Fujian Province is located on the continental margin of southeastern China (Figure 1). The collision between the Asiatic Plate and the Philippine Sea Plate forms the Taiwan compressional tectonic deformation zone, which is dominated by uplift in the Central Range, one of the most intense crustal tectonic deformation areas. Because of this plate activity, the Fujian area adjacent to Taiwan is an active tectonic zone that has the most intense earthquake activity in southeastern China, and several large historic earthquake have damaged this area .
The fault structure in this area is well developed and exhibits a series of fault zones in different directions [2–6]. Because the area presents considerable vegetation cover and inconvenient traffic patterns, field investigations are difficult. Therefore, the regional tectonic features are still controversial.
With the development of satellite technology, optical remote sensing has been widely used to interpret the structures of neotectonic fault zones and the surface morphology of tectonic landforms, and radar remote sensing technology has been successfully used to monitor ground deformations. In this study, we used image enhancement and 3D representation to characterize the main geological features at the west of the Taiwan Strait, and the PS-InSAR method was used to monitor the fault activity for earthquake risk estimation in Fujian Province.
In particular, Synthetic Aperture Radar Interferometry (InSAR) can be used under all weather conditions and has the advantage of continuous spatial coverage. Ground deformations on the centimetre to millimetre scale can be measured using Differential Synthetic Aperture Radar Interferometry (D-InSAR) [7–9]. However, conventional D-InSAR deformation monitoring technology has limited accuracy because of spatiotemporal and atmospheric delay effects, which severely restrict the application of D-InSAR technology for monitoring ground deformations.
To avoid the limitations associated with conventional D-InSAR technology, such as interferometric phase decorrelations and atmospheric effects, researchers have proposed a permanent scatterer interference measurement technology [10–15]. Permanent Scatterers (PS) are points that have stable reflection characteristics over long time scales, and they can effectively reduce the influence of spatiotemporal decorrelations and atmospheric effects on conventional differential interferometric radar. Additionally, when conventional D-InSAR cannot form interference fringes, a time series of radar images of Permanent Scatterers can be used to obtain the deformation rate of each discrete PS pixel, which can be interpolated to determine the surface deformation of the entire experimental area. This process suppresses the influence of spatiotemporal baseline decorrelations and atmospheric effects, and the long-term continuous monitoring of deformations will be the main application of this technology in the future [16–22].
The most active surface deformations west of the Taiwan Strait occur in the Quanzhou area where the Yongan-Jinjiang and Changle-Naoao fault zones intersect. Because of the area’s dense population and highly developed industries, these faults pose a major risk. Therefore, PS-InSAR monitoring of the surface deformation of these fault zones has a guiding significance for territorial planning, industrial layouts, urban development and major construction at the west of the Taiwan Strait.
2 Data and methods
2.1 Neotectonic interpretation
Remote sensing is a powerful tool for earth scientists and can be used in various geological applications. This technology records quantitative observational parameters, such as spectral radiance, and the collected data have the potential for use as a rich source of information on landforms, land cover, and tectonic features. Many workers have successfully applied 3D techniques and DEM-based studies to understand tectonic geomorphology and active faults [23–28].
Resource data (Landsat ETM+, QuickBird) were the basic data sources in the present study. A digital elevation model (DEM) provided by the Shuttle Radar Topography Mission (SRTM) was also used to map a number of the major structures in the area. These data sources provided the basic information required to characterize the topographic attributes of the terrain (Figure 2). In this study, DEMs were used to create three dimensional (3D) models of inferred faults using the best possible settings of vertical exaggerations and sun viewing angles. A 3D visualization method was used to interpret the 3D images and resolve the complex fracture pattern of this area because of its closely spaced and interconnected fault systems. The fault morphology was determined using Landsat ETM+ images overlaid on a SRTM (Figure 3).
Digital image processing includes image restoration, enhancement, quantization, and spatial filtering. In this study, we used standard image processing techniques, such as contrast stretching, spatial filtering, image sharpening, and principal component analyses.
Regional interpretations were performed to map the lineaments of the study area onto the images. Geomorphic evidence for fault activity includes offset ridges, offset drainage, displaced beds, and displaced alluvial fans [29–31]. Satellite and field observations have confirmed that these features are indicative of geological structures. Spectral criteria for identifying these lineaments include differences in tone and contrast. The tonality varies as a function of differences in vegetation cover, lithological composition (rock units), soil water content, permeability and rock strength .
