Identification of multiple faulting events of the Median Tectonic Line active fault system in the Tokushima Plain, Japan, based on close-interval radiocarbon dating

Identification of multiple faulting events of the Median Tectonic Line active fault system in the Tokushima Plain, Japan, based on close-interval radiocarbon dating
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Identification of multiple faulting events of theMedian Tectonic Line active fault system in theTokushima Plain, Japan, based on close-intervalradiocarbon dating Toshimichi Nakanishi  a,* , Keiji Takemura  b , Atsumasa Okada  a ,Michio Morino  c , Akira Hayashida  d , Masanobu Nakamura  e , Yuji Tazawa  e ,Koya Ogino  f  , Hiroshi Matsumoto  e , Masanori Hirose  e a Department of Geophysics, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan b Institute for Geothermal Sciences, Graduate School of Science, Kyoto University, Noguchibaru, Beppu, Oita 874-0903, Japan c OYO Corporation, Toro-cho 2-61-5, Kita-ku, Saitama 331-0804, Japan d Science and Engineering Research Institute, Doshisha University, Kyoto 610-0321, Japan e Department of Physics, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan f  Department of Nuclear Engineering, Graduate School of Engineering, Kyoto University, Kyoto 606-8501, Japan Abstract Close-interval radiocarbon dating of terrestrial macrofossils from sediment cores across a fault is an instructive toolfor identifying timing of multiple faulting events. We performed accelerator mass spectrometric (AMS)  14 C measure-ments and high-resolution sedimentological analysis. At least six paleoseismic events of the Median Tectonic Lineactive fault system in the Tokushima Plain, Japan, are detected during the last 16000 years: around 850 cal BP, 2850 calBP, 5300 cal BP, 7500 cal BP, 9000 cal BP, 16000 cal BP.   2004 Elsevier B.V. All rights reserved. PACS:  83.70.H; 91.30; 91.65; 93.85 Keywords:  Paleoseismology ; Sedimentology; AMS  14 C dating; Holocene 1. Introduction Most Holocene depositional basins in Japanhave evolved in association with sea-level trans-gression and tectonic movement such as faulting.Sediments of these basins along active faults re-cord patterns of historical surface-rupturingearthquakes. Based on high-density radiocarbondating of sediments from boreholes across a fault,it is possible to identify timing of these faultingevents [1].The Median Tectonic Line active fault system(MTL) is a 300-km-long, arc-parallel, right-lateralstrike-slip fault related to oblique subduction of the Philippine Sea plate beneath the Eurasian plate * Corresponding author. Tel.: +81-75-753-3945; fax: +81-75-753-4291. E-mail address: (T. Nakanishi).0168-583X/$ - see front matter    2004 Elsevier B.V. All rights reserved.doi:10.1016/j.nimb.2004.04.106Nuclear Instruments and Methods in Physics Research B 223–224 (2004) 573–  (Fig. 1). The MTL is the longest continuous activefault system onshore Japan with one of the highestlate Quaternary slip rates [2]. It has 1000–3000years earthquake recurrence interval [3]. Here, wereport activity of the MTL during the late Pleis-tocene to the Holocene in the Tokushima Plain(Fig. 1). 2. Methods Under the Tokushima Plain, vertical displace-ment along MTL is recorded in ca. 40-m-thickHolocene sediments [4]. Three 50–75-m-long sedi-ment cores (D-1, D-2 and D-4) were obtained fromthe present delta plain crossing the Naruto-minamifault [5], which is part of the MTL. Using D-2 andD-4 sediment cores taken on both sides of thefault, we detected vertical displacement along theNaruto-minami fault based on sedimentologicalmarker horizons and the depositional ages.  2.1. High-resolution sedimentological analysis In order to estimate depositional environmentsof these sediment cores, we conducted continuousand close-interval analyses of lithology (cm-order),initial magnetic susceptibility (2-cm-interval),grain size (10-cm-interval) and composition of thevery fine sand (v.f.s.) fraction (20-cm-interval).Initial magnetic susceptibility values were mea-sured at the laboratory of Prof. Hayashida inDoshisha University. Grain-size characteristicswere determined on each approximately 10 g and10-cm-thick slice of the cores using 0.063 and 0.125mm sieves. Using these separated samples, modalanalysis of the v.f.s. fraction including counting(normally 200 grains) of volcanic glass, bioclasts,plant fragments, heavy minerals, light mineralsand other grains was carried out (followed [6]).Some of the results were already shown in [7].  