Logo 2016 Dragon 3 Final Results Symposium

Conference Agenda

Overview and details of the sessions and sub-session of this conference. Please select a date or session to show only sub-sessions at that day or location. Please select a single sub-session for detailed view (with abstracts and downloads if available).

Session Overview
D2: ID.10607 Crustal Deformation & Infrastructures
Tuesday, 05/Jul/2016:
4:30pm - 5:30pm

Session Chair: Cécile Lasserre
Session Chair: Qiming Zeng
Workshop: Terrain & Geoid Measurement
Location: A-Li Mountain Room, 5.5 Floor, Junyi Dynasty Hotel

Show help for 'Increase or decrease the abstract text size'
Oral presentation

The Recent Crustal Movement Characteristics Analysis of Lhasa Middle Block Based on GPS and InSAR Technology

Jingfa Zhang, Yunfeng Tian, Yunsheng Li, Yi Luo

Institute of Crustal Dynamics, China Earthquake Administration;

Comparing and analyzing the macroscopic dynamics of integral faulting activity of Lhasa middle tectonic block is useful to judge the danger zone for strong earthquakes.

The paper analyses the crustal vertical and horizontal deformations in this region by means of InSAR technique and GPS data:

(1)the new GPS stations and mobile observation points, combined with the existing GPS data, access to the crust movement rate, monitoring tiny crust seasonal shock, try to detect slow slip events;

Collected in mainland China tectonic environment monitoring network in the study area and the surrounding layout of the calendar year observation data of GPS reference station, regional station.Active fault zones in the Tibetan capital Lhasa block in the middle of important layout of GPS observation network flow 21 points, had completed the first and the second phase of the field of GPS observations.
To 2013, 2014 and 2015 observations of three LGN network flow of the GPS observation data are analyzed, get ITRF2008 LGN under the framework of the global network of each station position time series in a single day.Based on the analysis of the existing GPS data, the current horizontal motion characteristics of the study area and the surrounding.In the northeast - xidatan tectonic belt, wood county, southern showed weaker tensile crust movement, north station show some right-lateral strike-slip properties.Dongya - xidatan tectonic belt yambajan segment showed the earth's crust extensional movement.Dongya - xidatan male segment of tectonic belt, the GPS on either side of the fracture velocity profile reflects weaker nature of normal fault.

(2)based on the accumulation of various radar data since 1991, the strong earthquake in the area are obtained using InSAR technology deformation parameters and the movement characteristics of major active faults;

On the basis of traditional coherent point extracting method, this paper proposes a new extraction method, and use the network delay phase respectively for three types of atmospheric correction.
Respectively two Track data for time series analysis, and the two data sets is joining together of the time series analysis results of the whole area of time-series analysis results, show that there are obvious in both sides of the fault fracture zone under the crustal relative motion, collapse the nature of the fault rupture is dextral strike-slip fault, before this also verified by the geological survey of the conclusions.
Since the 1990 s was studied by InSAR technique in research of small earthquakes in the area of the deformation field.On March 24, 2010 in a Tibetan region Mw5.6 earthquake as an example, using the traditional acoustic difference method to get the coseismic deformation field of the earthquake.In InSAR coseismic deformation observation data as constraint, the earthquake sliding space geometric features of fault and analysis of the distribution fault slip distribution on the surface of the.

(3)based on high resolution images and the method of tectonic geomorphology, focused on active faults and tectonic geomorphology late quaternary deformation characteristics of quantitative research, study fault block, etc.;

Damxung - Yambajan rift is south north and south to one of the most active tectonic units in the rift valley.Tectonic geomorphology study reveals evolution characteristics since the quaternary tectonic activities and tectonic activity.Using rock channel profile analysis method of river across Nyenchen-Tanglha space form, and is able to analysis rock uplift. Damxung - Yambajan fracture in plays a significant role in regulating regional activities, including section with left-lateral strike-slip movement is given priority to, and in the southern section is given priority to with normal fault activity effect. And in the south than the north is undergoing rock lifting effect is more apparent.Related riverbed profile analysis results reflect the instability characteristics of the region.

Relevant results will provide the Lhasa middle block and surrounding areas of the earth's crust movement mechanism of dynamics analysis provides a wealth of scientific data.

Oral presentation

An Improved InSAR Time-series Approach for extracting Small and Long-wavelength Land Surface Movements

Julia Stockamp1, Zhenhong Li2, Paul Bishop1, Alistair Rennie3

1School of Geographical and Earth Sciences, University of Glasgow, Glasgow, UK; 2School of Civil Engineering and Geosciences, Newcastle University, Newcastle upon Tyne, UK; 3Scottish Natural Heritage, Inverness, UK;

Differential SAR Interferometry (D-InSAR) is an established technique for analysing crustal motion and land surface deformation caused by co- and interseismic processes, volcanic activity, landslides or subsidence. When it comes to the detection of small, long-wavelength displacements, the application of D-InSAR becomes challenging and a very high quality standard in terms of precision and accuracy is necessary to make it a competitive tool to established geodetic techniques, such as GNSS. Under the right conditions conventional D-InSAR can be sensitive to land deformations in the mm level. The accuracy of D-InSAR time series techniques can reach 1 mm/yr or even higher for mean deformation velocities (Hammond et al. 2012), but is dependent on weather conditions, the number of SAR acquisitions available, and the total time span of images. In general, the accuracy of the extracted deformation signal is only as good as the accuracy of the interferometric phase, which is influenced by several error sources, such as orbital, atmospheric and topographic artefacts.

This study investigates the applicability of a D-InSAR time-series technique in its ability to monitor wide-scale, subtle ground movement over a long-term period under challenging environmental conditions (atmospheric water vapour, temporal decorrelation). It thereby contributes to the improvement of D-InSAR time-series methodology for deformation monitoring. The research area is Scotland, where the vertical land movement lies in the low mm-level and is mainly caused by glacial isostatic adjustment, as well as neo-tectonic effects and subsidence due to mining activities. The vertical crustal motion there has important implications for the interpretation of past and future relative sea-level trends at the coast. The findings of this study can provide important information for the application of InSAR time-series techniques and the mapping of natural hazards in areas world-wide that exhibit similar ground deformation characteristics with a low magnitude and on a broad spatial scale, such as interseismic strain accumulation or subsidence phenomena.

Different SAR satellites and frequencies are used in this investigation in order to cover a time frame of about 20 years. They include ESA’s ERS-1/2, ESA’s Envisat ASAR, ESA’s Sentinel-1 and JAXA’s ALOS PALSAR.

The focus lies on the Small Baseline technique for inverting a ground deformation time-series signal. This includes the advancement of error correction techniques for orbital and atmospheric disturbances within the time-series inversion processing chain that require to be addressed in order to allow the extraction of any deformation signal. Orbital artefacts are especially problematic when it comes to measuring long-wavelength wide-scale ground deformation, since both signals often show similar spatial patterns, which are difficult to distinguish. To reduce orbit errors in the InSAR interferograms, an improved network correction technique has been tested with good results (Stockamp et al. 2014). It combines the conventional approach after Biggs et al. (2007) with phase loop triplets of interferograms in order to introduce further observation equations that constrain parameters better. In addition, further methods are being explored for the separation of the desired deformation signal from residual orbital and atmospheric errors, such as Principal Component Analysis and Independent Component Analysis in order to take advantage of the different signals’ spatial and temporal characteristics.

Biggs, J., Wright, T., Lu, Z. & Parsons, B. 2007. Multi-interferogram method for measuring interseismic deformation: Denali Fault, Alaska. Geophysical Journal International 170, 1165-1179.

Hammond, W. C., Blewitt, G., Li, Z., Plag, H.-P. & Kreemer, C. 2012. Contemporary uplift of the Sierra Nevada, western United States, from GPS and InSAR measurements. Geology 40, 667-670.

Stockamp, J., Li, Z., Bishop, P., Hansom, J., Rennie, A., Petrie, E., Tanaka, A., Bingley, R. & Hansen, D. 2015. Investigating Glacial Isostatic Adjustment in Scotland with InSAR and GPS Observations. Proceedings Fringe 2015 Workshop, Frascati, Italy. European Space Agency (Special Publication) ESA SP-731.

Stockamp-An Improved InSAR Time-series Approach for extracting Small and Long-wavelength Land Surface_Cn_version.pdf

Oral presentation

Source Parameters of the 2016 Mw 6.4 Taiwan Earthquake from Multi-platform InSAR Observations

Jiajun Chen1, John Elliott2, Wanpeng Feng3, Zhenhong Li1, Sergey Samsonov3, Peter Clarke1

1COMET, School of Civil Engineering and Geosciences, Newcastle University, Newcastle upon Tyne, NE1 7RU, United Kingdom; 2COMET, Department of Earth Sciences, University of Oxford, Oxford, OX1 3AN, UK; 3Canada Centre for Mapping and Earth Observation, Natural Resources Canada, 560 Rochester Street, Ottawa, ON K1A 0E4, Canada;

The 6th February 2016 Taiwan Mw 6.4 earthquake occurred in southern Taiwan, causing widespread damage and 117 deaths. According to the Central Weather Bureau (CWB) and U.S. Geological Survey (USGS), the focal depth was ~23km. The focal mechanism, aftershocks and shaking distribution indicate this event likely resulted from a blind thrust fault. The mainshock is only 15km west to the 2010 Jiashian Mw 6.3 earthquake, and both events show many similarities. Geodetic and seismic inversions suggested the Jiashian event corresponded to the Chishan Transfer Fault Zone (CTFZ) at surface, but the source at depth remains unclear (Huang et al., 2013). It is believed that geodetic inversions for the 2016 event will help us better understand the geological faulting in southern Taiwan.

Coseismic interferograms from multi-platforms including Sentinel-1A, RADARSAT-2, and ALOS-2 were used to investigate the fault geometry and slip distribution of this event. Our best model from the geodetic inversion package PSOKINV (Feng et al., 2013) suggests a primary slip patch of about 19 km ´7 km with a peak slip of 0.87m at a depth of about 11km, and a cumulative geodetic moment of 4.47´1018 Nm, corresponding to Mw 6.37. The fault geometry is N30°W in strike and 23° dipping to NE, and the major slip is concentrated at the depth of 10-15km located to the NW of the hypocentre, which is almost at the northern end of the Chishan Fault (CSF) and close to the Chaochou Fault (CCF). To better explain observations of this deep source event, a layered model will be used to determine more realistic slip distribution (Wang et al., 2006). We will also calculate Coulomb stress changes to assess potential earthquake hazards in southern Taiwan.

Feng, W., Li, Z., Elliott, J.R., Fukushima, Y., Hoey, T., Singleton, A., Cook, R. and Xu, Z., 2013. The 2011 MW 6.8 Burma earthquake: fault constraints provided by multiple SAR techniques. Geophysical Journal International, p.ggt254.

Huang, M.H., Dreger, D., Bürgmann, R., Yoo, S.H. and Hashimoto, M., 2013. Joint inversion of seismic and geodetic data for the source of the 2010 March 4, Mw 6.3 Jia-Shian, SW Taiwan, earthquake. Geophysical Journal International, 193(3), pp.1608-1626.

Wang, R., Lorenzo-Martín, F. and Roth, F., 2006. PSGRN/PSCMP—a new code for calculating co-and post-seismic deformation, geoid and gravity changes based on the viscoelastic-gravitational dislocation theory. Computers & Geosciences, 32(4), pp.527-541.

Chen-Source Parameters of the 2016 Mw 64 Taiwan Earthquake_Cn_version.pdf

Oral presentation

Precise Measurement of Coseismic Deformation by Multi-Temporal Radar Interferometry

Lei Zhang1, Yangmao Wen2, Xiaoli Ding3, Zhong Lu4, Jun Hu5

1The Hong Kong Polytechnic University, Hong Kong S.A.R. (China); 2Wuhan University; 3The Hong Kong Polytechnic University, Hong Kong S.A.R. (China); 4Southern Methodist University (USA); 5Central South University (China);

Conventional Differential InSAR derived coseismic measurements are often perturbed by topography residuals, long wavelength orbital ramps,potential phase unwrapping errors and atmospheric artifacts. These nuisance components degrade the precision of the inverted fault slips. We outline here a multi-temporal InSAR (MT-InSAR) model to obtain precise coseismic deformation, with application to 6 October 2008 Dangxiong earthquake (Mw 6.3). By analyzing a set of pre-seismic SAR images and at least one post-seismic image our model can jointly estimate the coseismic deformation together with DEM and orbital errors, as well as pre-seismic (and post-seismic) deformation rate. The parameterization of orbital errors in our model is based primarily on its spatially smooth and temporally random characteristics. The joint model makes the separation of long wavelength orbital error, DEM residuals and seismic deformation possible. Furthermore, our method performs linear parameter estimation directly on the wrapped differential phases at arcs (coherent point pairs) implying no phase unwrapping procedure is needed, neither is a solution space searching needed. The slip due to Dangxiong earthquake inverted respectively based on MT-InSAR and DInSAR derived coseismic deformation shows up to several tens of cm differences indicating the topography error and distortions introduced by application of a best fitting orbital phase plane to the coseismic interferogram can bias the fault slip inversion.

Zhang-Precise Measurement of Coseismic Deformation by Multi-Temporal Radar Interferometry_Cn_version.pdf


Earthquake-Induced Building Damage Assessment Based on SAR Correlation and Texture

Lixia Gong, Qiang Li, Jingfa Zhang

Institute of Crustal Dynamics, China Earthquake Administration, China, People's Republic of;

Comparing with optical Remote Sensing, the Synthetic Aperture Radar (SAR) has unique advantages as applied to seismic hazard monitoring and evaluation. SAR can be helpful in the whole process of after an earthquake, which can be divided into three stages. On the first stage, pre-disaster imagery provides history information of the attacked area. On the mid-term stage, up-to-date thematic maps are provided for disaster relief. On the later stage, information is provided to assist secondary disaster monitoring, post-disaster assessment and reconstruction second stage. In recent years, SAR has become an important data source of earthquake damage analysis and evaluation.

Correlation between pre- and post-event SAR images is considered to be related with building damage. There will be a correlation decrease when the building collapsed in a shock. Whereas correlation decrease does not definitely indicate building changes. Correlation is also affected by perpendicular baseline, the ground coverage type, atmospheric change and other natural conditions, data processing and other factors. In this paper, building samples in the earthquake are used to discriminate the relation between damage degree and SAR correlation. First, the pre- and post-event SAR images are registered. Second, the vegetation area is excluded and the urban area with buildings is masked as ROI. Then the correlation coefficient and the backward scattering difference are calculated over the samples, aiming to fit the function between damage degree and these two parameters. on the basis of the discriminant function, damage levels of the whole ROI is calculated and classified.

There is an uncertain factor in the above method. After an earthquake, due to building collapse, SAR intensity may increase in some parts while decrease in others. The correlation between intensity and building damages can be positive or negative, which leads to uncertainty in the function. So SAR texture difference is utilized to build the discriminant function in place of intensity difference. Whether SAR intensity of a damaged building increases or decreases, the texture parameter will always decrease.

ENVISAT ASAR data of Dujiangyan City are used as experimental data. According to the results of field survey, 1113 building samples are selected to amend the parameters of empirical function, which discriminate the building damage on the basis of SAR intensity. Statistical analysis are carried out on SAR texture features of building damage, and it is found that variance, homogeneous and other parameters are good at characterizing the building damage. Then a discrimination model based on SAR texture is proposed. The corresponding function is built and applied, resulting in a good consistence with the field survey. And the accuracy has improved significantly compared with that of empirical function.

Gong-Earthquake-Induced Building Damage Assessment Based_Cn_version.pdf


Extraction of Building Damage due to Nepal Earthquake Based on Texture from Sentinel-1A

Qiang Li1,2, Jingfa Zhang2

1Institute of Engineering Mechanics, China Earthquake Administration,China,People's Republic of;; 2Institute of Crustal Dynamics, China Earthquake Administration, China,People's Republic of;;

In April 25, 2015, Nepal occurred Ms8.1 earthquake, which caused a number of buildings in Kathmandu collapsed. According to reports, the earthquake caused 8792 people to die, injured more than 23000 people, nearly 300 thousand people were affected. Kathmandu area is densely populated. How to evaluate the disaster of buildings quickly and effectively, is a useful part of disaster relief, but also the key to economic loss assessment and post-disaster reconstruction. The European Space Agency (ESA) launched Sentinel-1A satellite mainly used for environmental monitoring in April 3, 2014. The satellite is not affected by weather conditions and can penetrate through clouds, Moreover, in the event of a crisis when it can respond quickly to get disaster image. After the Nepal earthquake, Sentinel-1A accessed to the remote sensing image of Kathmandu area through rapidly adjustment.

In the radar wave irradiation, the scattering phenomenon of the building area mainly includes: Seculars reflection of roof, double scattering between the smooth surface and wall surface, scattering of collapsed buildings, strong reflection of collapsed buildings exposed steel and multiple scattering among building, building wall and the ground. Therefore, this phenomenon will cause similar to the scattering of different objects in the SAR images which may exhibit similar gray value, The difference of gray value before and after the earthquake would not fully reflect the actual object changes coupled with speckle noise. SAR images have rich texture features of buildings, which contain significant amount of information such as the land cover structure and the relationship with the environment. The texture feature is relatively stable and less influenced by other factors, Therefore in the SAR image analysis, using the internal texture information can make up for the inadequacy of grayscale characteristics and improve the identification accuracy.

This paper selects building samples of different damage levels to count on the gray level co-occurrence features. The maximum/minimum value, average value and standard deviation were counted. Based on this, the characteristic parameter image which has strong ability of characterizing information on the damage degree of buildings is selected, which is used as the reference image for the following classification. Finally, the information of seismic damage is extracted by the classification of reference image.

Taking Sentinel-1A data of Kathmandu as an example, Contrast feature is selected as the reference image to detect the buildings. The recognition results are compared with the visual interpretation results of optical images, Experimental result shows that the method based on texture feature has a high recognition rate and proves the validity of the algorithm.

ACKNOWLEDGEMENTS: The SAR data used in this paper are provided by the European Space Agency (ESA), Optical data were provided by the China Center For Resources Satellite Data and Application. The author would like to thank them all .

Li-Extraction of Building Damage due to Nepal Earthquake Based_Cn_version.pdf


Haiyuan fault deformation monitoring based on the FRAM - SBAS time series analysis method

Yingxu Song1,2, Ruiqing Niu1, Jingfa Zhang2, Yongsheng Li2, Luo Yi2, Yan Chen2

1China University of Geosciences(Wuhan); 2China Earthquake Apartment, China, People's Republic of;


Haiyuan fault zone in the west of China is not only the important activity block boundary tectonic belt, but also an important strong earthquake activity belt. The Ms8.5 Haiyuan earthquake occurred in that fault zone in 1920 had made a sinistral strike-slip displacement of 10.5m and almost a through fracture of the fault zone (Institute of Geology of China Earthquake Department, 1990).Studying the fracture activity and the seismic risk is meaningful and necessary. Scholars had made a lot research on the activity rate of the Haiyuan fault zone. The landform characteristics and dating results based on geology method indicate that the sinistral strike-slip rate is about 3~8mm/yr (Zhang Zhen-pei,2003; Li et al.,2009). The results obtained from GPS based on geodetic survey methods show that the sinistral strike-slip rate is about 3~10 mm/yr(Gan et al.,2005;Cui et al.,2009). All the results above are observed by sparse discrete points, which is different as the observation location and the density of the point. On the one hand,this may because in different location of the fault deformation rate is different. On the other hand, a single point can also cause inconsistent observations without mutual restriction. A wide range of continuous observation sequence of InSAR technique can obtain the overall image and the dynamic processes of fault zone, which is a good supplement of existing fixed point method. Therefore, we adopt the time series method for the experimental research to get the fault activity rate of the experimental area and provide the basis for the earthquake risk assessment of the experimental area.

The current differential interferometric synthetic aperture radar(DInSAR)technique cannot work well in measuring the long—term accumulated small crustal deformation because of serious constraints from the phase decorrelations in time and space.Therefore,we use a new method called SBAS-InSAR(Small Baseline Subset InSAR) to solve the problem.We will introduce the principle,algorithm and data processing procedures of FRAM-SBAS first in this paper.Then we take the Haiyuan fault zone as the experimental area and use this technique to calculate the deformation rate.24 scenes of ALOS PalSAR data from 2007 to 2010 and the FRAM-SBAS technology presented above are used to examine the small crustal deformation on the experimental area.The final result shows that most coherent point targets show a slip rate about 5~7 mm/yr, which is similar to that from GPS measurements and geological investigations.Then we can say the SBAS method is capable of detecting long-term accumulated crustal deformation on the Haiyuan active fault zone.

Key words: Haiyuan fault zone; SBAS; ALOS PalSAR; Deformation velocity; Groundsurface deformation monitoring

Song-Haiyuan fault deformation monitoring based on the FRAM_Cn_version.pdf


Estimation of the Optimum Phase for Distributed Scatterers by Alternating Projection Algorithm

Fengyu Yao, Ruya Xiao, Mi Jiang, Xiufeng He

Hohai University, China, People's Republic of;

To overcome the shortage of Permanent Scatterers Interferometry (PSI) that there is less PS in non-urban areas, the distributed scatters (DS) was proposed. The spatial density of measurement points was extremely enhanced by mean of jointly process PS and DS. In order to jointly process, the phase for DS must be optimized. Previous studies, namely, the PTA algorithm, have showed their advantages in improving the estimation accuracy of phase values by all possible interferograms. However, the method could be sensitive to initial values. Moreover, the method is computationally intensive. In this paper, we present an efficient algorithm for computing the exact maximum likelihood estimator of the phase values for distributed targets. The method is based on the statistically homogeneous pixels (SHPs) selection and covariance matrix estimation by the FaSHPS algorithm. Then, the maximum likelihood estimator can be constructed by the complex random vector and covariance matrix. Finally, using the appropriate initial values, the alternating projection is performed to estimate the optimum phase values. By using the simulation data and real SAR data, it is exhibited that we are able to retrieve credible phase values. A series of qualitative and quantitative results are presented to demonstrate the effectiveness of our method.

keywords InSAR, distributed scatterers, FaSHPS, alternating projection algorithm.

Yao-Estimation of the Optimum Phase for Distributed Scatterers_Cn_version.pdf


Source parameters for the 2015 Nepal Earthquake revealed by InSAR observations

Yongsheng Li1, Jingfa Zhang1, Zhenhong Li2

1Institute of Crustal Dynamics, China Earthquake Administrator, China; 2COMET,School of Civil Engineering and Geosciences, Newcastle University;

On April 25 2015, an Mw7.8earthquake occurred in Nepal, which is the largest since the 1934 Bihar Earthquake. 2015 Nepal earthquake occurred in subduction thrust interface due to collision of Indian Plate and Eurasian Plate, which is the main seismogenic fault structure named Main Himalaya Thrust fault (MHT).The earthquake released a huge amount of energy, causing huge losses of people and infrastructure.Using a combination of ALOS-2 ScanSAR wide mode data and SENTINEL-1A Interferometric Wide Swath data, we constructed maps of what happened on and below Earth's surface during the Mw7.8 earthquake in Nepal. Wide scan SAR data is very suitable for obtaining a wide range of deformation fields, especially for Nepal earthquake. The earthquake size is often described as slip over a larger fault area. The primarily model implies that the dimension of the fault is about 120 km × 80 km in size (length × width). The conventional Strip Map mode with maximum 100km coverage is unable to obtain the complete deformation region of Nepal earthquake. In this paper, we use the ALOS-2 and SENTINEL-1A wide scan data to obtain the wide coverage of the coseismic deformation field, then a two-step inversion strategy is employed to determine fault geometry and slip distribution, at last the seismic intensity surface motion is inverted from the distributed slip model. InSAR results show that the maximum uplift of ~1.3 m and the maximum subsidence of ~0.7 m, suggests that earthquake caused a huge ground deformation.