The data processing procedure associated with PS-InSAR technology is described below. First, one scene is selected from the image data as the common master image, and all of the remaining images are treated as subordinate images and registered and sampled again in the master image space to form an initial time series of differential interferograms . Next, a PS point target identification method is used to identify a series of PS point targets with stable scattering characteristics, and then the interference phase of each PS point target is extracted. The SAR phase simulated by the DEM is gradually reduced through the adoption of an external DEM, and the DEM coordinate system is then transformed, the image is registered, and the SAR phase is simulated, thus forming a second-time series of differential interferograms. Finally, the surface deformation rate and DEM elevation correction values are determined by parameter estimations.
During deformation monitoring, the terrain phase was removed from the interferometric phase by adopting SRTM-DEM data SRTM3 (3 arc-seconds), which were jointly measured by the National Aeronautics and Space Administration (NASA) and National Imagery and Mapping Agency (NIMA), as the external DEM data. In this study, we used the version V4.1 data product, which was publicly launched in 2003 and has a data range from 24.2° to 25.2°N and 108° to 109°E.
The spatial and temporal baselines and Doppler centre frequency, among other factors, were comprehensively estimated to select the master image, and finally the optimal master image adopted in this study was the SAR data scene 19980403. Other images as well as the master image constitute a spatial vertical baseline and a temporal baseline of 21 interference couples (Figure 5). The SAR image data, external DEM data and the master SAR image data were registered and sampled again. After selecting the reference master image, all the subordinate images had to be registered to the master image to create a one-to-one correspondence between the pixel positions corresponding to the same surface in the master and subordinate images. The registration accuracy of the image pair directly affected the phase coherence of the subsequent interference couples. Therefore, the registration accuracy of the image pairs in PS-InSAR processing must reach 0.2 pixels.
An analysis of the deformation time series used the Stanford Method for Persistent Scatterers/Multi-Temporal InSAR (StaMPS/MTI) interference superposition analysis software. This software focuses on highly coherent points with stable scattering properties on the ground within a certain period. The software includes several InSAR programs, such as ROI_PAC, DORIS, Snaphu, and Triangle. ROI_PAC was used to process the SLC image data in a standard CEOS format that is generated by RAW format data; DORIS was used to read and register the SLC data, convert the DEM phases and generate interferograms; Snaphu was used for interferogram unwrapping and PS three-dimensional unwrapping; and Triangle was used to compose PS points in a Delaunay triangulation network. Phase unwrapping eliminated errors associated with atmosphere and other parameters through iterative processing. Next, the SVD method was used to estimate the deformation results. Finally, a yearly average deformation rate map (Figure 6) of this area from 1996-1999 was obtained using a time series analysis of highly coherent points.
Tectonic deformation of the Quanzhou area, in the west of the Taiwan Strait, was monitored and analysed using PS-InSAR technology. Highly coherent PS points has been used for a phase regression analysis in the study area, including the Changle-Naoao and Yongan-Jinjiang fault zones. Based on the interpretation of remote sensing data and PS-InSAR monitoring, some parameters of the neotectonic and crustal deformations in the study area were observed.
Digitally enhanced remote sensing data helped to delineate spatial features because they can display many of the morphological characteristics of the earth’s crust with better surface expression [34, 35]. The images were useful in conclusively delineating the Quaternary structures of the study area, and prominent lineaments and fault traces were differentiated from various landforms, which are indicative of active tectonics.
Interference processing with the master image yielded 21 interferograms, which demonstrated that the overall coherence was poor. However, because PS InSAR technology conducts time series analyses based on highly coherent points, the baseline data requirements were expanded to a certain extent and the effects of atmospheric delay on the results were reduced.
The monitoring results show that the overall deformation of this area was small and presented an annual deformation magnitude between 8 and 9 mm. However, the complex terrain in the area, abundant water vapour, and atmospheric delay still affected the results. The phase unwrapping graph (Figure 6) showed that the unwrapping phase at the PS point had a relatively smooth spatial distribution; however, because the potential for a greater error was small, the unwrapping phase can be considered relatively reliable. To further determine the reliability of the results, we created a standard deviation distribution graph of the annual average rate. The graph showed that the overall results for the annual average rate were reliable and had a standard deviation of approximately 1–2 mm/yr for most PS points.
Fujian Province, located near convergence boundary of the Eurasian plate and the Philippines plate. Due to Philippine Sea Plate squeezed to the west and Asiatic Plate upthrusted to the east, the two plates interaction on the island of Taiwan, and the deformation zone of Taiwan, particular in the central mountain range uplift, became one of the most intense tectonic deformation zones in the earth’s crust. Influenced by these two tectonic plates, the Fujian area adjacent to the Taiwan Sea is the most active tectonic zone of seismic activity in southeast coastal of mainland China .