2.2. AMS   14 C dating  In order to establish a stratigraphically consis-tent chronology of D-2 and D-4 cores, we con-ducted AMS  14 C measurements on 31 samples,and added to previously reported by [8].Terrestrial macrofossils were chosen for  14 Cdating as far as available, to avoid ‘‘marine res-ervoir effects’’, and  14 C soil dating is too complex.Suitable samples were selected on sedimentlogicalobservation, and washed repeatedly with ultra-sonic cleaner and then cleaned chemically byacid–alkali–acid or acid treatments to removesecondary contamination. These samples weresealed in an evacuated Vycor tube with CuO, andcombusted in an electronic furnace. The resultingCO 2  was purified cryogenically in a high-vacuumpreparation system and then converted to graphiteby reducing the CO 2  on Fe-powder with hydrogengas in a sealed Vycor tube [9]. The sample pre-treatment was conducted at the laboratory of Prof.Kitagawa in Nagoya University. Fig. 1. Shaded relief map of the Median Tectonic Line activefault system (MTL) in eastern Shikoku and southern KinkiDistricts, and geomorphology and the location of coring sitesalong the MTL in northern Tokushima Plain (modified after[7]). Location of core sites is from [4].574  T. Nakanishi et al. / Nucl. Instr. and Meth. in Phys. Res. B 223–224 (2004) 573–578  14 C ages of the samples were measured alongwith the standard (NIST HOx II) at the KyotoUniversity AMS facility [10,11]. We corrected forcarbon isotopic fractionation via  d 13 C measuredon the AMS facility. We calculated  14 C ages fol-lowing [12], and converted them into calendardates using OxCal v3.8 [13–16]. 3. Results The upper side of Fig. 2 shows simplifiedlithofacies of D-1, D-2, and D-4 across the Nar-uto-minami fault. The upper Mesozoic basementrocks thrust over the upper Pleistocene deposits ataltitude of   ) 64.84 m in D-2. This fact and locationof surficial fault scarp suggest that Naruto-minamifault dips 69   northward. The sediment cores canbe divided into nine depositional units, cultivatedsoil and A to H units from top to bottom, on thebasis of sediment facies. Lithofacies, remarks,depositional environments, and ages of units A toH are given in Table 1.We correlated D-2 and D-4 cores, and detectedsix syntectonic sediments that are thicker on thefootwall side of the fault, based on high-resolu-tion sedimentological analysis (Fig. 2). A cleartephra horizon identified by the high concentra-tion of volcanic glass. This horizon is correlatedwith the widespread Kikai–Akahoya tephra (K-Ah; [17]: 7330 cal BP; [18]). Deeper markerhorizons show greater depth separation. Since theother part seems to have been deposited hori-zontally, and since any compaction has beenhorizontally uniform, we assume that the amountof vertical offset is due wholly to vertical faultmovement.The ages of these syntectonic sediments wereestimated from age–altitude plots (accumulationcurves) for each side of the fault (Fig. 3). We ex-pect that the faulting occur with the base of theeach sediment, because accommodation space forsediment is required. These accumulation curvesagree with this sedimentary facies-based correla-tion (Table 1). For example, D-2 and D-4 sedi-ments of a slow depositional period around K-Ahage contain the benthic foraminifers, we inter-preted these sediments formed simultaneously onprodeltaic environment. A timing of abrupt accu-mulation correlates well with relative sea-levelcurve in the Osaka Bay area [19].On the heavy sedimentation periods such asthe ages of syntectonic sediments 4 and 5, close-interval radiocarbon dating becomes most effectivetool for detection of paleoearthquakes. Moreover,we can estimate precise fault recurrence interval.These