This paper uses coseismic interferogram as inversion constrained and discovers the details of fault geometric parameters and slip distribute of this earthquake. The source parameters and variable slip distribution are determined by a two-step inversion strategy. We applied equal-weighting for these discrete points in the modeling. Firstly, assuming the quake was caused by a single ruptured fault plane and determines the fault location and principal focal mechanisms with a uniform fault, and then a linear inversion is used to estimate the slip distribution along the fixed straight fault plane.The best-fit slip inversion model suggests that the major seismogenic fault is a thrust fault with a strike of ~291° , a dip of ~7.6° and an average rake angle of ~95°. The slip is mainly distributed of 12-18 km in the depth and of 140 km in length. The maximum amount of slip is up to 5.3 m in the depth of 15 km. The accumulative seismic moment was up to 6.5×1020 N﹒m, equivalent to a magnitude of Mw 7.8.

Li-Source parameters for the 2015 Nepal Earthquake revealed_Cn_version.pdf


Application of Combined GPS and InSAR Data for Estimating Three-Dimensional Surface Motion Maps of Beijing

Yi Luo1, Yunfeng Tian1, Yongsheng Li1, Jingfa Zhang1, Leyin Hu2

1Institute of Crustal Dynamics, China Earthquake Administration; 2Seismological Bureau of Beijing , China Earthquake Administration;

We gather continuous GPS network data (17 CGPS Stations,15 seconds sampling ) between Jan.2005 and April.2012 in Beijing area, and 180 scenes SAR data (143 scenes ENVISAT ASAR data between 2003 and 2010: Track490/Track447/Track218/Track354;37 scenes TerraSAR-X data between 2010 and 2011, Track008/Track009) ,which footprints covered Beijing area. GPS data process by GAMIT, and then we get three dimensional deformation and the perceptible water vapor(PWV) component of those sites. For SAR data , SBAS and PS InSAR process to acquire LOS deformation and Atmospheric phase screen (APS). Then we selecte data acquisition time and coverage which exists overlapping in the data set, and test through space and time interpolation and GPS deformation as the control constraints to get 3D surface deformation maps. The work have not been fully completed, so the abstract just introduce some problems in processing which we found.

Luo-Application of Combined GPS and InSAR Data for Estimating Three-Dimensional Surface Motion Maps of Beijing_Cn_version.pdf


Seismic Damage of Building Detection Based on Ground Deformation Statistics

Qiang Li1,2, Jingfa Zhang2, Tengfei Xue1,2

1Institute of Engineering Mechanics, China Earthquake Administration,China,People's Republic of;; 2Institute of Crustal Dynamics, China Earthquake Administration, China, People's Republic of;

Earthquake is one of the most serious natural disasters which cause the loss of people's lives and property. Strong earthquake often caused a large area of the building collapsed, The instantaneous deformation of the ground is one of the main reasons for the serious damage to the building, Especially the buildings along the fault zone. In part ,the instantaneous deformation of the ground caused the failure of the building foundation, resulting in sinking, tilting, collapse of the building, On the other hand, deformation caused the distortion of building structure, which caused damage to the main structure of the building. The damage caused by the stress of the main structure of the building which is beyond the capacity of the building. At present, the work of earthquake emergency assessment mainly depends on the traditional work mode, which can not meet the requirements of timeliness of earthquake disaster emergency. With the continuous development of space information acquisition technology, the advantages of remote sensing technology must be fully developed, and the existing work mode of earthquake emergency should be exceeded. Thus, Rapid acquisition of ground deformation variables based on remote sensing method can effectively improve the efficiency of seismic damage detection, and improve the ability of earthquake emergency and assessment.

InSAR(Interferometric Synthetic Aperture Radar) is a new technology in observing the earth from space ,which can quantitatively study the change of natural environment, and is a useful supplement to the traditional remote sensing technology and measurement methods with the advantages of high spatial resolution, high precision and surface matrix measurement. Regional surface deformation monitoring such as earthquake, ground subsidence, landslide, volcano movement, glacial activity and other surface is one of the widely used field, especially in the field of geological disaster monitoring. Great achievements have been made in the study of urban land subsidence, high-speed rail deformation, the crustal deformation rate of the key seismic belt, the ground subsidence and landslide activity and volcanic activity. However, there is little application in the detection of damage degree of buildings, and there are few visible results.

It is difficult to obtain the remote sensing image of earthquake region especially the meizoseismal region after a short period of time. Therefore, in the early stage of earthquake emergency, People are at the stage of evaluation blind for the damaged buildings and casualties. With the continuous improvement of radar satellite spatial resolution, measurement accuracy, revisit period and track control, fast acquisition of high precision surface variables can be achieved after the earthquake. Rapid detection and evaluation of building damage based on ground deformation by building the interrelation between surface deformation variable and damage degree of building effectively, which is conducive to the distribution of emergency rescue forces, rescue decision making and seismic damage assessment.

In this paper, we will collect and collate the information of surface deformation and damage degree of buildings, On this basis, The deformation field data and the damage degree of buildings are refined and the deformation variables are statistically analyzed, and finally the damage degree of buildings with different variables is extracted. The quantitative relationship between the deformation variables and the damage degree of building can be constructed and can be used to indicate the relationship between the damaged buildings and the ground deformation effectively based on the method of regression analysis, Taking Yushu earthquake as an example, the results are compared with the results of visual interpretation and the accuracy and usability of the quantitative relation are verified, so that it can play a role in the strong earthquake in the future.

Li-Seismic Damage of Building Detection Based on Ground Deformation Statistics_Cn_version.pdf


Monitering coastal area ground subsidence of Cangzhou city by using IPTA mehod

Tengfei Xue, Jingfa Zhang, Yi Luo

China Earthquake Administration, China, People's Republic of;

Since the 90s of the 20th century, European Space Agency, Japan, Canada, Germany and China have launched several SAR satellites, such as ERS-1/2, ENVISAT, Sentinel-1A, JERS-1, ALOS, RadarSat, TerraSAR and Cosmo-SkyMed, there are plenty of SAR data sources for further development of Geodesy research, the research about application of radar remote sensing in Geodetic reached an unprecedented climax. The differential interferometric synthetic aperture radar differential interferometry synthetic aperture radar, i.e. DInSAR technique can obtain as high accuracy ground surface deformation information as traditional measurement method. However, affected by decoherence due to time/spatial baseline and atmosphere effect , DInSAR technology is limited in monitoring slow surface deformation of long time period, therefore the research focus gradually shifted to the long time series of interferometric synthetic aperture radar measurement, i.e. InSAR time series analysis methods. A variety of algorithms are proposed such as coherent point target extraction algorithm, 3D phase unwrapping algorithm, the error removal algorithm. InSAR time series analysis method of the technology is more and more comprehensive, and the measurement accuracy is getting higher and higher. InSAR time series analysis can obtain the surface deformation information of a long time series, it has already been used to urban groundwater extraction caused by settlement deformation, surface mining subsidence and fault activity monitoring and many other fields.

In 2003, Werner, and Wegmuller et al proposed the interferometric point target analysis method, by analyzing phase characteristics of point target in time and space on interferogram, the interference graph deformation phase, the topographic phase, orbit error phase and air phase can be separated effectively and precisely. Method mainly comprises the following steps: 1) select the appropriate master image, and all SAR images are registered based on the master image; 2) candidate target points extraction by using single image spectrum stability and the scattering intensity time series stability; 3) estimate target point elevation error and linear deformation variable using two-dimensional linear recursive model as well as time and spatial baseline; 4) obtain unwrapped interferometric phase images, and correct satellite orbit baseline, in order to obtain a more accurate deformation variables and elevation error; 5) separate nonlinear deformation and atmospheric phase delay by filtering of the interferometric phase residual. The IPTA method obtains the inversion results including elevation error, linear deformation rate, atmospheric delay, phase correction, baseline satellite quality evaluation index and nonlinear deformation time series.

In this paper, the coastal area of Cangzhou City, Hebei Province, China is chosen as a test area, and 18 SAR images during 2008-2010 are processed by using IPTA and SABS time series analysis method, while annual average deformation rate map was generated. Comparing IPTA and SABS results, it is found that the coherence point targets extract from study area using IPTA method is less than SBAS method, especially in the center of subsidence area which SBAS monitored. But there are more points in Huanghua City, Zhongjie farm living area and construction area along the coast by using IPTA method, and the subsidence rate is similar with the results using SBAS method. Analysis shows that the point extraction module of IPTA method is far different with SBAS method, and the calculation model of these two methods has significant difference. IPTA method uses master image as a benchmark respectively generate interferometric images, and SBAS uses small baseline combination to generate interferometric images.

Line charts of two points in Huanghua city extracting and calculating by IPTA and SBAS method is compared, the result shows that the deformation trends of these two methods are similar, and the subsidence rate is close, which indicates that the results using IPTA and SBAS method are consistent in this test.

Xue-Monitering coastal area ground subsidence of Cangzhou city_Cn_version.pdf
Xue-Monitering coastal area ground subsidence of Cangzhou city_ppt_present.pdf


Present Geodetic Slip Rate of Beng Co Fault Zone in Central Lhasa Block, Tibet

Yunfeng Tian, Yongsheng Li, Yi Luo, Wenliang Jiang, Hongbo Jiang, Wenhao Shen, Qisong Jiao, Dehua Wang, Qiang Li, Jingfa Zhang

Institute of Crustal Dynamics, China Earthquake Administration, China, People's Republic of;


Synchronous with a series of large scale S-N rift systems in southern Tibet, the “chord” connecting the eastern and western Himalayan syntaxes, which is known as KJFZ (Karakorum Jiali Fault Zone), is supposed to experience rapid (10-20mm/a) right-lateral strike slip (Armijo et al., 1989), accommodating the E-W extension and S-N shortening of Tibetan plateau between Himalayan arc in the south and Tarim basin in the north. The Jiali fault and Beng Co fault consist of the easternmost segment of KJFZ, and they interplay with the Yadong-Gulu rift near the Gulu county. Several strong earthquakes occurred in this region. For example, the 1951 M8 Beng Co earthquake which produced a maximum right-slip seismic offset of ~7.3m.

However, because of the harsh environment and the sparse modern geodetic network in this region, the recent activity of Beng Co fault zone is not well known, especially for the southeast segment (i.e. southeast the Beng Co lake). Therefore, we use both GNSS (Global Navigation Satellite System) and InSAR (Interferometric Synthetic Aperture Radar) techniques to study the current activity of Beng Co fault system, which combines the advantages of both the high-accuracy three dimensional positioning from GNSS and the area coverage of LOS velocity from InSAR.


We setup three campaign GNSS sites (LS08, LS09, and LS10) along the southeast Beng Co fault segment in 2013, and installed another two sites (LS22 and LS23) at the two sides of the northwest segment in 2014. All stations are located on bedrock and supposed to represent the motion of the underlying crust. We already carried out 2-3 sessions of field observations for several stations, with a minimal continuous observation of 2-3 days in each session. In addition, the GNSS data from Chinese “Lutai” observation project are also included in our analysis. The raw GNSS data were processed with GAMIT/GLOBK software with a conventional processing strategy. After deriving the daily positions for each session, the linear rates are estimated using the least square method.

At the same time, we also use InSAR time series analysis to derive the relative motion between the hanging wall and footing wall of Beng Co fault zone. Data from ENVISAT ASAR, ALOS-1, and Sentinel-1A are acquired. A phase closure method is used to detect and correct phase unwrapping errors for discrete coherent points in InSAR time series analysis. A test analysis using ASAR descending data during 2007-2010 was carried out to derive the crustal motion in this fault zone.


The GNSS profile across the northwest Beng Co fault show a 3.5mm/a right-lateral strike slip rate, with ~1mm/a crust shortening in the S-N direction. While across the southeast segment of Beng Co fault, the right-strike slip rate is ~1.3 mm/a, together with a 2~3 mm/a shortening rate perpendicular to the fault trace. Our InSAR derived rate estimates are consistent with GNSS results, with ~1.5mm/a and ~2mm/a right-lateral strike-slip rates for the southeast and northwest segments, respectively. Such a result does not agree with some previous conclusions, e.g. Garthwaite et al. (2013), which suggests a larger slip rate for the southeast segment and attributes it to the postseismic displacements following the 1951 M8 event. Furthermore, existing GNSS velocity map show a compressional environment along the Beng Co fault zone, which agrees with the squeezed V-shaped wedge model proposed by Yin and Taylor (2011) but is inconsistent with the extension found by field survey for the 1951 M8 quake.

To date, the above results are still rather preliminary because of the sparse GNSS network in this area and short observation time span (1-2 years for our new sites). Therefore, we are planning to install more sites and conduct longer field observations, with the aim to figure out the current activity of Beng Co fault zone.


We thank Ying-Shun Hu, the head of Naqu Seismic Station, for his kind assistance in our field work and station maintenance. Graduate student Ou Ku from Delft University helped collecting GNSS data in 2015. This work is funded jointly by National Science Foundation of China (no. 41204004, 41104001) and Institute of Crustal Dynamics (no. ZDJ2013-22).


Armijo, R., P. Tapponnier, and T. Han (1989), Late cenozoic right-lateral strike-slip faulting in southern Tibet, J. Geophys. Res., 94(B3), 2787–2838, doi:10.1029/JB094iB03p02787

Garthwaite, M. C., H. Wang, and T. J. Wright (2013), Broadscale interseismic deformation and fault slip rates in the central Tibetan Plateau observed using InSAR, J. Geophys. Res. Solid Earth,118, 5071–5083, doi:10.1002/jgrb.50348

Yin, A., and M. H. Taylor (2011), Mechanics of V-shaped conjugate strike-slip faults and the corresponding continuum mode of continental deformation, Geol. Soc. Am. Bull., 123(9–10), 1798–1821

Tian-Present Geodetic Slip Rate of Beng Co Fault Zone in Central Lhasa Block, Tibet_Cn_version.pdf


Highway subsidence analysis based on the advanced InSAR time series analysis method

Qingyun Zhang1,2, Jingfa Zhang2, Yongsheng Li2

1Institute of Crustal Dynamics, China Earthquake Administration, China, People's Republic of; 2Key Laboratory of Crustal Dynamics, Institute of Crustal Dynamics, China Earthquake Administration, Beijing, China;

Abstract: The synthetic aperture radar (InSAR) measurements are developed to be a potential new microwave remote sensing technology in recent years. It have the advantages of all-weather, wide range, high precision on the surface deformation monitoring. Highway as an important index of modern social and economic development, the quality and deformation changes in the process of using have a significant impact in the social development and people's life and property security. In practical applications the InSAR technology should do a variety of error correction analysis. This paper, by using a new analysis method – FRAM-SBAS time-series analysis method, to analyze the settlement of highway on Yanzhou area by the ALOS PALSAR datas. FRAM - SBAS time-series analysis method is a kind of short baseline of set time series analysis strategy which combined with a variety of InSAR error sources correction method. By comparing the FRAM - SBAS time-series analysis method and the traditional PS-InSAR method, proved that the advanced InSAR time-series analysis method have the advantages of simple and high accuracy, and verified the different degree of ground subsidence of Yanzhou due to coal mining and so on.

Keywords : FRAM-SBAS time series analysis, PS-InSAR, Highway subsidence

1. research status

With the development of economy and society, as well as the implement of the national strategy <The national highway network planning>. The highway construction is booming. But restricted by the landform and road grade, some highway should through underground goaf left by resources exploitation. Because of the existence of mined-out area these area will be formed on the surface subsidence basin, and easy to produce the phenomenon such as cracks and fissures, which bring great negative impact on the stability of the roadbed. And in a different location, because the influence of actual conditions such as the traffic, vehicle loads and geomorphic factors. Roadbed will appear different degree of damage in the process of using. Therefore, settlement monitoring of highway and to make sure the normal operation of highway is becoming more and more important.

PS-InSAR technology can identify point target scattering characteristics (PS) which with stable coherent. Set up the phase coherent point model, using mathematical method to separate the surface deformation information, imprecise DEM contribution and contribution to the atmospheric delay signal [4].This method can avoid the time coherence and space coherence, it was carried out stack analysis on the time series of SAR images (interferogram), realize inversion of surface deformation time series. But this method is a more complex InSAR processing method, it needs at least 30 scene data. So, at the less on data area, is unable to use this method.

FRAM-SBAS (Full Rank Matrix SBAS InSAR) time-series analysis method is based on full rank conditions of the sequential inversion matrix. By the least square method to get full rank coherent point by the inversion of deformation field, and then use the network do analyze. Not only can keep the resolution of SAR image, but also could not cause the missing of strong isolated scatterer.

At the same time the data requirement is lower than the PS - InSAR, only need more than 15 scene can get higher accuracy of the surface deformation analysis result. This method have a lot of advantages and potential in practice use.

The key steps of analysis method for sequence of FRAM-SBAS in use process are: 1) selecting specific baseline of time and spatial as the constrained conditions in the generation of interference graph, choosing suitable interferograms , to improve the coherence of the interferogram. 2) the principle of full rank matrix is used to select the coherent points to ensure the quality of the coherent points. 3) the atmospheric errors are corrected by the network method, which can eliminate the inconsistency of the atmospheric delay in space. 4) the method of closed loop error is used to correct the phase unwrapping error of coherent point, so as to improve the quality of unwrapping phase. 5) using the least squares method to solve the unwrapping interference atlas, which can get timing deformation of each coherent, so as to improve the quality of deformation monitoring.

2. research significance

The biggest feature of highway is the strip linear engineering, when detecting and analyzing the features, we need to meet some of the requirements: first, the most regional coverage, for the road is complex and we need common reference when monitoring, and simultaneous measurement of a wide range of requirements. Second, the requirements of the InSAR technology, this project requires the application of high precision InSAR data processing algorithm, the high resolution of fine detection. FRAM-SBAS timing analysis method is a new sequential inversion method is, which not only improved the dry point extraction method, and also improved the atmospheric correction and phase unwrapping wrapped around the method, and has verified the method in monitoring the earthquake deformation of application. The method not only has the advantages of low, high precision of deformation monitoring data demand requirements, and can also satisfy the deformation monitoring of highway high coverage, and the needs of high precision and high efficiency, so we consider using this method in Yanzhou area highway for deformation monitoring, contrasting with traditional PS InSAR method to carry on the contrast analysis, to make the validation of the method feasibility and the settlement characteristics of Yanzhou area.

Zhang-Highway subsidence analysis based on the advanced InSAR time series analysis method_Cn_version.pdf


Ground deformation monitoring based on CR-InSAR and time series InSAR

Yingxu Song1,2, Ruiqing Niu1, Jingfa Zhang2, Yongsheng Li2

1China University of Geosciences(Wuhan); 2Institude of Crustal Dynamics,China Earthquake Administraction, China, People's Republic of;

Ground deformation monitoring based on CR-InSAR and time series InSAR

Yingxu Song1,2,Ruiqing Niu1,Jingfa Zhang2,Yongsheng Li2,Yi Luo2,Yan Chen2

1.China University of Geosciences(Wuhan),430074

2.Institute of Crustal Dynamics, China Earthquake Administration,100085

The current differential interferometric synthetic aperture radar(DInSAR)technique cannot work well in measuring the long—term accumulated small crustal deformation because of serious constraints from the phase decorrelations in time and space,which heavily influence the detecting precision of the ground deformation monitoring. We know that hard target such as power lines, buildings and metal surface can cause strong corner reflection effect, enhancing the echo power of the radar wave velocity,which characterized by bright spot or bright line on the radar images. Taking these hard targets as coherent points can greatly improve the accuracy of InSAR deformation measurement.

At present, the CR technology has been widely used in the area where phase decorrelations in space and unstable terrain in research, especially in urban areas of geological disaster monitoring research. In the crustal deformation observation, CR can be installed on both sides of the fault as a supplement form for high precision control points of the network backbone.

This article will use the AlOS-1/ASAR image data and GPS data to calculate the deformation of the Qinghai-Tibet Plateau. Considering the Qinghai-Tibet Railway, highway, transmission lines and other national key projects and its subsidiaries (construction) or natural structures as corner reflector (CR) or persisten scatterers (PS).Based on these PS or CR points together with GPS stations, build the InSAR/GPS deformation monitoring network for long linear regions under a unified framework and analyse the deformation time series and the long line of section deformation comparison studies. According to deformation analysis, divide and monitor the super-deformed dangerous areas of the national key projects such as Qinghai-Tibet railway, highway, transmission lines forearly warning. At the same time, for deformation monitoring of the national key projects, classified dangerous degree also has important practical significance and social value.

Key words:CR;time series analysis;Ground deformation monitoring;Qinghai-Tibet Plateau;decorrelation

Song-Ground deformation monitoring based on CR-InSAR and time series InSAR_Cn_version.pdf


Development and Implementation for Calculation Model of Measuring Co-seismic Deformation Field by Using Ascending and Descending Orbit SAR Data

Tengfei Xue, Jingfa Zhang

Institute of Crustal Dynamics, China Earthquake Administration;

As a promising remote sensing technique, Interferometry Synthetic Aperture Radar (InSAR) has been developed rapidly in recent years. Its advantages of requiring ground deformation have made it possible for researchers to monitor ground surface from the space. However, this technology has a limitation that an interferogram can only measure one component of the surface deformation in the satellite's line of sight (LOS) instead of that in vertical and horizontal directions, i.e. LOS Amphibious. Especially for the earthquake, it is necessary to reveal the three-dimensional (3D) deformation of ground surface.

At present, there are many methods to extract 3D deformation by using InSAR. The main methods include as following:

(1)Azimuth and Range pixel offset: This method mainly uses pixel offset technique to generate the deformation along azimuth and range direction, and then both two direction deformation are calculated to generate 3D deformation.

(2)D-InSAR and azimuth pixel offset: This method mainly uses pixel offset technique to generate the deformation along azimuth direction, and uses D-InSAR technique to generate the deformation along line of sight. Then both two direction deformation are calculated to generate 3D deformation.

(3)D-InSAR and MAI: This method mainly uses multiple aperture InSAR(MAI) technique to enerate the deformation along azimuth direction, and uses D-InSAR technique to generate the deformation along line of sight. Then both two direction deformation are calculated to generate 3D deformation.

(4)Left and right look InSAR: This method mainly uses InSAR pairs with different look direction, such as left look and right look, to generate four interferogram images, and the the 3D deformation is calculated using this four images.

Through the above four methods, we can see that in order to obtain three components of surface displacement, there must be at least three equations to calculate equations. And conventional D-InSAR interferogram can only extract the displacement along LOS, that isonly one equation, even if the use of ascending and descending orbit data can only get two equations; and data with three different incident angles and azimuth angles is very rare, so the common method is to obtain the deformation along azimuth direction and calculate with D-InSAR results to generate 3D deformation.

Based on the analysis above, we analyzed and summarized some methods that can measure the three-dimensional deformation of ground surface by using D-InSAR, developed the calculation model of measuring the three-dimensional co-seismic deformation filed by using the ascending and descending orbit SAR data, and applied this model by using ENVI, SARscape as well as Matlab software. Two main parts are included: (1) measuring surface deformation along the LOS directions and the satellites' azimuth directions by using conventional D-InSAR and Multiple Aperture InSAR (MAI) technique; (2) calculating the three-dimensional(3D: up, north and east) co-seismic deformation filed by using the methods of weighted least square. The achievements of this thesis are as follows:

(1) The Formula of left-looking (both ascending and descending orbit data), right-looking (both ascending and descending orbit data) and general expression were proposed;

(2) The subset method was applied on the earthquake struck at Italy, on April 6th, 2009 calculating the coincident area of ascending and descending SAR images;

(3) The model was applied on L'Aquila earthquake, and the results reveal that the earthquake has caused displacement in both vertical and horizontal directions, and the earthquake made the area down lift 16.8cm along the vertical direction. The characters of the surface reflected by the results are very consistent with the geological exploration.

