CN103399344A - Prediction method for predicating collapse disaster position after earthquake - Google Patents

Abstract

The invention provides a predication method for predicating a collapse disaster position after earthquake. The predication method comprises the steps that multiple basic geographic parameters, multiple geological parameters, earthquake parameters, the geographic position of at least one known collapse site and the geographic positions of multiple non-collapse sites of the space to be predicted are obtained; Newmark displacement Dn is computed on the basis of the Newmark displacement model; a binary Logistic regression equation is constructed according to the distance between the predicted position and a river, the distance between the predicted position and a fracture zone, the topographic relief and the Newmark displacement Dn; by taking the parameters of the known collapse site and the non-collapse sites and the computed Newmark displacement Dn as the known conditions, the partial regression coefficient of each independent variable factor in the regression equation is computed; and the computed partial regression coefficients are utilized to construct the collapse disaster prediction model of the space to be predicated.

Description

The Forecasting Methodology of slump disaster position occurs after a kind of earthquake

Technical field

The present invention relates to the risk assessment technology field of disaster reduction and prevention after earthquake and disaster, more specifically, the present invention relates to after a kind of earthquake occur the Forecasting Methodology of slump disaster position.

Background technology

Earthquake is as a kind of destructive disaster, and the mankind's safety of life and property has all been caused to great threat.Violent earthquake, when above ground structure is damaged, also usually brings out secondary disaster.Particularly in mountain area, earthquake usually triggers a large amount of Secondary Geological Hazards, the disasters such as these avalanches, landslide, rubble flow increased the weight of earthquake loss, its loss that causes has even surpassed the direct loss that earthquake itself causes sometimes.Avalanche after shake, landslide disaster are class geologic hazards that continues and trigger at first after the earthquake, are also the basis, the thing source of having established of rubble flow, checked-up lake disaster simultaneously.Avalanche after shake, landslide disaster can be both that the generation with earthquake produces simultaneously, can be also after certain hour, just to occur after earthquake.

China is positioned at the intersection of circum-Pacific seismic zone and Mediterranean-Himalaya seismic zone, is one of the world multiple state of earthquake.Simultaneously, China is also the country on mountain more than, and mountain region, hills area account for 70% of area.The stack on many shakes and many mountains, make the Secondary Geological Hazards such as avalanche, landslide of China's earthquake-induced take place frequently, and lose huge, harm is serious.From between 1500-1949, the earthquake that has precise record to produce secondary avalanche, landslide disaster has 134 times.China is one of country of avalanche in the world, landslide, rubble flow especially severe, and the annual direct economic loss that causes because of geologic hazard accounts for more than 20% of disaster total losses.The fact shows, the violent earthquake that occurs in the Southwestern China mountain area tends to cause large-scale avalanche or landslide, causes serious economic loss and casualties.On May 12nd, 2008, the Wenchuan violent earthquake has caused the Secondary Geological Hazards such as ten hundreds of avalanches, landslide, caused painful loss for disaster area people's life and property safety, and within the quite a long time, people's lives and properties and communal facility safety are still existed to grave danger.According to estimates, the geologic hazard point that Wenchuan earthquake triggers has 3～40,000 places, take avalanche, landslide, Rolling Stone as main.Yin Yue equality people's investigation discovery, Wenchuan earthquake has brought out nearly 15000 places avalanche, landslide, mud-stone flow disaster, has caused approximately 20000 people's death.Sometimes, the Loss of Geological Hazards that earthquake triggers is also larger than the loss that earthquake itself causes, and for example, on September 7th, 2012, the victim's great majority in the earthquake of 5.7 grades of the Yiliangs of the Shao Tongshi of Yunnan Province are to die from the Rolling Stone disaster that earthquake triggers.The Lushan earthquake that on April 20th, 2013 occurred in the Ya'an, Sichuan Province has also triggered a large amount of avalanches, landslide disaster, stops up and cut off traffic, has had a strong impact on transporting of rescue personnel and goods and materials.Visible, the slump disaster after earthquake is the Secondary Geological Hazards that Southwest Mountainous Areas can not be ignored.

Slip and fall usually is accompanied, they have the contact that can't cut apart, result from identical geological tectonic environment and identical formation lithology structural environment under, and identical triggering factors is arranged, the area that easily produces landslide is also the Yi Faqu of avalanche, so landslide type avalanche or avalanche type landslide are also referred to as slump stream.Avalanche, landslide can bring out mutually under certain condition, transform mutually.The avalanche that in this paper context, earthquake causes and landslide also are called for short the slump disaster.

The distribution characteristics of understanding the rear slump disaster of shake is all significant to post-disaster reconstruction and risk assessment.At present, research for slump disaster distribution characteristics after shake has a lot, main method is by field investigation and remote sensing image decipher, obtain the rear landslide distribution information of shake, and based on the landslide distribution situation under GIS research varying environment background, statistical study is come down under the conditions such as different intensity area, different elevation, different gradient, different lithology distribution situation, explore its regularity of distribution.After shake, the research of slump disaster distribution characteristics is that further exploration discovery earthquake-landslide influence factor has been established important foundation.Existing viewpoint thinks that the range of influence of Earthquake-landslide is the area of the outer boundary institute enclosing region of all landslide points, and thinks that the earthquake motion intensity that stand in these zones is enough to induced landslide and avalanche.The research discovery, the influence factor on earthquake-landslide mainly contains: seismologic parameter, tectonic structure, Rock Nature, topography and geomorphology, hydrogeology, mankind's activity etc.Based on these influence factors, build the model of probability of happening of slump disaster after the prediction earthquake, effectively after the definite area shake, the slump disaster is dangerous distributes, for commander takes precautions against natural calamities, disaster relief work provides decision-making foundation, to the directive significance that restoration and reconstruction after calamity, planning and economic development have, also for further disaster risk investigation, lay a good foundation simultaneously.

The development and application of high precision remote sensing technology and GIS technology, for distribution characteristics and the Study on influencing factors of shaking rear slump disaster provides favourable technical support, for understanding earthquake-mechanism of the occurrence of landslide, further carry out risk assessment and have vital role.

From existing research method, the danger of slump disaster can be divided into two kinds: the one, and statistical analysis method, the 2nd, slope stability analysis method.

Statistical analysis method is based on the principle of mathematical statistics and method and the forecast model set up, general process is the controlling factor that at first the impact landslide is occurred, as employing mathematical statistics methods such as geology, inclination angle, terrain feature, the generation on landslide and the spatial relationship of influence factor are analyzed, and attempt finding out statistical law, the danger that finally based on corresponding rule, following landslide disaster the is occurred judgement of making prediction.Adopt mathematical statistics method to analyze the generation of shaking rear landslide, avalanche and the relation of factor of influence, find out statistical law, and in the process of predict future danger the most frequently used to method be the Logistic regretional analysis.The Logistic regretional analysis is a kind of probabilistic type nonlinear regression model (NLRM), be used to a kind of multivariable technique of the Relations Among of studying classification observations and some influence factors.For the classification observations, only have the Logistic of two classified variables of two states to return just being called is that binary Logistic returns.In binary Logistic returned, the dependent variable in regression equation was in fact probability, rather than variable itself.

Binary Logistic regression equation is expressed as follows:

$P=\frac{\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_j{x}_j)}{1+\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_j{x}_j)}$ Formula I

In formula, P is dependent variable, is the probability of happening of the independent variable factor with respect to a certain event, and span is [0,1]; x _jBe the independent variable factor, j is positive integer, and being affects the factor that event occurs; β ₀... β _jPartial regression coefficient, reflection independent variable factor x _jCapability of influence size to P.In the binary Logistic regression equation that uses in slump disaster liability, hazard assessment, dependent variable is 1,0 variable, expression " slump ", " not slump " implication, and independent variable is the influence factor that slump occurs.

Zhao Bin is in the article of " based on the Wenchuan earthquake Study of Hazard Evaluation of GIS " at exercise question, see [D]. Capital Normal University, 2011, take Wenchuan County as research object, based on GIS and Logistic regression model, set up geologic hazard sensitivity assessment model, and attempted having carried out dangerous zoning based on sensitivity analysis.Zhou Wei is in the article of " based on the Bailong River Basin Landslide Hazard Assessment research of Logistic recurrence and SINMAP model " at exercise question, see [D]. Lanzhou University, 2012, on the basis of having analyzed the landslide disaster factor of influence, 16 landslide contribution factors have been chosen, for example comprise elevation, NDVI, the gradient, lithology, the river distance, 60 minutes average rainfalls and soil utilization etc., the minor effect factor is respectively 24 hourly average rainfall amounts and slope aspect etc., under the support of GIS, Logistic regression model and SINMAP (Stability Index Mapping) model are applied in the Landslide Hazard Assessment of Bailong River Basin, research finds that Logistic regression model precision is 70.24%, Regional Landslide Hazard Risk Assessment effect is better than to the SINMAP model.Utilizing the danger of statistical analysis method Study of Seismic landslide, collapse hazard is more general method in present earthquake slump Hazard Risk Assessment, effect is good, but this method is based on the statistical law of mathematics mostly, has ignored the kinetic mechanism of earthquake slump disaster.

The slope stability analysis method is based on regional geotechnical property and mechanical analysis, adopts traditional slope stability computation model to predict liability, the danger of Regional Landslide.In the slump Disaster Study, the slope stability method can be analyzed the slump disaster on driven mechanical mechanism after shake.The quantitative test of the physics origin cause of formation has been carried out in the reservoir dam failure danger that the people such as nineteen sixty-five Newmark may cause for earthquake, proposes a kind of short-cut method of predicting under geological process the landslide displacement, and the stability of side slope is judged by critical acceleration.The Newmark model later stage constantly is improved, and is used widely in the Earthquake-landslide risk assessment.For example, after the people such as Jibson had studied 1994 Nian Bei ridge earthquakes, the Newmark value of moving distributed, and built the Earthquake-landslide prediction curve in conjunction with actual landslide cataloguing, see Jibson R.W., Harp E.L., Michael J.A. " A method for producing digital probabilistic seismic landslide hazard maps ", [J] .Engineering Geology, 2000,58 (3 – 4): 271-289.The people such as further Jibson have studied Newmark displacement Dn and (1) Critical Seismic acceleration; (2) critical acceleration rate and magnitude of earthquake; (3) Arias intensity level and critical acceleration; (4) relation between four aspects of Arias intensity level and critical acceleration rate, see Jibson R.W. " Regression models for estimating coseismic landslide displacement ", [J] .Engineering Geology, 2007,91 (2 – 4): 209-218.The research in past shows, Newmark displacement computation model is the prediction earthquake-comparatively effective method that comes down, but its scope of application has certain limitation, mainly is applicable on the one hand the shallow failure forecast analysis of rock mass; On the other hand, because model is higher to the accuracy requirement of correlation parameter, therefore be difficult to be applied to the forecast analysis in large zone.In addition, Newmark displacement computation model is based on desirable slope and builds, and is also comprehensive not to the consideration of the earthquake slump disaster influence factor of reality.

In sum, statistical analysis method and Newmark displacement computation model cut both ways in slump disaster risk analysis after shake.In present research, the utilization of statistical analysis method and Newmark model is independently basically, and prior art only adopts a kind of method wherein to carry out computational analysis.

Therefore, need to build a kind of not only considered landslide make a difference the factor space distribution but also considered the shake of Slope Failure mechanics model after slump disaster hazard prediction method, to the danger of slump disaster after correctly prediction shake.

Summary of the invention

In order to solve the problems of the technologies described above, a kind of not only considered landslide make a difference the factor space distribution but also considered the earthquake of Slope Failure mechanics model after the Forecasting Methodology of slump disaster occurs.

According to an aspect of the present invention, provide the Forecasting Methodology that slump disaster position occurs after a kind of earthquake, the method comprises the following steps:

Obtain a plurality of fundamental geological parameters in space to be predicted, a plurality of geologic parameters, seismologic parameter, the geographic position of the geographic position of at least one known slump point and a plurality of not slump points;

Based on the Newmark displacement model, calculate Newmark displacement D _n

With at least a portion in a plurality of fundamental geological correlation parameters and Newmark displacement D _nAs the independent variable factor, build the binary Logistic regression equation of slump hazard prediction model;

By the described geographical correlation parameter of described known slump point and ,Ge position, described not slump point geographic position and the Newmark displacement D that calculates _nAs the known conditions of described binary Logistic regression equation, calculate in described binary Logistic regression equation the partial regression coefficient of Variable Factors separately;

The partial regression coefficient that utilization calculates builds the slump hazard prediction model in this space to be predicted.

Preferably, described fundamental geological parameter comprises one or more in the gradient, topographic relief degree, zone of fracture position, position, river; Described geologic parameter comprises one or more in rock mass completeness, rock-mass quality, rock mass physical mathematic(al) parameter and rock group intensity; Described seismologic parameter comprises earthquake magnitude, one or more in source depth and epicenter coordinate.

Preferably, the binary Logistic regression equation of described forecast model is as follows:

$P=\frac{\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}{1+\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}$

Wherein, P is the probability of happening of the rear slump of arbitrary position shake in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

Preferably, described non-slump point utilizes create random points instrument random generation the outside actual slump point in ArcGIS.

Preferably, described non-slump point generates at random in the scope outside actual slump point 200m.

Preferably, according to following formula, calculate Newmark displacement D _n,

log?D _n＝1.521log?I _a-1.993log?A _c-1.546±0.375

Wherein, I _aFor Arias intensity, m/s; A _cFor critical acceleration.

Preferably, according to following formula, calculate the Arias intensity I _a,

$\log I_a=M-2\log \sqrt{r^2{+7.5}^2}-3.99\±0.5$

Wherein, M is moment magnitude, and r is hypocentral distance.

Preferably, using the center line in maximum earthquake intensity zone as linear focus, calculate hypocentral distance.

Preferably, the method is applicable to the slump hazard prediction of sliding mass thickness less than 6m.

According to a further aspect in the invention, provide the Forecasting Methodology that slump disaster position occurs after the earthquake of a kind of Southwest China, comprising:

Obtain a plurality of fundamental geological parameters in space to be predicted, a plurality of geologic parameters, seismologic parameter, the geographic position of the geographic position of at least one known slump point and a plurality of not slump points;

Based on the Newmark displacement model, calculate Newmark displacement D _n

Utilize the probability of generation slump disaster in position to be predicted in following formula calculating prediction space,

$P=\frac{\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}{1+\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}$

In formula, P is the probability of happening of the rear slump of arbitrary position shake in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n), wherein,

β ₀=2.438～1.297；

β ₁=-0.798～-0.524；

β ₂=-0.431～-0.436；

β ₃=1.272～1.134；

β ₄=0.318～0.314。

Under the support of GIS technology and SPSS statistical analysis software, after the Related Environmental Factors of avalanche after extracting the impact shake, landslide disaster generation and the spatial relationship of Related Environmental Factors and slump disaster, can be tentatively definite, the gradient, topographic relief degree, rock and soil properties, weathering Erosion degree, magnitude of earthquake etc. are all the key factors that avalanche after the impact shake, landslide disaster occur.Simultaneously, can observe river and the zone of fracture very strong control action that is distributed with for the slump disaster, mainly due to the existence to river and zone of fracture, to have strengthened erosion and the weathering of both sides, slope ground, make the Rock And Soil anti-shear ability descend, thereby cause this areal geology environmental abnormality fragility, avalanche, landslide disaster very easily occur.

The method according to this invention, on the basis to slump disaster Analysis on Main Influence Factors after shaking, utilize the Newmark displacement calculation result of Newmark displacement model, utilizes binary Logistic regretional analysis to carry out model foundation and probabilistic forecasting.By choosing the Dn value, apart from river distance, apart from zone of fracture distance and four independent variable factors of topographic relief degree, set up probability model, calculate the partial regression coefficient of each independent variable factor and build Probabilistic Prediction Model, for the forecast analysis of earthquake rear region slump disaster probability of happening.

The present invention occurs in respectively the M of Sichuan Province China province Wenchuan County with on May 12nd, 2008 _s8.0 Wenchuan County, Beichuan County and Mianzhu County after the level special violent earthquake are example, have set up the probability model of the rear slump disaster generation of prediction shake that preferably is applicable to Southwestern China section zone.By with the comparing and ROC check of actual slump point, the probability model prediction accuracy that the present invention sets up reaches more than 75%, prediction effect is good.

The accompanying drawing explanation

Fig. 1 illustrates the process flow diagram according to Forecasting Methodology of the present invention;

Fig. 2 illustrates the seismic intensity distribution plan of various embodiments of the present invention relevant range;

Fig. 3 illustrates the engineering rock component cloth of embodiment 1 relevant range;

Fig. 4 illustrates the terrain slope distribution plan of embodiment 1 relevant range;

The side slope static security coefficient that Fig. 5 illustrates embodiment 1 relevant range distributes;

The side slope critical acceleration that Fig. 6 illustrates embodiment 1 relevant range distributes;

Fig. 7 illustrates the Arias intensity distributions of embodiment 1 relevant range;

Fig. 8 illustrates the Newmark accumulation displacement D of embodiment 1 relevant range _nDistribute;

Fig. 9 illustrates the ROC curve of the forecast model that obtains according to embodiment 1;

Figure 10 illustrates according to the disaster probability of happening of the relevant range of embodiment 1 and distributes and the distribution of slump point;

Embodiment 1 relevant range Dn is shown Figure 11 and slump point position distributes;

Embodiment 2 relevant range Dn are shown Figure 12 and slump point position distributes;

Figure 13 illustrates the ROC curve of embodiment 2;

Figure 14 illustrate embodiment 3 the ROC curve;

The disaster probability of happening that Figure 15 illustrates the relevant range of embodiment 3 distributes and the distribution of slump point

Figure 16 illustrates the ROC curve of embodiment 4;

Relevant range Dn and slump point position that Figure 17 illustrates embodiment 5 distribute;

Figure 18 illustrates the ROC curve of the forecast model of embodiment 5;

The disaster probability of happening that Figure 19 illustrates the relevant range of embodiment 5 distributes and the distribution of slump point.

Embodiment

Come a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings and in conjunction with embodiment.

According to technical scheme of the present invention, by Newmark displacement and river, zone of fracture, topographic relief degree are combined, utilize binary Logistic regression model to set up probability function, carry out the probability of happening prediction of the rear slump disaster of zone shake.

Below with reference to Fig. 1, describe in detail according to the Forecasting Methodology of slump disaster position occurs after earthquake of the present invention.

Step S110, after the earthquake, obtain a plurality of fundamental geological parameters in the space to be predicted that the slump disaster may occur, a plurality of geologic parameters, seismologic parameter, the geographic position of the geographic position of at least one known slump point and a plurality of not slump points.

As known to those skilled in the art, the factor that after the impact shake, the slump disaster occurs has geographic factor, as elevation, the gradient, topographic relief degree, zone of fracture position and position, river etc.; Geologic parameter, as formation lithology, comprise rock mass completeness, rock-mass quality, rock mass physical mathematic(al) parameter etc.; And various seismologic parameters, as earthquake magnitude, source depth and epicenter coordinate etc.Can utilize the data in the atlas such as the emergent investigation of earthquake-stricken area spacer remote sensing of the establishment such as land Resources Department, China Geological Survey Bureau to show, obtain a slump point latitude and longitude coordinates.Can utilize ArcGIS a plurality of non-slump points of random generation outside for example 200m of actual slump point, determine non-slump point position latitude and longitude coordinates.Remote sensing images analysis according to relevant area, determine the fundamental geological correlation parameter of each slump point and non-slump point and fundamental geological parameter correlation, such as the elevation of each position, the gradient, topographic relief degree, formation lithology, apart from the zone of fracture distance with apart from the river distance etc.

It is mainly magnitude of earthquake, epicentral location and source depth that Newmark simplifies the seismologic parameter that relates in the displacement computation model.Can obtain seismologic parameter according to basic seismologic parameter and the earthquake intensity figure of the earthquake of State Seismological Bureau issue.

Step 120, calculate Newmark displacement D based on the Newmark displacement model _n.

At Newmark, simplify the factor of considering in displacement the gradient, formation lithology and seismic intensity are arranged.The displacement D of Newmark model _nCan be regarded as the influence factor that the rear slump disaster of shake occurs, it combines the impact of terrain slope, formation lithology and the Earthquake Intensity factor, and the possibility that the slump disaster occurs is with D _nIncrease and increase.

The theoretical foundation that model construction is calculated in the Newmark displacement is the slope limit equilibrium theory, is applicable to the shallow failure research of earthquake-induced.This model is analyzed slope stability based on areal geology, geomorphological environment, thereby obtain the critical acceleration that area slope is subjected to displacement, when earthquake occurs, after the acceleration of the stressed generation of area slope surpasses critical acceleration, slope will lose stable gradually, along the destruction face, slide, produce permanent displacement, the permanent displacement value has characterized the possibility size that after the shake, landslide occurs to a certain extent.The permanent displacement value here is by the difference to earthquake external force acceleration and critical acceleration, partly to carry out quadratic integral to obtain, and its formula can be expressed as:

$D_n={\∫}_t{\∫}_t[a\left(t\right)-a_c]\mathrm{dt}$ Formula I-1

In formula I-1, Newmark permanent displacement amount D _nFactor of determination be Earthquake Intensity and critical acceleration a _c.For critical acceleration a _cCalculating, normally utilize infinite slope method computationally secure coefficient (F _s), more indirectly solve critical acceleration a _c, computation process is as follows:

Calculate slope static security coefficient F _s:

Formula I-2

C'---effective cohesion, z---destroy the face degree of depth, m;

γ---Rock And Soil severe, N/m ³z _w---the groundwater level depth that the destruction face is above,

M---z _w/ z, dimensionless; β---side slope surface inclination angle, (°);

The severe of γ---water, N/m ³

---effective angle of inner friction, (°);

Calculate critical acceleration a _c:

a _c=(F _s-1) g sin β formula I-3

Wherein, F _sFor the static security coefficient, g is acceleration of gravity, and β is the side slope surface inclination angle.

Earthquake Intensity in formula I-1 utilizes earthquake motion peak accelerator (PGA) and peak velocity (PGV) to describe usually, and the two all depends on the pulse in short-term of STRONG MOTION DATA medium-high frequency.But there are relation the destructive power of earthquake and vibration frequency, amplitude, duration, only use the destruction result that vibration frequency can not complete reflection earthquake.

1970, American scientist Arias proposed a comprehensive amount of weighing Earthquake Intensity, and namely Arias intensity (Arias Intensity), comprised vibration frequency, amplitude and duration full detail, more comprehensively reflected the overall condition of earthquake motion.Arias intensity by earthquake ground motion acceleration in STRONG MOTION DATA square when macroseism is held in to time integral, then multiplication by constants is determined, sees formula I-4:

$I_a=\frac{\mathrm{\π}}{2g}\underset{0}{\overset{T_d}{\∫}}{\left[a\left(t\right)\right]}^2\mathrm{dt}$ Formula I-4

Wherein, I _aFor Arias intensity, unit is m/s; A (t) is strong-motion instrument record component acceleration time series; T _dIt is the duration of strong-motion instrument record; T is the time take second as unit; G is acceleration of gravity.

The people such as Wilson and Keefer introduces Arias intensity in the research on landslide of earthquake triggering at first, utilizes repeatedly STRONG MOTION DATA to obtain the attenuation relation of Arias intensity earthquake magnitude, distance.The people such as Wilson utilizes the available records data afterwards, by numerical analysis method, has improved the zone decay experimental formula of Arias intensity, obtains formula I-5.

$\log I_a=M-2\log \sqrt{r^2{+7.5}^2}-3.99\±0.5$ Formula I-5

Wherein, I _aFor Arias intensity, M is moment magnitude, and r is hypocentral distance.

Because Arias intensity can complete description Earthquake Intensity, therefore be introduced in Newmark displacement computation model, in conjunction with critical acceleration a _c, set up corresponding functional equation prediction Newmark displacement D _nThereby, for the risk assessment on regional earthquake landslide.

Jibson, R.W. waiting the people is being " A Method for Producing Digital Probabilistic Seismic Landslide Hazard Maps:An Example from the Los Angeles at exercise question respectively, California, Area ", Technical report[R] .US Geological Survey Open-File report, 1997:98-113, with exercise question, be " Evaluating Earthquake-Triggered Landslide Hazard at the Basin Scale Through Gis in the Upper Sele River Valley ", [J] .Surveys in Geophysics, 2002, the following logarithm regression equation I-6 that sets up in the article of 23:595-625 is most widely used:

Log D _n=1.521log I _a-1.993log A _c-1.546 ± 0.375 formula I-6

Wherein, D _nFor Newmark shift value, cm; I _aFor Arias intensity, m/s; a _cFor critical acceleration.

The ultimate principle of Newmark displacement computation model is conventionally known to one of skill in the art, and the Newmark model after the simplification shown in above-mentioned formula reduces the requirement of seismologic parameter, makes its application be more prone to.

Step S130, with a plurality of fundamental geological correlation parameters and Newmark displacement D _nAs the independent variable factor, build the binary Logistic regression equation I-7 of slump hazard prediction model,

$P=\frac{\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}{1+\exp ({\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+\·\·\·+{\mathrm{\β}}_4{x}_4)}$ Formula I-7

In formula, P is the probability of happening on avalanche after the shake of arbitrary position, landslide in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

As previously mentioned, owing in the Newmark model, failing to consider the influence factor that the rear slump disaster of shake occurs comprehensively, in this step, utilize the Logistic regression model, the Newmark displacement D that is calculating _nOn basis, supplement factor of influence, set up the rear slump hazard prediction model of shake.In the present invention, the factor of influence that supplements remains predicted position apart from the distance in river, apart from zone of fracture distance and topographic relief degree.

Step S140, utilize known conditions to calculate in described binary Logistic regression equation the partial regression coefficient of Variable Factors separately.

In this step, for the slump point that has obtained, in binary Logistic regretional analysis, can be 1 by the dependent variable assignment.For non-slump point, in binary Logistic regretional analysis, can be 0 by the dependent variable assignment.Based on the coordinate of slump point and non-slump point, the topographic relief degree on the respective point position, apart from zone of fracture distance, apart from river distance and Newmark displacement D _nDeng the numerical value of the independent variable factor, can calculate the partial regression coefficient of the independent variable factor.

Step S150, utilize the partial regression coefficient that calculates to build the slump hazard prediction model of this slump disaster space to be predicted.

The coefficient substitution equation I-7 that will obtain at step S140, obtain according to slump hazard prediction model of the present invention.

According to a useful aspect of the present invention, can be from the statistics of the earthquake that occurred and slump disaster, obtaining relevant geologic parameter and seismologic parameter and building forecast model, be applied to the forecast model that obtains with the zone of the region adjacent that earthquake occurs or slump hazard prediction after having the shake in area of similar geographic entity.According to another useful aspect of the present invention; can utilize method of the present invention from the relevant geologic parameter of Real-time Obtaining and seismologic parameter occurent earthquake and slump disaster and build forecast model; and after utilizing constructed forecast model to the contingent shake of this earthquake, predict the position of slump disaster, effectively to protect the safety of country and individual lives and properties.

Below in conjunction with the preferred embodiment of the present invention, illustrate the Forecasting Methodology according to slump disaster after earthquake of the present invention.

Embodiment 1

Below will, take slump disaster analysis after China's Wenchuan County in Sichuan Province shake as example, illustrate according to the forecast model construction method of slump disaster occurs after earthquake of the present invention.

According to the data in " the emergent investigation of Wenchuan earthquake disaster area spacer remote sensing " atlas of Ministry of Land and Resources, China Geological Survey Bureau's establishment, show, after the shake of Wenchuan County, actual slump point has 1904, in binary Logistic regretional analysis, by the dependent variable assignment of these points, be 1, expression slump point.For non-slump point, in this embodiment, utilize the create random points instrument in ArcGIS to generate at random.Consider actual slump point distributes in certain area coverage uncertainty, embodiment is for example take actual slump point as the center of circle, and 200m is the occurrence scope that radius marks actual slump point, supposes outside this scope for the zone on avalanche, landslide does not occur.Generate 2000 some positions at random in zone on avalanche, landslide does not occur for this, assignment is 0, represents non-slump point.

At first, based on slump point as above and non-slump point, extract coordinate on the respective point position, topographic relief degree, apart from the zone of fracture distance, apart from the river distance, and be used to calculating Newmark displacement D _nVarious geography information and seismologic parameter, for regretional analysis.The Wenchuan County River Data is to utilize ArcGIS herein, based on regional dem data, carries out the water system distributed data that extracts after hydrological analysis.

Subsequently, based on Newmark displacement computation model, determine Newmark displacement D _n.

