Subsidence monitoring in coal area using time-series InSAR combining persistent scatterers and distributed scatterers

doi:10.1016/j.jag.2015.02.007

International Journal of Applied Earth Observation and Geoinformation

Volume 39, July 2015, Pages 49-55

https://doi.org/10.1016/j.jag.2015.02.007 Get rights and content

Highlights

•
Both classification and statistical characteristics are applied for DS identification.
•
Two-scale network on PSs and DSs is proposed to improve the computational efficiency.
•
Spatial density of measurement points increases greatly especially in natural area.

Abstract

In coal mining areas, ground subsidence persistently happens, which produces serious environmental issues and affects the development of cities. To monitor the ground deformation due to coal mining, a modified time-series InSAR technique combining persistent scatterers (PSs) and distributed scatterers (DSs) is presented in this paper. In particular, DSs are efficiently identified using classified information and statistical characteristics. Furthermore, a two-scale network is introduced into traditional PSI to deal with PSs and DSs in a multi-layer framework by taking the advantage of the robust of PSs and the widely distribution of DSs. The proposed method is performed to investigate the subsidence of Huainan City, Anhui province (China), during 2012–2013 using 14 scenes of Radarsat-2 images. Experimental results show that the proposed method can ease the estimation complexity and significantly increase the spatial density of measurement points, which can provide more detailed deformation information. Result shows that there are obvious subsidence areas detected in the test site with subsidence velocity larger than 5 cm/year. The proposed method brings practical applications for non-urban area deformation monitoring.

Introduction

With the rapid development, the demand for mining underground energy resource becomes more and more enormous in China. The exploitation of underground coal often causes persistent land subsidence, which would lead to huge threats to infrastructure and safety in the coal area (Jung et al., 2007, Baek et al., 2008). Subsidence monitoring and surveys in coal areas are urgently needed, which can provide knowledge for the hazard warning system. Traditional leveling and GPS data are able to produce reliable measurement of ground subsidence (Demoulin et al., 2005). However, these field surveys are time-consuming and cannot provide deformation map with high spatial sampling density.

Synthetic aperture radar (SAR) interferometry (InSAR) is a powerful technology, which can obtain high precise elevation and surface deformation along the line of sight (LOS) using phase information from different SAR images. DInSAR technique has been applied to monitor the surface movement in coal mining areas (Ng et al., 2009, Ge et al., 2007). However, conventional DInSAR usually suffers from temporal, geometrical and atmospheric decorrelation (Zebker et al., 1997, Zebker and Villasenor, 1992). To overcome these limitations, several advanced techniques have been proposed, such as Persistent Scatterer Interferometry (PSI) (Ferretti et al., 2001, Hooper et al., 2004), small baseline subset algorithm (SBAS) (Berardino et al., 2002), coherent pixels technique (CT) (Mora et al., 2003), and partially coherent targets technique (P-CTs) (Prati et al., 2010, Tao et al., 2012). These techniques have been applied in several studies to retrieve ground movement due to underground mining (Ng et al., 2010, Perski et al., 2009, Jiang et al., 2012, Zakharov et al., 2013, Guéguen et al., 2009). However, these methods have some limitations in mapping the mining-induced subsidence due to low density of PSs or coherent points and their inhomogeneous distribution. In recent years, distributed scatterers interferometry (DSI) (Guarnieri and Tebaldini, 2008, Ferretti et al., 2011, Goel and Adam, 2014, Wang et al., 2012) has been proposed to monitor the surface deformation of large scale suburb areas. Most coal areas are located in suburbs areas, in where man-made objects are too sparse to obtain enough measure points for PSI. But the bare soil areas, land covered by sparse vegetable or debris areas (referring them as DSs) are widely distributed, which can be selected as measure points for subsidence monitoring.

In this work, we present an approach to obtain ground subsidence due to mining in Huainan City, China using time-series InSAR technique combining PSs and DSs by taking the advantage of the robust of PSs and the widely distribution of DSs. To increase the spatial density of measure points, a new processing method using both classified information and statistical characteristics is proposed to identify DSs. To decrease the computationally difficulty and control error propagation, a two-scale network is introduced in the processing of PSs and DSs, which deals with PSs and DSs in a multi-layer schema from high phase quality to low. Linear deformation rate and DEM error of PSs are first retrieved using conventional PSI (Ferretti et al., 2001), and then the deformation parameters of DSs are obtained under the constrain of the result of the PSs. A series of Radarsat-2HH polarization images collected in Huainan are used to verify the effectiveness of the proposed method. A dense ground surface deformation map is obtained. Experimental results show the potential application of the proposed method to retrieve ground movement in coal mining areas.

Section snippets

Proposed method

The block diagram of the proposed method is shown in Fig. 1. After all the SAR images are co-registered, the DSs are selected first by combining classified information and statistical characteristics. For simplicity, PSs are identified based on their coherence stability in the stack of the interferograms. Then, a two-scale network is constructed at PSs and DSs. Finally, linear deformation and nonlinear deformation components are retrieved using conventional PSI. It should be noticed that in the

Study area and data processing

Huainan, located in the Yangtze River Delta hinterland central of Anhui province, is rich in coal resource, accounted for 19% of the national vision of stocks of coal stock. There are total 14 pairs of key state-owned coal mines in Huainan. These mines are mostly located in central Huainan. Coal mining in Huainan is of great importance for local economic development. However, this anthropogenic activity has caused series land subsidence, which result in many series of environmental problems and

Conclusions

This paper has presented an approach to retrieve ground deformation over nonurban areas using time-series InSAR combining PSs and DSs. DSs can be efficiently identified by the new strategy using classified information and statistical characteristics. A two-scale network on PSs and DSs has been constructed, which may make parameter estimation easier and more efficient.

The approach has been applied to monitor the subsidence in coal mining areas of Huainan City, China from September 2012 to

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China under Grants 41271425 and 41331176.

We would like to thank the anonymous reviewers for carefully reading the manuscript and providing valuable comments and suggestions.

References (24)

Y. Guéguen et al.
Monitoring residual mining subsidence of Nord/Pas-de-Calais coal basin from differential and Persistent Scatterer Interferometry (Northern France)
J. Appl. Geophys.
(2009)
H.C. Jung et al.
Satellite observation of coal mining subsidence by persistent scatterer analysis
Eng. Geol.
(2007)
A.H.-M. Ng et al.
Mapping accumulated mine subsidence using small stack of SAR differential interferograms in the Southern coalfield of New South Wales, Australia
Eng. Geol.
(2010)
Z. Perski et al.
InSAR analyses of terrain deformation near the Wieliczka Salt Mine, Poland
Eng. Geol.
(2009)
C. Prati et al.
Recent advances on surface ground deformation measurement by means of repeated space-borne SAR observations
J. Geodyn.
(2010)
Y. Wang et al.
Retrieval of phase history parameters from distributed scatterers in urban areas using very high resolution SAR data
ISPRS J. Photogram. Remote Sens.
(2012)
J. Baek et al.
Analysis of ground subsidence in coal mining area using SAR interferometry
Geosci. J.
(2008)
P. Berardino et al.
A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms
IEEE Trans. Geosci. Remote Sens.
(2002)
A. Demoulin et al.
GPS monitoring of vertical ground motion in northern Ardenne–Eifel: five campaigns (1999–2003) of the HARD project
Int. J. Earth Sci.
(2005)
A. Ferretti et al.
Permanent scatterers in SAR interferometry
IEEE Trans. Geosci. Remote Sens.
(2001)

A. Ferretti et al.

A new algorithm for pro-cessing interferometric data-stacks: Squeesar

IEEE Trans. Geosci. Remote Sens.

(2011)

L. Ge et al.

Mine subsidence monitoring using multi-source satellite SAR images

Photogramm. Eng. Remote Sens.

(2007)

Cited by (37)

Research on the spatiotemporal prediction of mining deformation with subcritical extraction integrated with D-InSAR technology
2023, Advances in Space Research
The prediction of surface movement and deformation in the mining area is of great significance to the production of the mining. Since the differential interferometry synthetic aperture radar measurement (D-InSAR) technology cannot predict the deformation caused by mining face, and the existing probability integral method (PIM) prediction model can only predict final mining deformation under critical extraction. Therefore, on the basis of PIM prediction model, the simplified Boltzmann function and Gompertz time function were introduced to form the BPIM dynamic prediction model for the analysis of temporal and spatial characteristics of the mining area. The parameters of BPIM model are inversed by D-InSAR monitoring data combined with optimization algorithm, and the mining of the working face is followed by the dynamic forward prediction of the changes in temporal and spatial characteristics. In addition, we analyzed the sensitivity of the main parameters in the model to the predicted results and the variation relationship of the model subsidence coefficient with the mining of the working face. When the model was applied to the working face of Guobei coal mine, the RMSE of the BPIM dynamic prediction model is 25.1 mm. Constructed with the previous dynamic prediction model established by conventional methods, its deformation accuracy under the prediction of subcritical extraction on the surface of the mining area is improved by 58.7%. The results demonstrated that the BPIM dynamic model effectively avoids the problem of large prediction results under subcritical extraction, and can provide some suggestions for the planning and production of the mining area.
Review of satellite radar interferometry for subsidence analysis
2022, Earth-Science Reviews
Citation Excerpt :
Analysis of linear infrastructures has been successfully carried out with the interferometric technique (Parcharidis et al., 2009; Raspini et al., 2013) mainly to identify differential displacements and ground fracture delineation, essential for urban development planning (Castellazzi et al., 2017). Despite well-known limitations, SAR interferometry has been used to give a two-dimensional spatial coverage over non-urban, natural areas (Aly et al., 2012; Da Lio et al., 2018), mainly thanks to applications of different bands (Wegmuller et al., 2004; Wempen and McCarter, 2017; Zhou et al., 2018), adoption of current SAR satellites which perform better than their predecessors (Cascini et al., 2013; Ng et al., 2017) and, finally through the combination of different processing algorithms (Poitevin et al., 2019) and synergy of persistent and distributed scatterers (Zhang et al., 2019a; Zhang et al., 2015b). Monitoring means the regular measure of the surface displacement field induced by the event and reconstruction of its evolution history.
This paper includes a critical review of the existing literature on the use of satellite SAR imagery for subsidence analysis. Land subsidence, related to multiple natural and human-induced processes, is observed globally in an increasing number of areas. Potentially leading to severe impacts on economics and the environment, subsidence has attracted growing scientific attention and, over the last decades, new tools and methods have been developed for accurately measuring the spatial and temporal evolution of surface deformations associated with subsidence phenomena. The collection of the existing scientific literature on the satellite InSAR for subsidence analysis was conducted in January 2022 exploiting the WoS's freely accessible web search engine. An extensive database of 1059 scientific contributions was compiled, covering the period 1997–2021. The content of each record in the literature database has been critically examined to collect and store information regarding the study area location, microwave band adopted, satellite used, processing approach, subsidence cause, application type, field evidence and strategies to validate and compare InSAR data.
Analysis of temporal distribution revealed a substantial growth in scientific production and an increasing interest of geoscientists, with a mean value of 21 articles per year from 1997 to 2014, rising to about 100 articles per year between 2015 and 2021. All continents include at least a study area, with Asia and Europe having the largest number of case studies, with 586 and 281 analyses in their territory, respectively, and revealing a clear geographical bias in subsidence study locations. Graphical visualizations and syntheses of current applications are presented. The large availability of different acquisition bands, the increasing imaging capabilities, refinement of processing approaches, and growing expertise in data interpretation allowed InSAR data to be used at different scales of analysis, for different purposes and subsidence types, in a wide range of physiographic settings.
This review highlights that satellite InSAR has moved from being a niche topic to an operative tool with a major role in subsidence studies. Despite more than 25 years of progress and advancements, technical and operational challenges remain to be faced. Leveraging on the analysis of the literature review and authors' experience, recommendations and perspectives are provided for a more effective use of InSAR data.
A phase-decomposition-based polarimetric coherence optimization method
2022, International Journal of Applied Earth Observation and Geoinformation
The interference between different scattering components in distributed scatterer (DS) makes it difficult to estimate the phase from DS. Recently, some methods based on eigenvalue decomposition (EVD) are proposed to separate scatterer components from DS, such as component extraction and selection SAR (CAESAR) and phase-decomposition-based persistent scatterer (PS) InSAR (PD-PSInSAR). Meanwhile, many polarimetric radar satellites launched in recent years provide abundant polarization data, which promotes the research of polarimetric coherence optimization (PCO). The capability of PCO to improve the amount and quality of PSs has been verified. In this paper, a phase-decomposition-based polarimetric coherence optimization (PD-PCO) is proposed to explore the potential of PCO to improve the phase quality of DS. Firstly, the contribution of the first principal component on a coherence matrix is obtained by searching for the optimum projection vector. Then, the new scattering coefficient is generated by combining the polarimetric scattering vector and the projection vector. Finally, the interferometric phase of dominant scattering mechanism component is separated from the phase component by EVD. Twenty-eight dual-pol Sentinel-1A SAR images have been adopted to validate the effectiveness of the proposed method. Phase derivative variation and coherence of the proposed method has been compared with other algorithms. The result shows that the quality of phase estimated by the PD-PCO is superior to the CAESAR, PD-PSInSAR, and the mean coherence polarimetric optimization based on equivalent scattering mechanism (ESM-R). The experimental results show that the proposed algorithm successfully improves the phase quality of DSCs.
Investigation of deformation patterns by DS-InSAR in a coal resource-exhausted region with spaceborne SAR imagery
2021, Journal of Asian Earth Sciences: X
Citation Excerpt :
But the DS-InSAR (distributed scatters-DS) technology could effectively overcome the shortcomings of the conventional time-series InSAR method applied to rural environments by integrating permanent scatterers (PS) and distributed targets (DS) (Cao et al., 2015; Ferretti et al., 2011). It has been proved to be able to effectively increase the number of observed points in oilfield, landslides and other study regions (Jiang et al., 2016; Zhang et al., 2015; Zhao et al., 2019). In addition, considering both the low coherence along with distributed targets and multiple scattering mechanisms within one resolution units of low- and medium-resolution SAR images, Fornaro et al. (2014) and Cao et al. (2016)successively optimized the phase value and obtained available results with DS-InSAR technology based on the eigenvalue decomposition of the covariance/coherence matrix.
Along with the impact of energy structure adjustment as well as the coal resource exhaustion in the old mining region, the deformation over the abandoned mine region has severely restricted both the reuse of abandoned lands and the sustainable urban development. Therefore, it is essential to conduct the long-term surface deformation observation in the abandoned coal region. In order to overcome the incoherence factors influencing performance of the traditional InSAR time series technique, the new InSAR time series approach integrating the distributed scatter targets (DS) was presented and applied in observation and analysis of surface deformation characteristics in coal resource exhaustion region. With the help of DS identification and corresponding phase optimization, the number of measurement points (MPs) increase from 92,539 to 781,364 in Peibei coal region with Sentinel-1 TOPS imagery, which make up for the deficiency caused by the uneven distribution of few points. Compared to the traditional TS-InSAR method, DS-InSAR results could efficiently improve the deformation identification ability with high reliability in coal regions. The monitoring result reveals that the surface became steady gradually along with the underground mining activities stop, but there is still some instability over the closed goaf because of the unstability of coal pillar and the improving surface loads. Thus the long-term surface monitoring should be carried out in old coal region, and the DS-InSAR would provide an efficient and reliable technical support for surface stability investigation in coal resource exhausted region.
Monitoring active open-pit mine stability in the Rhenish coalfields of Germany using a coherence-based SBAS method
2020, International Journal of Applied Earth Observation and Geoinformation
Citation Excerpt :
They are also used to monitor groundwater-induced subsidence (Haghshenas Haghighi and Motagh, 2019; Motagh et al., 2017, 2008, 2007; Tang et al., 2016), glacier movements (Sánchez-Gámez and Navarro, 2017), and nuclear tests (Wang et al., 2018). These advanced InSAR techniques have also been used for ground deformation monitoring in underground and open-pit mining areas (Carlà et al., 2017; Colombo and Macdonald, 2015; Mura et al., 2016; Przylucka et al., 2015;Yang et al., 2017a, 2017b;Zhang et al., 2015a, 2015b). However, the resolutions (10-30 m) and long revisit times (24-, 35-, and 46-day repeats) prevent this technique from becoming commonly applied in the mining industry (Esch et al., 2020, 2019), and only recently have a few studies evaluated the potential of data from the Sentinel-1A/B satellite constellation for early warning and aiding decision making in an open-pit mine (Intrieri et al., 2019).
