| The Changhe River Basin in Jincheng city of Shanxi Province is located in the southeast edge of the Loess Plateau,which belongs to the typical loess hilly area.It is rich in coal resources.The mining areas are densely distributed in the region,and the urban and rural residential areas are concentrated.Agricultural production and the development and utilization of coal resources are highly complex.Coal mining has caused a certain degree of damage to local buildings,farmland,agricultural facilities and water resources,resulting in the decline of agricultural productivity and the deterioration of regional ecological environment.In recent years,although the relevant departments have taken some measures in the aspects of environmental control measures and ecological restoration,the environmental problems still exist in the mining area.Therefore,based on the time series multi-source remote sensing monitoring technology,it is necessary to scientifically and deeply analyze the influence degree and distribution characteristics of surface deformation caused by coal mining in the Changhe River Basin of the Loess Plateau,the temporal and spatial distribution characteristics of surface soil water caused by coal mining disturbance and its relationship with terrain and environmental factors,and analyze the reclaimed soil water content characteristics in the mining subsidence area on the field scale The reasonable mining,environmental protection,land consolidation and other ecological reconstruction measures in the coal mining area are the strong support.In this study,33 Sentinel-1 radar images and 24 Sentinel-2 optical images were collected from the Changhe River Basin.Meanwhile,through field investigation,collection and sampling,the data of coal resource development and utilization schemes in 12 mining areas,as well as the data of soil characteristics,land use and terrain characteristics in the region were obtained.Based on these data,the contents and results of this study are as follows:(1)By using gray level co-occurrence matrix method and Sentinel-1 of VV and VH dual polarization as data source,the time series analysis of backscattering characteristics and texture characteristics of four typical ground objects in the study area under different polarization modes was carried out.The results showed that the texture characteristics of the four typical features were different under different polarization modes.On the whole,the value of backscattering coefficient was buildings > forest and grassland> farmland > water body,and the value of backscattering coefficient was VV > VH under different polarization modes.The order of complexity of texture information of four typical features was forest and grassland > farmland > buildings >water.In the VV polarization mode,the change of the time series of synergy was relatively stable,and the synergy of VV polarization was less than that of VH polarization,which indicated that the texture of VV polarization image was not as uniform as that of VH polarization.The variance of VV polarization was greater than that of VH polarization,which indicated that the texture intensity of ground objects was more dispersed in VV polarization mode.Based on the analysis of the time series characteristics of typical features in the study area,the results showed that cultivated land accounts for 42.87%,forest and grassland 44.5%,buildings 11.83%,water 0.81%,the total classification accuracy is 87.4%,and kappa coefficient was 0.823.(2)The Permanent Scatterer Measurement Technology(PS-In SAR)of synthetic aperture radar interferometry was used to process 33 Sentinel-1 images,and the annual average subsidence rate of the surface in the study area from April 12,2017 to December 22,2019 was obtained.The average subsidence rate varied from-15.8mm/yr to 9.5mm/yr.The subsidence area was mainly in the northwest of the goaf,and the lo of the area is relatively small S displacement variable shows a significant downward trend.Through the comparative analysis of field survey points and PS points in the study area,the influence range of surface deformation obtained by PS-In SAR technology is consistent with the field survey,and the damage degree is consistent with the damage degree of surface and buildings.Among the 12 mining areas in the study area,the surface deformation of the Chengzhuang Coal Mine,Tian’an Runhong and Tiantai Kunda were more serious;the deformation from the Chengzhuang Coal Mine was the most serious,with a deformation value that ranged from-20.22 mm/yr to 16.65 mm/yr.The deformation of the remaining nine mining areas was small,and the annual average subsidence rate ranged from-8.59mm/yr to 5.82mm/yr.Under the influence of neotectonic movement,the surface of undisturbed area showed a rising trend,while the surface of disturbed area showed a decreasing trend under the influence of underground coal mining.The surface deformation values of the three mining areas with the most serious surface deformation were not only related to neotectonic movement and precipitation,but they were also affected by underground mining activities.Precipitation is also an important factor affecting regional surface deformation.The R2 between the deformation and precipitation for IMF1 and IMF2 both were 0.30 in the small and medium scale(3.1 and5.9 M).The influence increased in the large scale(IMF3,18.9 M),and the R2 was-0.45.In the undisturbed area,the influence of precipitation on the deformation was not obvious at small scale(IMF1,3.9 M),but gradually increased at medium scale(IMF2,10.3 M)and large scale(IMF3,24.2 M),with R2 of 0.33 and 0.46,respectively.This result showed that the precipitation had more significant effect on the surface deformation in the disturbed area in a short period of time compared with the undisturbed area.