results prompted application of paleoseis-mological detection method to improve long-term Fig. 2. Correlation of cores based on sedimentological analysisacross the Naruto-minami fault (modified after [7]). See Table 1for features of the units. Shaded trapezoids show syntectonicsediments that are thicker on the footwall side of the fault.Abbreviations of indexes are K-Ah, Kikai–Akahoya tephra;v.f.s., very fine sand; sus., initial magnetic susceptibility. T. Nakanishi et al. / Nucl. Instr. and Meth. in Phys. Res. B 223–224 (2004) 573–578  575  earthquake prediction. On the other hand, longinterval and large separation in the ages of syn-tectonic sediments 3 and 6 would suggest accu-mulated faulting events. In such parts, we needmore detail analyses or investigations for thedetection.The vertical displacement plots of D-2 and D-4 suggests fault movement (Fig. 4). At least six Table 1Lithofacies, remarks, environments, and  14 C ages of units A to H (modified after [7])Unit Lithofacies Remarks DepositionalenvironmentsAge(cal BP) a A Silt to silty sand Interbedded clay Backswamp Present to 2800Including rootlet with iron rimand wood fragmentsB Fine to medium sand Coarsening upward Delta plain (topset) 2800 to 4200–5000Well sortedInterbedded coarse sand in the upperportionIncluding plant fragments andno shell fragmentBioturbatedC Interbedded very fine tofine sand and siltCoarsening upward Delta front (foreset) 4200–5000 to 8000Including plant and shell frag-mentsShell fragments and bioclastsdecrement toward the upper por-tionIncluding few foraminifers inbottomKikai–Akahoya tephraIncreasing plant fragments in theupper unitBioturbatedD Interbedded silt and clay Coarsening upward Prodelta (bottomset) 8000 to 9600Interbedded four to five shellfragments and conglomerate thinlayersShell fragments and bioclastsincrement toward the upper portionIncluding few foraminifers on topBioturbatedE Fine to medium sand Fining upward Estuary 9600 to 11000Well sortedIncluding few shell fragmentsin the lower portionPlant and shell fragments arecommonWeakly bioturbatedF Interbedded coarse sandand conglomerateCoarse sand to conglomerate Channel 11000 to 13000G Silt Several vivianite within peatybedding in the middle portionBackswamp and pond 13000 to 14500Plant fragments are commonVery thin bedded in D-2 coreH Interbedded coarse sandand conglomerateCoarse sand to conglomerate Channel 14500 to ? a Ages of units A to H based on age–altitude curves constrained by  14 C ages.576  T. Nakanishi et al. / Nucl. Instr. and Meth. in Phys. Res. B 223–224 (2004) 573–578  surface-rupturing earthquakes are detected dur-ing the last 16000 years: around 850 cal BP,2850 cal BP, 5300 cal BP, 7500 cal BP, 9000 calBP, 16000 cal BP. The average dip-slip rate of the Naruto-minami fault is estimated at 1.1 mm/year. Fig. 3. Ages of syntectonic sediments 1–6 based on age–altitude curve constrained by  14 C ages. Shaded zone shows determined faultingage distribution constrained by these sediments (Fig. 2) and age–altitude curves. We expect that the faulting occur with the base of theeach sediment, because accommodation space for sediment is required.  14 C ages are quoted with 2-sigma errors whereas the calibratedages are given in periods of which the boundaries correspond to these ages 2-sigma. The gray and black mounts represent the calibrated 14 C age probability distribution of D-2 and D-4 calculated by OxCal v3.8 [13–16]. Org., shell, and no denoted mounts show samples of organic sediments, shell, terrestrial macrofossils. Dashed curve shows ralative sea-level change in the Osaka Bay area after [19]. The ageof Kikai–Akahoya tephra horizon (K-Ah; stars) is used 7330 cal BP, estimated by counting varve sediments after [18]. Any age-offset(the lag between the age of sample and that of deposition) and sediment compaction effects are not taken into account. However, theseeffects would be small, since the age curves are consistent with K-Ah age. T. Nakanishi et al. / Nucl. Instr. and Meth. in Phys. Res. B 223–224 (2004) 573–578  577
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks