| The Karakoram Highway(KKH)connecting Pakistan and China is a key section of the China-Pakistan Economic Corridor,and it traverses the Kunlun Mountains,Pamir Plateau,Karakoram Mountains,and the Himalayas,which are regions with the most complex geo-environment,various landforms,and intense seismic activities on earth.The price paid for the KKH construction was huge.It’s described that along the KKH,one person was injured per kilometer,and one Chinese person gave his life every 3kilometers.Some sections of KKH have been relocated and reconstructed.The serious landslide disaster directly affects the safe operation of the China-Pakistan Highway and the safety of people in villages and towns along the highway.Landslide identification and inventory are the basis and preliminary steps toward landslide susceptibility,hazard,and risk assessment.Landslide susceptibility and hazard mapping are the key tasks of geohazard management,prevention,and risk mitigation.However,the technical problem of landslide identification along KKH is still a pending issue,the comprehensive potential landslide inventory is indeed needed.The landslide development characteristics and genetic mechanisms have not been revealed systematically.There is a lack of up-to-date studies on potential landslide-based susceptibility and seismic landslide hazard assessment along KKH.Therefore,the scientific problem of distribution of landslide susceptibility and hazards along KKH,its differences in spatial,and intrinsic causes,and genetic mechanism have not been clarified clearly.In view of these technical and scientific problems,it is urgent to carry out a comprehensive and systematic study to provide scientific support for landslide hazard prevention and safe operation of the China-Pakistan Highway and China-Pakistan Economic Corridor.Based on the data collection of the geological environment,and historical disaster in the study area,by applying multi-source Remote Sensing monitoring and identification,field verification and investigation,geo-statistics,and spatial analysis,the landslide identification,distribution characteristics,development regularity,and genetic mechanism along the KKH were systematically analyzed.A slope unit division program based on image pixel processing was developed.Studies on landslide susceptibility and seismic landslide hazard assessment were carried out along KKH based on slope units and machine learning modeling and the Newmark model.The main achievements and conclusions condensed in this study are as follows:(1)A total of 762 potential landslides(57 complex landslides,126 rock falls,167debris slides,and 412 unstable slopes)were identified.The accuracy of landslide identification is 66.74%.Potential landslides are mainly concentratedly distributed in the southern Khunjerab Valley,Hunza Valley in Pakistan,Tashkurgan River Valley,and Gaizi River Valley.The features of area differences were found that the landslide area along the Pakistan section is larger while the landslide area along the domestic section is small.This paper summarizes four levels of landslide distribution along the KKH:the three sections from Bulunkou to Tashkurgan,Hunza Valley(Khunjerab to Chalt),and Sazin to Besham are landslide-intensive areas;the Gaizi Valley,Jaglot to Chilas are landslide sub-intensive areas;the Tashkurgan River Valley(southern section),Chilas to Sazin are landslide relative intensive areas;and the Chalt to Jaglot and Besham to Thakot are landslide sparse areas.(2)The landslide’s development regularity and the genetic mechanism were investigated and revealed.Firstly,landslide development is closely correlated with Geomorphology factors.In the slope with higher relief(200 m≤H<550 m),degree of25-40°aspect range of east-southeastern-south-southwest,landslides are concentrated developed.Secondly,landslides are clustering distributed in the area with soft rock mass.Along the KKH in China,they are mainly located in the deposition of Quaternary.Along the KKH in Pakistan,Mudstone,Slate,and other soft rock are prone to landslide development.Thirdly,the landslides are mainly distributed in the pattern of clustering and stripping along the faults zone.The coupling of earthquakes with faults and terrain amplification effect promotes landslide development.Fourthly,the area with intensive human activities is coincidently the area with the intense development of landslides.Fifthly,the power exponent relationship and prediction formula of landslide slope height,maximum altitude,and landslide area parameters are revealed.(3)The multi-angle InSAR observations and optical Remote Sensing interpretation were applied for the first time to complete the fine monitoring and identification of landslides,in the Hunza Valley area with a dense population,high building density,and serious disasters,as well as the study on the characteristics and causes of landslide development.Firstly,the development and distribution of landslides in Hunza Valley is mainly controlled by lithology,elevation,relief,slope gradient,and aspect.The more proximity to roads,water systems,faults,and epicenters,the more prone to landslides.Secondly,the power law formula of landslide area and volume is established to predict the landslide volume identified in Hunza Valley.Thirdly,this study innovatively reveals the activity characteristics of high altitude--large area--large relief--large slope--high deformation rate of active landslides in Hunza Valley,and preliminarily summarizes the genetic mechanism of integrated control of active fault--altitude--seasonal frozen-soil freeze-thaw--temperature-erosion.Fourthly,landslides are mainly distributed in the northern terrain of the Hunza River and are closely related to erosion processes and human activities.(4)The slope unit is chosen as the evaluating unit for landslide susceptibility mapping(LSM)by Support Vector Classifier machine on the 10 km buffer area along the KKH and in Hunza Valley.Firstly,the evaluation results showed that the area of very high susceptibility and high susceptibility accounted for 12.93%of the total area of the study area,and was mainly distributed in Gaizi Valley,Khunjerab,Hunza Valley,and Sost,which was in good match with the dense distribution of identified landslides.These regions are characterized by steep topography,high mountains and deep valleys,and intense glacier influence,which makes landslide development more serious.There are 753 of 762 potential landslides(98.82%)in the high and very high susceptibility regions.Secondly,in Hunza Valley,the slope units with high susceptibility were mainly distributed on the north bank of the Hunza River,which was consistent with the distribution of identified landslides.The north bank of Hunza Valley has many villages,high population density,and dense buildings.These landslides events happened recently verified the high landslide susceptibility section along the KKH.As a supplement to landslide identification cataloging,the evaluation of landslide susceptibility has great significance for landslide disaster prevention and risk prediction.(5)This study completed the seismic landslide hazard assessment with an exceeding probability of 10%of 50 years.The result indicates that the slope units in the high,and very high hazard levels are 2931,with an area of 3945.72 km~2,accounting for21.95%of the total study area.The study finds that 252 of 762 landslides identified by this study are in the high and very high hazards level,accounting for 33.07%.There are several segments in the high and very high hazards level along KKH in Pakistan,such as Sost to Hunza Valley,part of the Gilgit and Jaglot region,and part of the Sazin and Dasu region.Along KKH in China,some slopes in the front of Pamir mountains in Tashkurgan,and most slope units of Gaizi Valley are in high and very high hazard levels.In these areas,fault and seismic points are densely distributed,which provides the internal power source for landslide development.Finally,based on the results of the assessment and the characteristics of landslide damage to the China-Pakistan Highway investigated during the study,the author proposed some suggestions for prevention and control measures.The latest research results of potential landslide identification,cataloging,and evaluation can be directly applied to the detailed investigation and evaluation of disaster risk and the planning and research of prevention and control measures along the KKH. |
PDF Full Text Request