1. seismologic parameter

It is mainly magnitude of earthquake, epicentral location and source depth that Newmark simplifies the seismologic parameter that relates in the displacement computation model.5.12 after the violent earthquake of Wenchuan, basic seismologic parameter and the earthquake intensity figure that Wenchuan earthquake comprises earthquake magnitude, time, source depth and epicenter coordinate etc. all issued at China Seismology Bureau and US Geological Survey country's earthquake information center, sees Fig. 2.In the Newmark simplified model, the parameter in Arias strength calculation formula I-5 is moment magnitude, so the magnitude parameter in the present embodiment adopts 7.9 grades of US Geological Survey, the result that other seismologic parameter adopts China Seismology Bureau to announce.

From Fig. 2, find out, the Wenchuan earthquake earthquake centre is positioned at the Ying Xiu town, source depth 14km(China Seismology Bureau), the east of breaking northwards is to expansion.Further, northwards east is to extension from earthquake centre in the magistoseismic area of XI earthquake intensity, and area is larger, only with the hypocentral distance distance, describes earthquake impact and the reality of surface rupture is not inconsistent.Therefore, the present embodiment consider to adopt earthquake intensity be the center line in XI degree zone as linear focus, source depth adopts 14km, calculates hypocentral distance r.

2. engineering geology parameters

In the Newmark simplified model, participate in determining critical acceleration a _cImportant parameter---slope static security coefficient F _sCalculating need according to local landform and lithology data.Wherein lithology data, adopt engineering rock group data to replace in the present embodiment.Engineering rock group data consist of engineering rock component Butut and corresponding physical and mechanical parameter.The present embodiment is determined Wenchuan County engineering rock component Butut with reference to the Wenchuan earthquake severely afflicated area engineering petrofabric diagram of people's compilations such as Qi Shengwen, see Fig. 3, referring to Wenchuan earthquake utmost point severely afflicated area geologic background and secondary slope disaster space law of development [J]. engineering geology journal, 2009, (1): 39-49.

Corresponding engineering rock group physical and mechanical parameter, as effective angle of internal friction ( ), effectively the main reference " Standard for classification of engineering rock masses " such as cohesion (C'), severe (γ) are (GB50218-94); " Code for design of building " (GB50007-2002); " Code for investigation of geotechnical engineering " (GB50021-2001), shows 1-1,1-2, the standards such as 1-3.Due to the actual rock mass discontinuity that represented of the potential slipping plane in the Newmark displacement model, and structural plane has been controlled rock mass strength to a great extent, so the present embodiment uses the empirical parameter of structural plane to carry out computational analysis.

According to the field study result, show general 3～4 groups of cracks, the rock crushing of growing of the rock side slope that the hard rock in northeast, Wenchuan County forms, integrality is poor, according to table 1-1,1-2,1-3, adjust accordingly its lithologic parameter, and parameter corresponding to each engineering rock group of Wenchuan County is as shown in table 1-4.

The qualitative division of table 1-1 rock mass completeness

The classification of table 1-2 rock mass basic quality

Table 1-3 rock mass and structural plane physical and mechanical parameter

Table 1-4 Wenchuan County engineering rock group physical and mechanical parameter

Lithology	The lithology code	Severe (N/m 3)	Angle of internal friction (°)	Cohesive strength (Pa)
					Hard rock	Ⅰ	27500	29	120000
Than hard rock	Ⅱ	26500	24	100000
					Than soft rock	Ⅲ	25500	19	80000
Soft rock	Ⅳ	23500	16	70000
					The dead-soft rock	Ⅴ	21500	13	50000

Table 1-5 Wenchuan County engineering geology rock group strength reduction factor

In addition, consider that rock mass meets water correction, the effective cohesion of its intensity index and angle of internal friction are in fact low than the empirical value in table 1-4, so the empirical value in his-and-hers watches 1-4 carries out respectively reduction, and it is table 1-5 that reduction is closed.

3. other parameter

Determine side slope critical acceleration a _cValue, be mainly to depend on areal geology, landforms and hydrological environment.From the expression formula of above-mentioned formula I-2, formula I-3, can find out: domatic inclination angle, ground shear strength, the water-bearing zone degree of depth and the landslide surface degree of depth are determining a _cSize.Wherein, regional ground shear strength parameter is discussed in engineering geology parameters one joint, mainly solve remaining other parameters here.

Side slope surface inclination angle (β) calculates by regional DEM altitude figures.This paper dem data derives from the digital elevation data product of the 30m resolution of national science data service platform issue, by data splicing, fusion, cutting, obtain Wenchuan County DEM elevation distribution plan, thereby calculate the terrain slope distribution plan (30m * 30m), see Fig. 4 in the whole county.The most of regional terrain slope in Wenchuan County is more than 30 °, and is less less than the slope distribution area of 10 °, mainly concentrates on De Xuankou town, the southeast, water mill town one band.

The landslide surface degree of depth in the Newmark simplified model is landslide thickness, dissimilar landslide, its sliding mass thickness difference.After the Wenchuan earthquake shake, the SURVEYING OF LANDSLIDE analytical data is bright, and the landslide degree of depth in area, Wenchuan is generally less than 3m.In the Classification of Landslides that carries out of thickness of landslide, sliding mass thickness<6m's be shallow failure, and therefore, the landslide after Wenchuan shakes mainly belongs to shallow failure, just in time meets the optimum condition that Newmark displacement computation model is applicable to shallow failure.But for zone, still there is uncertainty in landslide thickness, the people such as Khazai have proposed the relation of grade of side slope and sliding broken thickness, landslide thickness reduces with the increase of the gradient, sees Khazai, B., Sitar, N.Landsliding in native ground:a GIS-basedapproach to regional seismic slope stability assessment, report Http:// www.ce.berkeley.edu/*khazai/Research/, 2000.The present embodiment is determined landslide thickness according to this relation, in Table 1-6.

Table 1-6 Wenchuan landslide thickness and gradient relation

Grade of side slope	Landslide thickness
		0-30°	4m
30-40°	3m
		40-60°	2m
>60°	1m

The water-bearing zone thickness of sliding mass is also the parameter that will consider in calculating.Wenchuan County located in subtropical zone moist climate band, have a humid climate, and affected by SE Monsoon and southwest monsoon, be in again on windward slope in addition, and high temperature and rainy, annual precipitation is 700～1200mm, and mainly concentrates on for 5～September.Therefore, this paper considers that rainfall infiltration causes sliding mass saturated situation fully, gets the ratio m=1 of saturated sliding mass to integral thickness.

4. based on Newmark displacement D after the shake of Newmark model _n

Based on the ArcGIS software platform, utilize the Newmark simplified model can realize shaking the evaluation of rear slump disaster space liability.According to above-mentioned formula I-2, I-3, I-5 and formula I-6, according to regional geotechnical property and grade of side slope, analysis of slope is because avalanche, the landslide liability that self build-in attribute causes is critical acceleration a _cThe size of calculating the suffered Earthquake Intensity in different regions is Arias intensity; Based on the critical acceleration that calculates and Arias intensity, calculate Newmark displacement D _n.

(1) critical acceleration a _c

Based on regional rock-soil mechanics intensity and regional slope data, adopt formula I-2, I-3 to calculate static security coefficient and critical acceleration, result is as shown in Figure 5 and Figure 6.The critical acceleration distribution plan has characterized under identical seismic dynamic loading background, the slump easy-suffering level difference that causes due to the side slope intrinsic property.Do not consider the impact of earthquake motion on zones of different, the slump disaster more easily occurs in the zone that critical acceleration is less.

(2) Arias intensity I _a

According to above-mentioned formula I-5, calculate Arias intensity, the affect size of reflection earthquake motion on zones of different, result is shown in Figure 7.Because herein hypocentral distance is that distance between center line according to distance XI degree intensity area calculates, therefore, the decline trend of Arias intensity is similar to the earthquake intensity decline mode.

(3) Newmark displacement D _n

Based on above-mentioned critical acceleration a _cWith the Arias intensity I _aResult of calculation, utilize above-mentioned formula I-6 to calculate the Newmark displacement D of study area _nDistribution plan, be shown in Fig. 8.

Subsequently, with a plurality of fundamental geological correlation parameters and the Newmark displacement D that calculates _nAs the independent variable factor, build the binary Logistic regression equation of slump hazard prediction model.

The present embodiment is on the basis of Newmark model, the factors such as river, zone of fracture and landform of considering all affect the distribution of slump disaster, Newmark displacement and river, zone of fracture and topographic relief degree are combined, utilize binary Logistic regression model to set up probability function, carry out the probability of happening prediction of the rear slump disaster of zone shake, see formula I-7.

Each partial regression coefficient β in the Logistic regression equation of table 2-1 Wenchuan County

Variable	Factor beta	Standard error	Wald?χ 2	Degree of freedom	The level of signifiance	Exp(B)
							Constant	2.438	0.714	11.669	1	0.001	11.456
Ln (apart from the river distance)	-0.798	0.035	505.316	1	0.000	0.450
							Ln (apart from the zone of fracture distance)	-0.431	0.035	149.063	1	0.000	0.650
Ln (topographic relief degree)	1.272	0.139	83.786	1	0.000	3.568
							ln(Dn)	0.318	0.040	64.413	1	0.000	1.375

Subsequently, by 1904 slump points obtaining in step formerly and 2000 slump points apart from the river distance, apart from the zone of fracture distance, topographic relief degree and the D that calculates _nThe above-mentioned formula I-7 of value substitution, solve this equation I-7 and obtain each partial regression coefficient β in the Logistic regression equation of Wenchuan County, as shown in table 2-1.

Each partial regression coefficient substitution formula I-7 in table 2-1 is obtained to the occurrence Probability Model formula I-8 of the rear slump disaster of mountain area, Wenchuan County shake:

$P=\frac{\exp (2.438-0.798x_1-0.431x_2+1.272x_3+0.318x_4)}{1+\exp (2.438-0.798x_1-0.431x_2+1.272x_3+0.318x_4)}$ Formula I-8

In formula, P is the probability of happening of the rear slump of arbitrary position shake in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

From the table 2-1 partial regression coefficient can find out, the slump probability of happening with apart from river apart from, apart from zone of fracture apart from becoming negative correlativing relation, with Newmark displacement D _nBecome positive correlation.Exp (B) is odds ratio, and expression unit of the every variation of independent variable, the probability that slump occurs and the ratio of probability of occurrence not are the multiples of corresponding ratio before changing.Exp (B) has reflected the strength of association of influence factor and slump disaster, as Exp (B) > 1, the risk factor of influence factor and slump disaster increases, positive correlation; Exp (B)<1, the risk factor of influence factor and slump disaster reduces, negative correlation; Exp (B)=1, the risk factor of influence factor and slump disaster is irrelevant.2-1 can find out by table, D _nBe worth larger, less apart from river distance, apart from the less zone of zone of fracture distance, more easily come down.

The check of forecast model fitting effect

As mentioned above, by binary Logistic regretional analysis, set up the forecast model formula I-8 of the rear slump disaster of Wenchuan County shake.The present embodiment carrys out the accuracy of verification model by the ROC curve.

The ROC curve is the abbreviation of experimenter's operating characteristic (Receiver Operating Characteristic), and it is a kind of data statistical approach of widespread use.Application ROC curve can help the researcher to determine rational probabilistic classification point, is about to probability and is judged as result (or result does not occur) occurs greater than (or less than) what research object.When forecast result of model was best, the ROC curve should be from lower left corner vertical uplift to top, and then horizontal direction extends to the upper right corner to the right.Fig. 9 is the ROC curve that after the shake of Wenchuan County, slump disaster occurrence Probability Model formula I-8 is applied to the area, Wenchuan, and its form meets the ROC tracing pattern of optimum prediction effect substantially, and is visible, and this model is better to the prediction effect in area, Wenchuan.

