| The adverse engineering geological properties of saline soil(i.e.,salt swelling,collapsibility and corrosion)will cause destructive disasters to the natural environment and artificial structures.The disasters include slope instability,building rupture,uneven uplift or subsidence of roadbed,corrosion of underground metal pipes,etc.Therefore,when large-scale engineering construction and resource development are carried out in the regions affected by saline soil,especially large-scale and ribbon constructions such as railways,highways and highvoltage transmission lines crossing different geological units,there will be many potential geological disasters caused by the uneven deformation.Monitoring and interpreting the timeseries dynamic evolution process in the saline soil areas are very important for preventing and controlling the potential geological disasters.In addition,we can reasonably exploit the natural environment and guarantee the safety of large-scale engineering constructions.Time-series InSAR(Interferometric Synthetic Aperture Radar),as a new earth-observation technology,can extract the deformation with high spatiotemporal resolution and high precision over a wide area.As a result,time-series InSAR has been successfully applied to the temporal deformation monitoring under complicated environments,such as urban surface,landslide disaster,permafrost,and alpine glacier.With the launch of a large number of SAR satellites and the continuous accumulation of SAR images,we can monitor the ground deformation and reveal the evolution law of the saline soil area by time-series InSAR under complicated geological environments.However,the extraction accuracy and reliability of the saline-soil collapsibility/swelling are poor,the deformation model of the conventional time-series InSAR has poor applicability when the deformation factors are complex,the regional deformation evolution trend of saline soil under the influence of hydrothermal environment changes is unclear,the boundaries of different geological units which lead to the differences of surface to underground deformation are fuzzy,and the traditional time-series deformation analysis method of the small-baseline subset(SBAS)has some shortcomings,such as inefficient use of the SAR images and unsuitable calculation model in the calculation process.It presents a new challenge for the accurate detection of the ground deformation and the reasonable interpretation of the internal mechanism,especially for the effective prevention and control of potential geological disasters.To solve the above problems and challenges,this paper is oriented to detect the dynamic evolution law and prevent the potential geological disasters in the construction of the Belt and Road Initiative.Consequently,we selected two study areas as the research objects,which are the typical sulphate saline soil area traversed by the Tehran-Isfahan high-speed railway and the typical chlorine saline soil area traversed by the Qinghai-Tibet railway,respectively.The main research contents include the saline-soil deformation extraction based on an improved time-series InSAR method with optimized interferograms,time-series InSAR modeling and analysis of saline-soil deformation considering environmental factors,multi-factor coupling modeling and parameter inversion for the regional dynamic evolution of saline soil,division and verification of the hydrogeological units in the Qarhan Salt Lake region based on the hydrologic model.To improve the precision of the time-series deformation in the saline-soil areas with significant seasonal variation characteristics,we proposed an improved time-series InSAR method based on high-coherence interferograms,which are extracted by systematically analyzing the average coherences of the time-series interferograms.The seasonal variation of SAR scattering signals in the saline soil areas are obtained by the statistical analysis of the average coherence of 3,161 interferograms from October 2014 to July 2020 along the Soltan Salt Lake section of the high-speed railway in Iran,and 2,921 interferograms from May 2015 to May 2020 along the Qarhan Salt Lake section of the Qinghai-Tibet Railway.In addition,the high-coherence interferograms(i.e.,the optimized interferograms)were extracted to participate in the time-series InSAR calculation process,and the high-precision time-series deformation in the satellite Line of Sight(LOS)was obtained.Compared with the results of the traditional SBAS-InSAR method,the improved time-series InSAR method greatly increased the number of the effective coherent targets,enhanced the stability of interferogram network,and improved the accuracy and reliability of the calculation results.To solve the inapplicability of the LOS deformation obtained from single-platform SAR images in the fine modeling of the time-series deformation and high-precision inversion of relevant physical parameters,we proposed a method to extract the long time-series twodimensional deformation based on the ascending/descending SAR images and cubic spline interpolation.Before the two-dimensional deformation calculation,we have optimized the LOS deformation of single-platform SAR images.The high-precision two-dimensional time-series deformation with a uniform time span baseline is obtained by calculating the Sentinel-1A SAR images from 2015 to 2020 in the Qarhan Salt Lake region.As a consequence,the validity of the two-dimensional deformation calculation method based on cubic spline interpolation is verified.To solve the problem that the linear deformation model commonly used in the time-series InSAR analysis cannot be applied to the nonlinear time-series deformation extraction in the saline-soil areas,we proposed a deformation model combining trigonometric functions and environmental factors in the time-series InSAR processing.The long time-series deformation of the sulphate saline soil was extracted in the Soltan Salt Lake area by our proposed method.According to statistics,the maximum root mean square error and minimum root mean square error of the time-series deformation calculated by the proposed model are 8.6 mm and 3.3 mm,respectively.Compared with the cubic polynomial model and the trigonometric function model,the accuracy of our model is improved to some extent,and the linear deformation component,the seasonal deformation variation intensity and the deformation components affected by the precipitation and the temperature are successfully separated out.Consequently,we can better reveal the dynamic evolution law and internal mechanism of the saline soil areas.To reduce the difficulty for detecting the surface to deep dynamic evolution law in the saline soil areas under the influence of hydrothermal environment changes,in this paper,a timeseries InSAR deformation model and inversion method for related physical parameters are proposed by considering the influence of the hydraulic head changes and the regional surface deformation.The time-series deformation in the Qarhan Salt Lake area is selected as a typical research object to construct a coupling deformation model combining environmental parameters(temperature and precipitation)and hydrogeological parameters(magnitude coefficient and hydrogeological decay coefficient of the hydrodynamic function).The deformation components of the surface saline soil and the deformation caused by the hydraulic head changes in the underground confined aquifer are separated successfully.Through modeling and inversion of the hydrogeological decay coefficient,the location of existing faults and newly discovered faults are accurately detected.In addition,the hydrogeological units are partitioned and provide data support for the prevention and control of the potential geological disasters across different hydrogeological units.To make the fuzzy relationship more clear between the hydrogeological units derived from the decay coefficient and the deep underground structure,the relationship between the boundaries of hydrogeological units and the cumulative thickness of the aquitard is established by analyzing stratigraphic data from drilling wells.We collected 50 drilling wells data in the Qarhan Salt Lake area,and the spatial distribution of the cumulative aquitard thickness in the stratigraphic structure is simulated by the ordinary Kriging spatial interpolation method.It is found that the isoline with the cumulative thickness of 2.8 m in the aquitard is in good agreement with the boundary of the decay coefficient map with an abrupt change from 0 to-1.By selecting two sub-regions and calculating the Jaccard similarity coefficients between the hydrogeological decay coefficient map and the spatial distribution map of the aquitard,the quantitative relationship between the decay coefficient and the cumulative thickness of the aquitard is explored(The Jaccard similarity coefficient of zone A is 0.866,and the Jaccard similarity coefficient of zone B is 0.879).At the same time,the reliability of the hydrogeological decay coefficient is verified in dividing different hydrogeological units and determining the fault location.Furthermore,the important criterion of hydrogeological unit division is expanded by using the aquitard thickness.A series of the research results presented in this paper provide innovative methods and theoretical support for revealing the regional dynamic evolution law and internal mechanism in the inland saline soil areas.Furthermore,the innovative research results can be used for preventing the potential geological disasters and constructing large-scale engineering projects. |