Xue-Development and Implementation for Calculation Model_Cn_version.pdf
Xue-Development and Implementation for Calculation Model_ppt_present.pdf


Applied Reserch of In SAR Technology in Deformation Monitoring in Mining Subsidence

Qingyun Zhang1,2, Jingfa Zhang2, Yongsheng Li2

1Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration; 2Laboratory of Crustal Dynamics, Institute of Crustal Dynamics, China Earthquake Administration;

Applied Reserch of In SAR Technology in Deformation Monitoring in Mining Subsidence
Qingyun ZHANG1,2 , Jingfa ZHANG2,Yongsheng LI2
1. Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration, Harbin, China
2. Key Laboratory of Crustal Dynamics, Institute of Crustal Dynamics, China Earthquake Administration, Beijing, China
InSAR (Synthetic Aperture Radar Interferometry) and the D-InSAR (Differential Syn-thetic Aperture Radar Interferometry) as new microwave remote sensing technology, have developed rapidly in recent years, which has all-weather, all- covering a broad area with a high degree of automation and high-precision monitoring ground deformation capacity. More and more researchers are aimed at the study of the application field of the InSAR technology, and the technology was studied and improved, and put forward the differential interference technology such as PS-InSAR、SBAS-InSAR. It is applicated in various aspects, such as the earthquake emergency analysis 、the surface subsidence analysis and fault analysis, and were obtained high precision analysis results. In the surface deformation monitoring, the theory of value can be accurate to mm.
In reality, the large-scale exploitation of underground coal often causes the problem of serious mining subsidence. By rock and surface displacement and deformation, mining subsidence damages the farmlands, buildings, traffic and other facilities of mine, bring adverse effect to production, life and even sustainable development of mining area. The mining subsidence areas may cause serious influence the normal production of the mining area, at the same time on the surrounding residents property will exist unexpected hidden trouble. Timely monitoring and deep analysis of spatio-temporal characteristics of mining subsidence are helpful to finding out its causes and harms, and play an active role in its prevention and treatment.
In this paper, the Yanzhou area is chosen as the test area, and the ground subsidence caused by mining exploiting in Yanzhou area is monitored and studied in detail. Applying the PS-InSAR technique and improved SBAS-InSAR technique and the ten over pictures of ALOS PALSAR data to detect the subsided areas and analysis the raw data. After interference images generation, adaptive filter and coherence generation, unwrapping and geographic coding. To get the ground subsidence monitoring map of Yanzhou area. According to the final result, the ground subsidence degree and the changing tendency of Yanzhou were analyzed.
This paper compares the distribution of the ground subsidence in Yanzhou area with the distribution of mineral resources and the groundwater exploitation. Sedimentation volume rendering in color gradation in the studied area, to get the intuitive visual interpretation of settlement in the study area. Analyzed the merits of the two methods of surface deformation monitoring. Results show that both methods in yanzhou area subsidence monitoring has been relatively consistent with the monitoring results, Yanzhou of subsidence areas and distribution of coal mining area. According to the result of analysis shows that the current PS-InSAR technology and improved SBAS-InSAR technology in Yanzhou area, the ground subsidence monitoring has a higher application value, is one of relatively good monitoring methods.
Keywords : mining subsidence, InSAR technique, deformation monitoring

Zhang-Applied Reserch of In SAR Technology in Deformation Monitoring in Mining Subsidence_Cn_version.pdf


Systematic improvement of PSI geolocalization using a single acquisition corner reflector

Mengshi Yang1,2, Prabu Dheenathayalan1, Ling Chang1, Mingsheng Liao2, Ramon F. Hanssen1

1Delft University of Technology, Netherlands, The; 2Wuhan University,China;

Persistent Scatterer Interferometry(PSI) approach is an effective tool for measuring the kinematic behavior of ground targets with millimeter precision (Ferretti et al. , 2001). Compared to such a high precision of estimated kinematic behavior, the positioning precision of InSAR measurement points (i.e. PS) is relatively poor, which limits the operational applicability of satel- lite radar interferometry. Whereas it is demonstrated to be possible to find the position of a target on earth in a corresponding radar image (hence, go- ing from 3D coordinates to 2D coordinates), see e.g. (Eineder et al. , 2011; Dheenathayalan et al. , 2016), the inverse procedure (going from a point target in the radar image to the corresponding 3D position on earth) is not possible in an absolute sense.

Standard PSI processing estimates relative height differences between PS points. However, these heights are relative to an arbitrary reference point. To obtain precise absolute positions, the position of the reference point should be known, which is generally not the case. The geocoding process in PSI processing assumes that scatterers are located on the used digital elevation model. Consequently, since the reference point could be situated at some (tens of) meters elevation from the elevation model, we have horizontal errors in geocoding in the order of tens of meters as well.
In this investigation, we propose a pragmatic, efficient and effective way to improve the geolocalization of point scatterers, which helps us to make a better interpretation of the PSI results. We use a single corner reflector, deployed in only one SAR acquisition, and use the GNSS derived position, together with the sub-pixel position location in the SAR image, to correct the entire point cloud of PS points in 2 dimensions. Then, we correct for the last (cross-range) dimension using a high resolution digital surface model (DSM), using a cross-correlation between all PS points and the DSM.
We demonstrate our method in the city of Delft, the Netherlands by using both high and medium resolution data (i.e.~TerraSAR-X and Radarsat-2 data). Eventually, we show the improved PS geolocation in a 3D environment and link every PS point to the corresponding ground object.

Key words: Sub-pixel positioning, Data Visulisation


Dheenathayalan, Prabu, Small, David, Schubert, Adrian, & Hanssen, Ra- mon F. 2016. High-precision positioning of radar scatterers. Journal of Geodesy, 1–20.

Eineder, Michael, Minet, Christian, Steigenberger, Peter, Cong, Xiaoying, & Fritz, Thomas. 2011. Imaging geodesyToward centimeter-level rang- ing accuracy with TerraSAR-X. Geoscience and Remote Sensing, IEEE Transactions on, 49(2), 661–671.

Ferretti, Alessandro, Prati, Claudio, & Rocca, Fabio. 2001. Permanent Scat- terers in SAR Interferometry. IEEE Transactions on Geoscience and Re- mote Sensing, 39(1), 8–20.

Hanssen, Ramon F. 2001. Radar Interferometry: Data Interpretation and Error Analysis. Dordrecht: Kluwer Academic Publishers.

van Leijen, F.J. 2014. Persistent Scatterer Interferometry based on geodetic estimation theory.


Land Subsidence Monitoring Along Intercity High-speed Railway in Beijing Area Using Multi-platform SAR Imageries

Mi Chen1,2, Zhenhong Li2, Huili Gong1, Xiaojuan Li1

1Capital Normal University, China, People's Republic of; 2Newcastle University, UK;

Land subsidence is a severe hazard threating the safety of urban infrastructure. Beijing and Tianjin mega-cities have been suffering from land subsidence for ages, mainly due to over-exploitation of ground water. The project of Beijing-Tianjin intercity high-speed railway completed in 2008 plays an important role in these two major cities of China. It is well known that a tiny displacement of the rail could result in serious consequences. Therefore continuous monitoring of land subsidence along the Beijing-Tianjin intercity high-speed railway is vital to maintain its safety operation.

In this study, 41 Envisat ASAR images acquired between 2003 and 2010, and 45 TerraSAR-X stripmap images collected from 2010 to 2015 were used to investigate the land subsidence along the Beijing-Tianjin intercity high-speed railway, particularly in the Beijing region. Small baseline InSAR was applied to obtain land deformation information. The experimental results show that the intercity high-speed railway passes through the edge of settlement centre of Dongbalizhuang-Dajiaoting over eastern suburbs with a subsidence rate greater than 50 mm/yr. Comparisons of InSAR-derived subsidence rates from multi-platform SAR datasets show a high correlation coefficient with a small RMS difference, indicating the reliability of InSAR retrievals of land subsidence rates. GPS derived subsidence rates were also used to compare with InSAR retrievals, and a high consistency was also obtained, which provides additional supporting evidence for the robustness of InSAR retrievals.