The earth’s crust 3D structure model indicates that the phenomena of upper mantle uplift and crustal thinning are obvious along the coastal fault zone and the Jiulongjiang fault. According to the field investigation and geodynamical feature, these faults are high angle strike slip faults which have a certain vertical component. GPS observation results show that the active rate of faults in Fujian area is about 3 mm/a, and the focal mechanism resolutions of small earthquakes in study area show that the principal stress direction is NNW . In 1999, the vertical uplift is 2–7 m caused by the Chi-Chi earthquake in Taiwan . These facts demonstrate that Quaternary stress field and crustal movement direction in Fujian-Taiwan area is NNW-NW and, the movement of Philippines Plate might be the main driving force of crustal movement in Fujian area. The structure of the southeastern coastal area of Fujian Province is controlled by two groups of faults trending to the northeast and northwest respectively. The seismic exploration found that the faults are located in the boundary of normal continental crust and thinned continental crust, and the Taiwan strait west boundary fault might be caused by Cenozoic strong subsidence. The northeast trending fault group includes two fault zones, the Changle-Naoao and coastal fault zones, whereas the northeast trending fault group includes the Hanjiang, Jiulongjiang, Yongan-Jinjiang, Xinghuawan, and Minjiang fault zones. The Changle-Naoao fault zone is an important tectonic zone on China’s southeastern coast that formed prior to the Mesozoic Era with width of 50–70 km and was affected by a variety of activity thereafter. The Changle-Naoao fault zone extends southwest of Naoao to Guangdong Province; however, because of Quaternary cover in the Chaoshan Plain, there are few obvious signs of the fault zone at the surface. The fault zone is divided into inner, middle and outer zones. The inner zone extends from Fuzhou to Zhangzhou, Raoping to Chenghai and Shantou, and then to the south near Huilai. The middle zone shows obvious linear features to the east of Fuzhou, Quanzhou, and Naoao Island. The outer zone crosses the eastern Naoao Island and then merges with the middle zone near Quanzhou. In Fujian coastal area, the NW faults are lateral tension faults formed later. As a result of intense activity, the faults in other directions are almost have been cut off. The distribution of fault group is relatively regular, and seismic activity intensity increases southwards.
The faults in the study area such as Changle-Naoao and Yongan-Jinjiang fault zones are vertical and have horizontal striations and lateral dislocation drainages, which indicates that they are two groups of conjugate strike-slip faults resulting from oblique compressive stress from the collision between the Asiatic Plate and the Philippine Sea Plate . This collision formed two groups of conjugate shear facture zones trending to the northeast and northwest (Figure 7).
The deformation monitoring results showed that the annual displacement rates of the faults are 3–5 mm (Figure 6), indicating that the Changle-Naoao and Yongan-Jinjiang fault zones are still active and are at risk of a major earthquake in the future.
We are grateful to two anonymous reviewers for their constructive feedback on our manuscript. This study was financially supported by the Important Direction Project of Knowledge Innovation in the Resource and Environment Field of the Chinese Academy of Sciences (KZCX2-EW-QN112) and the fund from the Key Laboratory of Petroleum Resources, Gansu Province (No. 135CCJJ20160517). We thank Global Land Cover Facility (GLCF) for providing the SRTM DEM and Landsat ETM datasets (http://glcf.umd.edu/).
