 2 O %, Fe 2 O 3 6.4%, MnO 0.0%, MgO 2.5%, CaO 0.8%, Na 2 O 1.9%, K 2 O 2.9%, and P 2 O 5 0.0%; in host rock at 1, m depth, SiO %, TiO 2 1.0%, Al 2 O %, Fe 2 O 3 6.4%, MnO 0.0%, MgO 2.2%, CaO 1.4%, Na 2 O 1.5%, K 2 O 3.2%, and P 2 O 5 0.0%; all concentrations are in weight percent). Therefore, the loss of atoms with high atomic number is not likely the cause of the low CT numbers. Although the calculated XCT porosity should be regarded as only semiquantitative, a notable advantage JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

7 T. Hirono et al., of this method is the high vertical resolution (1 mm). This resolution allows detailed correlation of the porosity profile with fine structures. We determined the XCT porosity of a 2-cm-thick major slip zone thought to be associated with the 1999 Chi-Chi Earthquake (Ma et al., 2006) to be 33.0% (Fig. 6). High porosity in the slip zone can indicate dilative shear deformation during an earthquake. Lin et al. (2008) also suggested that fault zone materials might become noncohesive and that their porosity might increase immediately after a largedisplacement slip. Finally, we emphasize that the reconstruction of a fault zone porosity profile from XCT imagery presented here is the first attempt to do so, and it may provide a valid tool for detailed correlation of porosity with microstructure in shear and/or slip zones. Fig. 6 Photo image and sketch of core, X-ray CT image, and depth profiles of CT number, wet bulk density, and porosity for FZB1136. TDR porosity data from Lin et al. (2008) is also shown. BGZ: black gouge zone; GGZ: gray gouge zone; BZ: breccia zone; FDZ: fracture-damaged zone; MSZ: a 2-cm-thick major slip zone thought to be associated with the 1999 Chi-Chi earthquake, identified by Ma et al. (2006). JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22

8 Porosity profile reconstruction from XCT number Acknowledgments We would like to thank Kuo-Fong Ma and Jih- Hao Hung for their support and for their analyses of Hole B core samples. We also thank Kohtaro Ujiie and Kan Aoike for their constructive reviews. References Chen, K. C., B. S. Huang, J. H. Wang, W. G. Huang, T. M. Chang, R. D. Hwang, H. C. Chiu, and C. P. Tsai (2001), An observation of rupture pulses of the 20 September 1999 Chi-Chi, Taiwan, earthquake from near-field seismograms, Bulletin of the Seismological Society of America, 91, Hirono, T., W. Lin, E. Yeh, W. Soh, Y. Hashimoto, H. Sone, O. Matsubayashi, K. Aoike, H. Ito, M. Kinoshita, M. Murayama, S. Song, K. Ma, J. Hung, C. Wang, and Y. Tsai (2006), High magnetic susceptibility of fault gouge within Taiwan Chelungpu Fault: Nondestructive continuous measurements of physical and chemical properties in fault rocks recovered from Hole B, TCDP, Geophysical Research Letters, 33, L15303, doi: /2006gl Hirono, T., E. Yeh, W. Lin, H. Sone, T. Mishima, W. Soh, Y. Hashimoto, O. Matsubayashi, K. Aoike, H. Ito, M. Kinoshita, M. Murayama, S. Song, K. Ma, J. Hung, C. Wang, Y. Tsai, T. Kondo, M. Nishimura, S. Moriya, T. Tanaka, T. Fujiki, L. Maeda, H. Muraki, T. Kuramoto, K. Sugiyama, and T. Sugawara (2007), Nondestructive continuous physical property measurements of core samples recovered from Hole B, Taiwan Chelungpu Fault Drilling Project, Journal of Geophysical Research, 112, B07404, doi: /2006jb Hounsfield, G. N. (1973), Computerized transverse axial scanning (tomography), British Journal of Radiology, 46, Kano, Y., J. Mori, R. Fujio, H. Ito, T. Yanagidani, S. Nakao, and K. F. Ma (2006), Heat signature on the Chelungpu Fault associated with the 1999 Chi- Chi, Taiwan, earthquake, Geophysical Research Letters, 33, L14306, doi: /2006gl Kikuchi, M., Y. Yagi, and Y. Yamanaka (2000), Source process of the Chi-Chi, Taiwan, earthquake of September 21, 1999 inferred from teleseismic body waves, Bulletin of the Earthquake Research Institute, University of Tokyo, 75, Lee, Y., M. Hsieh, S. Lu, T. Shih, W. Wu, Y. Sugiyama, T. Azuma, and Y. Kariya (2003), Slip vectors of the surface rupture of the 1999 Chi- Chi Earthquake, western Taiwan, Journal of Structural Geology, 25, Lin, W., O. Matsubayashi, E. Yeh, T. Hirono, W. Tanikawa, W. Soh, C. Wang, S. Song, and M. Murayama (2008), Profiles of volumetric water content in fault zones retrieved from Hole B of the Taiwan Chelungpu Fault Drilling Project (TCDP), Geophysical Research Letters, 35, L01305, doi: /2007gl Ma, K. F., E. E. Brodsky, J. Mori, T. A. Song, and H. Kanamori (2003), Evidence for fault lubrication during the 1999 Chi-Chi, Taiwan, earthquake (M w 7.6), Geophysical Research Letters, 30, 1244, doi: /2002gl Ma, K. F., C. T. Lee, Y. B. Tsai, T. C. Shin, and J. Mori (1999), The Chi-Chi, Taiwan, earthquake: Large surface displacements on inland thrust fault, EOS, 80, Ma, K. F., T. R. Song, S. J. Lee, and S. I. Wu (2000), Spatial slip distribution of the September 20, 1999, Chi-Chi, Taiwan, earthquake: Inverted from teleseismic data, Geophysical Research Letters, 27, Ma, K. F., H. Tanaka, S. Song, C. Wang, J. Hung, Y. Song, E. Yeh, W. Soh, H. Sone, L. Kuo, and H. Wu (2006), Slip zone and energetics of a large earthquake from the Taiwan Chelungpu Fault Drilling Project, Nature, 444, Wu, H., K. Ma, M. Zoback, N. Boness, H. Ito, J. Hung, and S. Hickman (2007), Stress orientations of Taiwan Chelungpu Fault Drilling Project (TCDP) Hole A as observed from geophysical logs, Geophysical Research Letters, 34, L01303, doi: /2006gl JAMSTEC Rep. Res. Dev., Volume 9 Number 2, September 2009, 15 22