(3)24 periods of surface soil water content in the study area from January 2018 to December 2019 were retrieved by using radar signal change detection algorithm based on time series,and the accuracy was verified by using 12 field measured soil water content data.The R2 was 0.7246,and the fitting degree was good.The results showed that the time series radar signal change detection algorithm had good accuracy in estimating the surface soil water content in the study area.The surface soil water content in the study area had obvious seasonal variation characteristics,showing the characteristics of dry wet dry,which was obviously affected by precipitation and had hysteresis.From the perspective of spatial change,the impact of mining disturbance on surface soil moisture was not obvious.From the time series change,the drying process of soil in disturbed area was faster than that in undisturbed area.In the process of changing from dry to wet,the soil water in the disturbed area could be replenished quickly.(4)Based on the characteristics of topography,hydrology and soil in Changhe River Basin,EOF method was used to analyze the potential spatiotemporal variation of surface soil water.The spatial anomaly of soil moisture in the study area was determined by two spatial models,which had different contributions in different time,spatial EOF1 and spatial EOF1 mode explained 58.8% and 7.1% of the spatial variability of soil water,respectively.A total of 65.9% of the two EOFs in the data set were explained,which indicated that most of the observed spatial variability of soil water was stable in time.The ability of temporal EOF1 to explain the temporal variation of soil moisture was 52.0%,which was obviously related to the rainy season and dry season caused by the change of precipitation and potential evapotranspiration.The results showed that the EOF mode in the disturbed area was closer to that in the whole study area.Finally,the influencing factors of soil moisture in the disturbed area,undisturbed area and the whole study area were analyzed.It was found that the environmental factors affecting soil moisture content in the disturbed and undisturbed areas were different,and they were mainly concentrated in soil texture,bulk density,altitude and slope,which indicated that the soil texture of the two sub areas may be different at different altitudes,or may be different from that the original soil structure was changed by the remediation measures in the disturbed area.On the whole,the EOF mode of the whole study area was determined by the EOF mode of the disturbed area.(5)Through indoor simulation test,the parameters of GPR were adjusted and determined.Two typical disturbed fields(No.1 and No.2)and one typical undisturbed field(No.3)in the study area were selected as the research objects.The grid data of the selected fields were collected by GPR,and the point data are obtained by conventional methods for accuracy verification.Through radar image processing and interpretation,the different soil thickness of the field was obtained.By fitting the relationship between soil dielectric constant and radar signal data with field measured data,the prediction model between soil moisture measured data and amplitude data extracted from GPR image was established.Through the use of ground penetrating radar for non-destructive detection of the soil thickness of reclaimed farmland,combined with the laboratory test experience,after data processing and interpretation,it is found that the average thickness of the soil cover of No.1 field was less than that of No.2 field,and its evenness was greater than that of No.2 field.By studying the relationship between soil moisture content and amplitude,a logarithmic model suitable for soil environment in the study area was established.The results showed that GPR signal attribute could be used to measure soil moisture content,and early amplitude analysis(AEA)could be used to quantitatively estimate soil moisture content.Compared with TDR and drying method,AEA method could obtain relatively accurate spatial distribution of soil moisture in the study area,and had better advantages in reflecting the continuous change of surface water content.By compared the soil moisture of the three cultivated lands,it was found that the distribution of surface soil water content in No.1 and No.2fields was uneven compared with that in No.3 field,which indicated that in the field scale,the surface soil water distribution was uneven due to the random existence of underlying soil fissures after the surface damage caused by underground mining activities and the implementation of certain reclamation measures.The change of soil moisture in different locations was more affected by altitude. |
PDF Full Text Request