With the recent progress in synthetic aperture radar (SAR) technology, especially the new generation of SAR satellites (Sentinel-1 and TerraSAR-X), our ability to assess slope stability in open-pit mines has significantly improved. The main objective of this work is to map ground displacement and slope instability over three open-pit mines, namely, Hambach, Garzweiler and Inden, in the Rhenish coalfields of Germany to provide long-term monitoring solutions for open-pit mining operations and their surroundings. Three SAR datasets, including Sentinel-1A data in ascending and descending orbits and TerraSAR-X data in a descending orbit, were processed by a modified small baseline subset (SBAS) algorithm, called coherence-based SBAS, to retrieve ground displacement related to the three open-pit mines and their surroundings. Despite the continuously changing topography over these active open-pit mines, the small perpendicular baselines of both Sentinel-1A and TerraSAR-X data were not affected by DEM errors and hence could yield accurate estimates of surface displacement. Significant land subsidence was observed over reclaimed areas, with rates exceeding 500 mm/yr, 380 mm/yr, and 310 mm/yr for the Hambach, Garzweiler and Inden mine, respectively. The compaction process of waste materials is the main contributor to land subsidence. Land uplift was found over the areas near the active working parts of the mines, which was probably due to excavation activities. Horizontal displacement retrieved from the combination of ascending and descending data was analysed, revealing an eastward movement with a maximum rate of ∼120 mm/yr on the western flank and a westward movement with a maximum rate of ∼ 60 mm/yr on the eastern flank of the pit. Former open-pit mines Fortuna-Garsdorf and Berghein in the eastern part of Rhenish coalfields, already reclaimed for agriculture, also show subsidence, at locations reaching 150 mm/yr. The interferometric results were compared, whenever possible, with groundwater information to analyse the possible reasons for ground deformation over the mines and their surroundings.
Research on ground deformation monitoring method in mining areas using the probability integral model fusion D-InSAR, sub-band InSAR and offset-tracking
2020, International Journal of Applied Earth Observation and Geoinformation
Citation Excerpt :
Ground subsidence damages buildings, soil, vegetation, and water resources; sudden ground subsidence may cause earthquakes. Real-time monitoring of ground movements in mining areas can provide early warning of the above problems, which can effectively reduce the occurrence of environmental geological disasters, or at least reduce their impact (Ge et al., 2007; Ng et al., 2009, 2010, 2017; Zhang et al., 2015). The Interferometric Synthetic Aperture Radar (InSAR) is a non-contact and green monitoring technology that has the advantage of being available all day and in any weather conditions.
With the aim of addressing the problem of accurately monitoring complete deformation fields over mining areas by means of Synthetic Aperture Radar (SAR), this paper proposes a solution to obtain complete deformation fields using the probability integral model to fuse deformation data derived from Differential Interferometric SAR (D-InSAR), sub-band InSAR and offset-tracking. This method is used for small-scale, medium-scale and large-scale deformation monitoring using D-InSAR, sub-band InSAR and offset-tracking, respectively. Finally, the probability integral model is utilized to integrate the three deformation fields, and a complete deformation field with high-accuracy over the study area can be obtained. The method is tested on 13 TerraSAR-X (TSX) images from December 2, 2012 to April 24, 2013 of the working face 52,304 of the Daliuta mining area in Shaanxi province, China. The complete deformation field of the working face during the 113-day mining period is obtained. The results show that during the process of working face advancing, the subsidence basin has been expanding along the direction of excavation. The relationship between the average maximum subsidence rate and the advancing distance of the working face can be described by a quadratic polynomial. It has been also observed that, when the underground mining reaches the full mining condition, the maximum subsidence value does not increase further. The accuracy of the proposed method is verified against the global positioning system field survey data. The root mean square errors in the strike and dip directions are 0.134 m and 0.105 m, respectively. Due to the support provide by the reserved coal pillars, the subsidence value above the reserved coal pillars is smaller.

View all citing articles on Scopus

View full text

Subsidence monitoring in coal area using time-series InSAR combining persistent scatterers and distributed scatterers

Highlights

Abstract

Introduction

Section snippets

Proposed method

Study area and data processing

Conclusions

Acknowledgments

J. Appl. Geophys.

Eng. Geol.

Eng. Geol.

Eng. Geol.

J. Geodyn.

ISPRS J. Photogram. Remote Sens.

Analysis of ground subsidence in coal mining area using SAR interferometry

Geosci. J.

A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms

IEEE Trans. Geosci. Remote Sens.

GPS monitoring of vertical ground motion in northern Ardenne–Eifel: five campaigns (1999–2003) of the HARD project

Int. J. Earth Sci.

Permanent scatterers in SAR interferometry

IEEE Trans. Geosci. Remote Sens.

A new algorithm for pro-cessing interferometric data-stacks: Squeesar

IEEE Trans. Geosci. Remote Sens.

Mine subsidence monitoring using multi-source satellite SAR images

Photogramm. Eng. Remote Sens.