The spatio-temporal evolution characteristics of land subsidence and the non-uniform deformation along the Beijing-Tianjin intercity high-speed railway were analyzed together with hydrogeological data. The experimental results indicate the different characteristics of different frequency InSAR time series analysis and the complementarity of X-band and C-band SAR data for land subsidence monitoring along high-speed railway. C-band Envisat ASAR data provided land subsidence information with large spatial scale and X-band TerraSAR-X data provided localized information with a finer scale. The combined analysis of C-band with X-band data helped depict the subsidence funnel and reveal the causes of different subsidence rates, suggesting that it should be an effective technique to monitoring of land subsidence related to high-speed railway. The complete comprehension of the land subsidence on both large and small scales along high-speed railway is critical for predicting potential hazards and designing compensation strategies.

Chen-Land Subsidence Monitoring Along Intercity High-speed Railway_Cn_version.pdf


InSAR Technique is Applied to the Monitoring of the Qinghai-Tibet Railway

Jingfa Zhang, Yi Luo, Yongsheng Li

Institute of Crustal Dynamics, China Earthquake Administration;

The Qinghai-Tibet railway is the symbol of the national western development project, has important social and economic significance.But the railway crossing an average elevation of 4500 meters of the Tibetan plateau, the world's highest railway engineering, face frozen, active fault, seismic and geological disasters long-term threat, disaster reduction task is very heavy.

This paper choose some active fault zones and the qinghai-tibet railway intersection of Damxung - Yambajan region, mainly in the following several aspects work:
(1) active fault survey: by remote sensing interpretation, find out the active fault along Qinghai-Tibet railway spatial location, distribution, occurrence, mechanical property, direction of motion.
(2) the fault induced geological hazard survey, using remote sensing data, analysis of fault induced deformation of tectonic cracks, uneven freezing, etc. The structure of the geological disasters, forming conditions, evolution process, activity patterns.
(3) the surface deformation monitoring: using the accumulated ASAR, TerraSAR and Alos data, InSAR time-series analysis, determine the permafrost change with the seasons, on this basis, the analysis of the laws of fault activities.Using relevant results, but also determines the along Qinghai-Tibet railway and ancillary facilities for the recent several strong earthquakes in the near and far field of reaction.
Relevant results provide an important basis for prevention and control of disaster occurred.


Measurement of subsidence in the Yangbajain geothermal fields, Tibet, from InSAR time series analysis

Yongsheng Li1, Jingfa Zhang2, YI Luo3

1China Earthquake Administrator, China,; 2China Earthquake Administrator, China,; 3China Earthquake Administrator, China,;

Yangbajain contains the largest geothermal energy power station in China. Geothermal explorations in Yangbajain first started in 1976, and two plants were subsequently built in 1981 and 1986. A large amount of geothermal fluids have been extracted since then, leading to considerable surface subsidence around the geothermal fields. In this paper, InSAR time series analysis is applied to map the subsidence of the Yangbajain geothermal fields during the period from December 2011 to November 2012 using 16 senses of TerraSAR-X stripmap SAR images.

Due to its high resolution and short repeat cycle, TerraSAR-X provides detailed surface deformation information at the Yangbajain geothermal fields. The main conclusions of this work can be summarized as follows.

(1) Taking into consideration only the absolute deformation, except for the effect of SFG, we obtained a subsidence rate map and time series for the Yangbajain geothermal fields. The results

show two different patterns of the subsidence related to the geothermal fields. The first subsidence region next to power plant I shows an irregular subsidence pattern which is consistent with the presence of the road and pipelines. Over the 11-month period of monitoring, the maximum subsidence level reached was 25 mm. The second subsidence region is next to power plant II. The subsidence pattern is bowl shaped with a maximum of up to 8 mm.

(2) Taking into consideration influence of the SFG, it can be assumed that the long wavelength

deformation is caused by the seasonal changes in SFG. The ground surface reaches its maximum subsidence of −5 mm between August and September and its maximum uplift of 5 mm between February and March, which is highly correlated with the change in surface temperature.

(3) To continue sustainable development of the geothermal power plants, the following works are urgently needed. Firstly, a balance must be ensured between the fluid production rates and the recharge rates by injecting of fluids to compensate for the inadequacy of natural replenishment. Secondly, the deep geothermal reservoir must be exploited in order to reduce the usage of shallow thermal storage. Finally, the continuous monitoring of ground surface subsidence near the power stations and the surrounding regions must be carried out using geodetic surveys (GPS, InSAR etc.)


Systematic Anomaly Detection for Rails and Embankments using Satellite Radar Interferometry

Freek van Leijen, Ling Chang, Ramon Hanssen

Delft University of Technology, The Netherlands;

Satellite radar interferometry (InSAR) has the capability of monitoring rails and embankments over wide areas. It has been demonstrated that the sub-centimeter-scale deformation of millions of InSAR measurements over railway infrastructure can be measured routinely using InSAR. Yet, to handle such huge data volumes and to recognize anomalies (such as localized differential deformation) in an efficient way, still limits the potential for operational use in wide areas, e.g. for monitoring a nation-wide railway network.

In this work, we develop and demonstrate a systematic InSAR methodology that can scrutinize data and automatically detect anomalies. The method is mainly based on statistical testing theory. Particularly, we use a ‘short arc’ method to focus on detecting localized differential deformation between two nearby InSAR measurement points over the railway. Our approach is applied to the entire railway network of the Netherlands, with a total route length of more than 3000 km. We used 210 Radarsat-2 descending data from three different tracks which were acquired between 2010 and 2015. A differential deformation and anomaly map are produced. This method will be further investigated for all railways in China.

This research is supported by the Young Research Scientists Support Program, in the framework of the Dragon cooperation 2013 – 2016 (Dragon 3).


Monitoring road stability in permafrost environment with InSAR stacking method

Keren Dai, Guoxiang Liu, Deying Ma, Bing Yu, Xiaowen Wang

Southwest Jiaotong University, China, People's Republic of;

The cyclic freezing-thawing process is one typical characteristic of permafrost area, which would bring geologic hazard to the nearby infrastructure, such as the rapid deformation to the road or railway. Previous Leveling results show that in Qinghai province there are rapid and cyclic damage of ground deformation for the G214 national road in permafrost environment. Due to the severe weather and high altitude there, it is difficult to perform conventional manual monitoring to the G214 road. Time series InSAR methods such as PS-InSAR and SBAS were widely used in the last twenty years, which could provide the accurate ground deformation covering wide area. However, the performance of these traditional time series InSAR method is not good due to the low coherence and rapid deformation rate of this area. It was found that the spatial deformation gradient would be lost if only analyzing the persistent scatter (or other type of high coherent) points.

This paper will use InSAR stacking method to monitor the ground deformation along this road with TerraSAR-X(TSX) and Sentinel-1(S1) images. To keep the relatively high coherence area as wide as possible, the interferograms would be formed only from the differential pairs with shortest time baseline(i.e. 11days for TSX and 22days for S1). As a result, the deformation gradient of the whole area would be kept in these interferograms with high coherence. With use of the high precision DEM, the topographic effect can be totally removed. It should be noted that according to the previous leveling data and the interferograms results, the deformation in permafrost area is cyclic. From September to the next April, the permafrost would be freezing while in May to August it would be thawing. So the stacking process should be separately performed to evaluate the mean deformation rate separately.

The final results show that there is one subsidence bowl in this area, through which the road pass. The max subsidence rate reach to 30 cm/year from June to August, while the subsidence rate of other part is about 5~15 cm/year. The geology of this subsidence bowl is different from other parts from the . From the interferograms it can be conclude that the relative deformation of different area is totally inversed from September. The meteorological data verify that the road temperature become below 0℃ since September. It means the freezing process become active instead of the thawing process. As the InSAR technology measures the relative deformation, the stacking method show a better result to keep the whole deformation gradient in the whole area. The subsidence profile of the road was verified by leveling data, which could provide value information for road maintenance and preventing geologic damages.

Contact and Legal Notice · Contact Address:
Conference: 2016 Dragon 3 Final Results Symposium
Conference Software - ConfTool Pro 2.6.94
© 2001 - 2016 by H. Weinreich, Hamburg, Germany