Chen C., Study on special distribution characteristics of seismic activity and active fault in Fujian area, Earthquake, 2005, 25(3), 102-108(in Chinese). Google Scholar
Zhu J. and Huang Z., City active fault explore and seismic risk evaluation in Fujian province, Fujian Seismology, 2003, 19(2), 1-9(in Chinese). Google Scholar
Zhang L., Qu G., Zhu J., Li Y., and Huang Z., Model of Quaternary tectonic movement and dynamic setting of basins along the coast of Fujian, China, Geological Bulletin of China, 2007, 26(3), 275-288(in Chinese).Google Scholar
Ye Y., Huang Q., Wang Y., and Xue L., Quaternary active features of the NW-striking fault zone in Jiulong River down-stream area, Fujian Province, Northwestern Seismological Journal, 2012, 34(1), 64-71(in Chinese).Google Scholar
Chen Y., Wang Z., Ye J., Liu F., and Xu Y., Preliminary study of geotectonic stress filed in Fujiang area since late Pleistocene, Journal of Oceangraphy in Taiwan Strait, 1999, 18(4), 426-431(in Chinese).Google Scholar
Huang Q., Zhuang J., Xie Z., and Zhang Q., Active fault and volcanic activity in the Longhai-Zhangpu coastal area, Fujian Province, Earthquake Research in China, 2005, 19(2), 214-223(in Chinese). Google Scholar
Shan X., Ma J., Liu J., Wang C., and Song X., The theory of D-InSAR and its application to mapping the displacement fields of earthquake, Seismology and Geology, 2001, 23(3), 439-446(in Chinese). Google Scholar
Singleton A., Li Z., Hoey T., and Muller J.P., Evaluating subpixel offset techniques as an alternative to D-InSAR for monitoring episodic landslide movements in vegetated terrain, Remote Sens. Environ., 2014, 147, 133-144. CrossrefGoogle Scholar
Qu C.Y., Shan X.J., Song X.G., Zhang G.F., Zhang G.H., and Guo L.M., The PSInSAR technique and its application to the study on crustal deformation of the Haiyuan fault zone, Chinese Journal of Geophysics-Chinese Edition, 2011, 54(4), 984-993(in Chinese).Web of ScienceGoogle Scholar
Hooper A., Zebker H., Segall P., and Kampes B., A new method for measuring deformation on volcanoes and other natural terrains using InSAR persistent scatterers, Geophys. Res. Lett., 2004, 31(23), 1-5.Google Scholar
Zhang J., Gong L., and Jiang W., Application of PS In-SAR technique to measurement of long-term crustal deformation, Recent Developments in World Seismology, 2006, 11(6), 1-6(in Chinese).Google Scholar
Aobpaet A., Cuenca M.C., Hooper A., and Trisirisatayawong I., InSAR time-series analysis of land subsidence in Bangkok, Thailand, Int. J. Remote Sens., 2013, 34(8), 2969-2982.CrossrefWeb of ScienceGoogle Scholar
Massonnet D., Rossi M., Carmona C., Adragna F., Peltzer G., Feigl K., and Rabaute T., The displacement field of the Landers earthquake mapped by radar interferometry, Nature, 1993, 364, 138-142. CrossrefGoogle Scholar
Cakir Z. and Akoglu A.M., Synthetic aperture radar interferometry observations of the M=6.0 Orta earthquake of 6 June 2000 (NW Turkey): Reactivation of a listric fault, Geochem. Geophy. Geosy., 2008, 9, 1-20. Web of ScienceGoogle Scholar
Guerrieri L., Baer G., Hamiel Y., Amit R., Blumetti A.M., Comerci V., Di Manna P., Michetti A.M., Salamon A., Mushkin A., Sileo G., and Vittori E., InSAR data as a field guide for mapping minor earthquake surface ruptures: Ground displacements along the Paganica Fault during the 6 April 2009 L’Aquila earthquake, J. Geophys. Res., 2010, 115, 1-16.Web of ScienceGoogle Scholar
Sun J., Xu X., Shen Z., Shi Y., and Liang F., Parameter inversion of the 1997 Mani earthquake from INSAR co-seismic deformation field based on linear elastic dislocation model - I. Uniform slip inversion, Chinese Journal of Geophysics-Chinese Edition, 2007, 50(4), 1097-1110(in Chinese).Google Scholar
Wang H. and Zhou X., InSAR and seismic cycle monitoring, Geospatial Information, 2009, 7(3), 71-73(in Chinese).Google Scholar
Shi X., Zhang L., Liao M., and Balz T., Deformation monitoring of slow-moving landslide with L- and C-band SAR interferometry, Remote Sens. Lett., 2014, 5(11), 951-960.CrossrefWeb of ScienceGoogle Scholar
Zhao C., Lu Z., and Zhang Q., Time-series deformation monitoring over mining regions with SAR intensity-based offset measurements, Remote Sens. Lett., 2013, 4(5), 436-445.Web of ScienceCrossrefGoogle Scholar
