Index Selection And Rating For Debris Flow Hazard Assessment

Debris flow, which has the strong destructiveness, is a kind of more complex fluidwith ferocious momentum, fast speed, large flow, strong destructive power, short timeof occurrence and high recurrence rate. These sudden natural disasters tend to occur inthe slopes or valleys of the mountainous area. Under the comprehensive effect of allkinds of natural factors and human factors including geological structure, topography,geomorphology, meteorology, hydrology and human economic activities, etc, in ashort period of time there will be a lot of sediment and rock dumped outside themountains, causing a huge disaster to human production and life, hinderingengineering economic construction of the mountainous area.Thus, debris flow has already become a danger to mountainous towns, the energybase, water conservation and hydro-power engineering, road traffic, the ecologicalenvironment and the safety for people’s lives and property. In order to alleviate thedirect and indirect damages caused by debris flow and protect the safety for people’slives and property, studying on debris flow has become a critical problem to beaddressed.This dissertation is based on the debris flow as the research object which is locatedon the downstream of Jinsha river Wudongde hydropower station, analyzed theoverall development condition of debris flow combined with the local datainformation and field investigation. The dissertation builded the three-dimensionalmodel and interpret the three-dimensional characteristic geometrical parameters of the27debris flows, applied remote sensing images, DEM and GIS technology. Throughfield investigation and3S remote sensing, the dissertation obtained the initial indices selection of debris flow assessment according to regional characteristics of debrisflow in the study area and field investigation.Then extract and reduct the evaluation index selection respectively according torough sets and stepwise discriminant analysis. Apply rough sets to reduct as4groupsof evaluation index, avoiding the repetitive factors. Apply stepwise discriminantanalysis to extract1group of evaluation index. The factors we selected have sufficientinformation which can reflect the hazard degree of debris flow. Apply the5groups ofevaluation index extracted by two kinds of nonlinear method, and calculate thecombination weighting by analytic hierarchy process and factor analysis. Accordingto the classification criteria of all evaluation indexes, build a extenics model of debrisflow susceptibility assessment respectively, and divide rating for the hazard degree ofdebris flow in the study area.At the same time, this dissertation built the traditional Liuxilin model and evaluatedthe debris flow susceptibility in the study area in contrast with rough sets-extensiontheory model and stepwise discriminant analysis-extension theory model. Considerthe initial indices selection as a combination which did not extract the evaluationindices, then combine with extension theory to rate for debris flow hazard assessment.It turned out that two models are built: one is extension theory combined with agroup of the evaluation index extracted by rough sets including the main channeltortuosity, length proportion of the debris supplement section, average main channelgradient, material per square kilometer and contour roundness; another is stepwisediscriminant analysis–extension theory model. The two models of debris flow hazardassessment is in accordance with the geological model, and they can reflect the actualsituation of debris flow in the study area.