The Study On The Mapping And Hazard Assessment Of Shallow Landslides In Gully Loess Area

In the Loess Plateau,loess landslides are especially frequent and widespread geological disasters,causing serious casualties as well as huge property damage annually.Mapping inventories and assessing the stability of landslides are always an important topic for the study of loess disaster,which has significance on land planning,landslide forecast and treatment.Finding a low-cost and fast method to map different sizes of landslides will benefit on landslide prevention and control work for the loess region.And the three-dimensional deterministic model,which considers complex slope environment,could guide to get more actual results and alleviate effectively landslide hazard.This paper chose the Laolang Gully as study area,which has typical gully physiographic region and located in Lanzhou.Firstly,Google Earth satellite image that combined with field investigation was used to map the landslide inventory.To analyze the morphological characteristics of all landslide,we used a handheld laser range finder to accurately measure the landslide size.To calculate landslide volumes,a portable standard penetration instrument was used to measure the thickness of the landslide.Through establishing the fitting relationship between height of the main scarp and penetration results,the thickness of all landslides in the study area was evaluated and the result was used to calculate the volume.Then,different input parameters and resolution of digital elevation model data were used to forecast the stability of landslide by Scoops3 D.Finally,the confusion matrix and the success rate curve validated and evaluated the reliability of the results with different resolution DEM data.By mapping the landslide inventory based on satellite images,we found that satellite images can clearly identify the landslide with area changing from dozens of square meters to thousands of square meters in the loess area.The efficiency of mapping was affected by the quality of the satellite images.Landslides in the study area can be classed as rotational slides,translational slides and falls.The analysis revealed that the position,type and topography of landslides have an important effect on morphological characteristics,especially for landslides that are not affected by water erosion.Their length and width has obvious linear relationship.The results standard penetrometer and the high density electrical method showed that the standard penetration instrument can be used as the effective tool for measuring the thickness of the landslide.The relationship of penetration thickness and height of the main scarp showed a good linear correlation,and the results can be calculated detailed landslide volumes in the study area.The regression equation between landslide areas and volumes belongs to a geometric relationship,which can be used to estimate roughly the volumes and access landslide hazard.In landslide susceptibility assessment,the result of parameters sensitivity analysis revealed that both geotechnical-engineering parameters and DEM resolution have a significant impact on the safety factors.The prediction results show that the rainfall is also an important factor to affect the stability of landslide.Scoops3 D model is qualified to access shallow loess landslides,and point landslides may be more suitable for model adaptive test.The result of validation and evaluation by the confusion matrix and the success rate curve show that this model has a good adaptation to predict the stability of shallow loess landslides.Meanwhile,it could obtain reliable prediction accuracy by the high-resolution DEM data.