Font Size: a A A

Land Subsidence Monitoring And Building Risk Assessment Based On InSAR And Machine Learning In Lanzhou,China

Posted on:2024-02-29Degree:MasterType:Thesis
Country:ChinaCandidate:Y M XuFull Text:PDF
GTID:2530307064468834Subject:Solid Earth Physics
Abstract/Summary:PDF Full Text Request
Lanzhou,as a representative city located in the Loess Plateau region of China,is affected by various factors such as geological conditions,precipitation,seismic subsidence and engineering construction,with various types of geological hazards frequently occurring,especially building cracking and tilting accidents caused by land subsidence have occurred in hundreds of cases,which are the main geological hazards threatening the safety of infrastructure and building structures in the main urban area of Lanzhou.However,the causes of ground subsidence are incredibly complex.Obtaining long time series of subsidence information can help to grasp the changes of various ground hazards and provide guidance for the prevention and control of land subsidence.Based on this,this paper uses the Interferometric Synthetic Aperture Radar(InSAR)technique to monitor the land subsidence in the main urban area of Lanzhou from 26 October 2014 to 12 December 2021,combining data on groundwater level,stratigraphic lithology,geological faults,building loads and rail traffic in Lanzhou to analyze in detail the influencing factors and modes of action of land subsidence in Lanzhou.Meanwhile,a machine learning method was used to predict the distribution of risk levels of future land subsidence in Lanzhou.At the same time,vectorized data of buildings in Lanzhou City were added to obtain the subsidence risk accurate to the unit of buildings.The main research results achieved in this study are as follows:(1)Subsidence pattern:Lanzhou’s annual average deformation rate ranges from-18.74 to 12.78 mm/yr.During the monitoring period,most of the subsidence areas in Lanzhou are dominated by linear subsidence.The subsidence areas are mainly located along the Yellow River,along the railway lines,and in the villages and towns at the edge of the urban areas.The main areas where subsidence occurred were the eastern part of Chengguan District,along the railway line in Anning District,the southern part of Xigu District,and Qilihe Urban Area,accounting for 38.761%,43.455%,32.497%and 51.765%of the respective administrative areas respectively.(2)Causes of subsidence:The decrease in groundwater level led to land subsidence.Lanzhou experienced excessive groundwater drainage in some areas between 2015 and 2017.The most painful areas experienced a 1m drop in groundwater level in the last two years,while the surface subsided by 14.61mm.There is a large amount of substantial subsidence in the regional distribution of loess in Lanzhou,with the remaining moderately compressible soils experiencing a small amount of significant subsidence.The ground rise and fall in Lanzhou is related to the movement of faults,along with opposite ground movements on either side of the Retan River Fault and the Jinchengguan Fault.Subsidence occurred in most of the railway sections in Lanzhou,reaching a maximum of-11.68mm/yr.Also,over half of the super-tall building areas showed subsidence funnels.(3)Subsidence prediction:The random forest model analysis shows that the factors affecting land subsidence in Lanzhou are,in order of importance,the rate of ground settlement(weight 0.498),cumulative ground settlement(weight 0.213),the effect of precipitation(weight 0.054),the distribution of faults(weight 0.062),the lithology of the ground(weight 0.035),high-rise buildings(weight 0.026),the distance to rivers(weight 0.036)and railway construction(weight 0.076).The area at very high risk of future subsidence in Lanzhou is 22.02km~2 and the area at high risk of subsidence is 54.47km~2.The areas at greatest risk of future subsidence are Chengguan District and Qilihe District.The city covers a total of 51163 buildings in the very high-risk area,with about 44.47%of brick and timber houses,51.36%of old housing and 52.78%of super-tall buildings having a very high risk of sub-sidence,threatening the lives and property of the population.The deformation results of the large area well reveal the danger to building safety in Lanzhou,providing an essential basis for future urban development and construction.
Keywords/Search Tags:Land subsidence, InSAR, Machine learning, Lanzhou, risk assessment
PDF Full Text Request
Related items