Vassilopoulou S., Sakkas V., Wegmuller U., and Capes R., Long term and seasonal ground deformation monitoring of Larissa Plain (Central Greece) by persistent scattering interferometry, Open Geosci., 2013, 5(1), 61-76. Google Scholar
Roberts N. and Cunningham D., Automated alluvial fan discrimination, Quaternary fault identification, and the distribution of tectonically reactivated crust in the Gobi Altai region, southern Mongolia, Int. J. Remote Sens., 2008, 29(23), 6957-6969. CrossrefWeb of ScienceGoogle Scholar
Demirkesen A.C., Digital terrain analysis using Landsat 7 ETM+ imagery and SRTM DEM: a case study of Nevsehir province (Cappadocia), Turkey, Int. J. Remote Sens., 2008, 29(14), 4173-4188.CrossrefWeb of ScienceGoogle Scholar
Babaahmadi A., Yassaghi A., Naeimi A., Dini G.R., and Taghipour S., Mapping Quaternary faults in the west of Kavir Plain, northcentral Iran, from satellite imageries, Int. J. Remote Sens., 2010, 31(19), 5111-5125.CrossrefGoogle Scholar
Xypolias P. and Koukouvelas I.K., Fault trace parameters as a tool for analysing remotely sensed fault arrays: An example from the eastern Gulf of Corinth, Greece, Int. J. Remote Sens., 2004, 25(21), 4685-4699.CrossrefGoogle Scholar
Singh A.K., Parkash B., and Choudhury P.R., 3D perspective view of the northwestern part of the Dehradun valley, India, Int. J. Remote Sens., 2007, 28(11), 2401-2402. Web of ScienceCrossrefGoogle Scholar
Kamberis E., Bathrellos G.D., Kokinou E., and Skilodimou H.D., Correlation between the structural pattern and the development of the hydrographic network in a portion of the Western Thessaly Basin (Greece), Cent. Eur. J. Geosci., 2012, 4(3), 416-424. Web of ScienceGoogle Scholar
Migiros G., Bathrellos G.D., Skilodimou H.D., and Karamousalis T., Pinios (Peneus) River (Central Greece): Hydrological — Geomorphological elements and changes during the Quaternary, Cent. Eur. J. Geosci., 2011, 3(2), 215-228.Web of ScienceGoogle Scholar
Kokinou E., Hariklia D.S., George D.B., Antonarakou A., and Kamberis E., Morphotectonic analysis, structural evolution/pattern of a contractional ridge: Giouchtas Mt., Central Crete, Greece, J. Earth. Syst. Sci., 2015, 124(3), 587-602. CrossrefWeb of ScienceGoogle Scholar
Ioanna I., Dimitrios R., Theodora P., and Paris T., Geotechnical and mineralogical properties of weak rocks from Central Greece, Cent. Eur. J. Geosci., 2009, 1(4), 431-442. Web of ScienceGoogle Scholar
Bathrellos G.D., Gaki-Papanastassiou K., Skilodimou H.D., Papanastassiou D., and Chousianitis K.G., Potential suitability for urban planning and industry development using natural hazard maps and geological–geomorphological parameters, Environ. Earth Sci., 2011, 66(2), 537-548. Web of ScienceGoogle Scholar
Chousianitis K., Del Gaudio V., Sabatakakis N., Kavoura K., Drakatos G., Bathrellos G.D., and Skilodimou H.D., Assessment of earthquake induced landslide hazard in Greece: from arias intensity to spatial distribution of slope resistance demand, B. Seismol. Soc. Am., 2016, 106(1), 174-188. Web of ScienceCrossrefGoogle Scholar
Apostolidis E. and Koukis G., Engineering-geological conditions of the formations in the Western Thessaly basin, Greece, Open Geosci., 2013, 5(3), 407-422. Google Scholar
Wen X. and Xu X., Analysis of the two historical earthquakes of 1067 and 1574 in the southern coastal region of China, Seismology and Geology, 2005, 27(1), 1-10(in Chinese). Google Scholar
Zhou S., Shuai P., Guo F., Chen Z., Deng Q., Zhou X., and Chen T., The present quantitative study of crustal movement of Fujian and its marginal seas, Acta Seismological Sinica, 2000, 22(1), 66-72(in Chinese). Google Scholar
Chen Y.-G., Chen W.-S., Lee J.-C., Lee Y.-H., Lee C.-T., Chang H.-C., and Lo C.-H., Surface rupture of 1999 Chi-Chi earthquake yields insights on active tectonics of central Taiwan, B. Seismol. Soc. Am., 2001, 91(5), 977-985. Google Scholar
Wu J., Suppe J., Lu R., and Kanda R., Philippine Sea and East Asian plate tectonics since 52 Ma constrained by new subducted slab reconstruction methods, J. Geophys. Res., 2016, 121(6), 4670-4741. Web of ScienceCrossrefGoogle Scholar
About the article
Published Online: 2017-05-05
Citation Information: Open Geosciences, Volume 9, Issue 1, Pages 126–132, ISSN (Online) 2391-5447, DOI: https://doi.org/10.1515/geo-2017-0010.
© 2017 Jianming Guo et al.. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. BY-NC-ND 3.0