Usually, by calculating ROC area under curve (Area Under Curve is called for short AUC), carry out quantificational expression model prediction success ratio.Fig. 9 is carried out to AUC calculating, and in Table 2-2, the result that obtains is that the success rate prediction of the Logistic regression model set up of embodiment 1 reaches 85.3%, and prediction effect is good, can be used in slump disaster probability of happening prediction after actual shake.

Table 2-2AUC statistical study

Figure 10 has provided size and actual avalanche, a landslide disaster point distribution situation of avalanche after the shake of slump disaster occurrence Probability Model formula I-8 prediction after the shake of setting up according to embodiment 1, landslide disaster probability of happening.As can be seen from this figure, the zone that after shake, slump disaster probability of happening is high is relatively concentrated, and mainly stream, zone of fracture are the ribbon distribution along the river.All in all, in east and northern zone, the area in slump disaster zone occurred frequently is larger, mainly due to the structure zone of fracture, mainly is distributed in these areas, simultaneously Newmark shift value D _nAlso higher in eastern region, these combined factors have got up to control macroscopical distribution range of slump disaster.From the probabilistic forecasting result of slump disaster generation and the contrast effect of actual slump point position, the high-risk danger zone of slump disaster is more consistent with the distribution of actual slump disaster.

Comparative Examples 1

Figure 11 has provided Wenchuan County Newmark shift value D _nDistribute and actual avalanche, landslide disaster point position distribution situation.As can be seen from this figure, D _nThe distributed areas of the zone of 3cm and actual slump point position relatively coincide substantially, but by extracting the D of a slump point respective point _nValue, and statistics drops on D _nActual slump point number in the 3cm zone finds, D _nActual slump incidence in the 3cm zone is only 35% left and right, visible Newmark model is better to the prediction of slump disaster distribution trend, but lower to the accuracy of single slump hazard prediction.

Newmark simplifies the difference that displacement model mainly depends on side slope rock signature and earthquake motion impact, although more comprehensively considered engineering rock signature and triggering factors that avalanche, landslide occur, has ignored regional landform, landforms general layout feature.For example, lack the river cutting action is considered, the consideration that shortage affects rift structure etc.

Relatively Comparative Examples 1 can be found out, high to the predictablity rate of large-scale dangerous zoning drawn game section danger position according to slump disaster occurrence Probability Model I-8 after shake of setting up based on Newmark Displacement Analysis and binary Logistic regretional analysis of the present invention, prediction effect is good, can take into account two aspects of both macro and micro.

Embodiment 2

Prediction effect when this embodiment verifies that by the ROC curve forecast model formula I-8 that embodiment 1 sets up is applied to Sichuan Province's Beichuan County.

Beichuanqiangzu Autonomous County is positioned at In Northwest of Sichuan Basin.Geographic coordinate: 31 ° 35 '-31 ° 38 ' 2 of north latitude ", 104 ° 26 ' 15 of east longitude "-104 ° 29 ' 10 ".East connects Jiangyou City, and southern adjacent An County west depends on ，Bei Di Songpan, Mao County, Pingwu, and the territory area is 2867.83 sq-kms.

According to slump disaster occurrence Probability Model formula I-8 after the shake of inventive embodiments 1 foundation, the probability of happening forecasting process that avalanche, landslide disaster are brought out in regional earthquake to Beichuan County mainly is divided into three parts: obtain geographic factor, geologic parameter and seismologic parameter, calculate this regional Newmark displacement D _nWith the effect of estimating the method according to this invention and the forecast model of setting up and the model of setting up.

At first, according to step as described in Example 1, obtain the various correlation parameters of Beichuan County.

Subsequently, utilize the geologic parameter and the seismologic parameter that obtain to calculate Newmark displacement D _n.

The Beichuan County slope map, take the DEM of 30m * 30m as basis, obtains with the ratio of horizontal range by the difference of elevation that calculates between the adjacent cells unit.Correlation parameter is brought into to the Newmark accumulation shift value D that calculates each grid point value in Beichuan County in formula I-2, formula I-3, formula I-5 and formula I-6 _n, analyzed area is shaken the liability of rear slump disaster accordingly.Figure 12 is the Beichuan County Newmark accumulative displacement distribution situation that calculates.

Subsequently, utilize and calculate the Beichuan County geographic factor that the Newmark accumulative displacement distributes and obtains, the forecast model formula I-8 that sets up can be used for to the prediction to slump disaster probability of happening after the Beichuan County shake in embodiment 1.

In order to check the effect of forecast model at Beichuan County, the present embodiment, by ArcGIS 617 non-slump points of random generation outside the 200m scope of 617 actual slump points of Beichuan County, is analyzed the prediction effect of this model at Beichuan.

Equally, adopt the ROC curve to carry out the testing model prediction effect.Figure 13 is the ROC check curve of forecast model formula I-8 after the Beichuan County application that embodiment 1 sets up.From on tracing pattern, the forecast model of embodiment 1 has predictive ability preferably equally to the probability of happening of the rear slump disaster of Beichuan County shake.From AUC result of calculation, in Table 2-3, the forecast model that embodiment 1 sets up is about 80.3% to the success rate prediction of the rear slump disaster of Beichuan County shake.

Table 2-3AUC statistical study

In sum, the forecast model set up of embodiment 1 is better to the prediction effect of the avalanche of Beichuan County earthquake-induced, Landslide hazards.This model is based on the forecast model that the actual slump point in Wenchuan is set up, by this embodiment, at the application verification of Beichuan County, show that this model has and expands preferably applicability for similar area, can be used in the quantification risk assessment of the earthquake slump disaster in Southwestern China mountain area, for the control of shaking rear Secondary Geological Hazards lays the foundation.

Embodiment 3

Below will, take slump disaster analysis after China's Sichuan Province's Beichuan County shake as example, illustrate according to the forecast model construction method of slump disaster occurs after earthquake of the present invention.

At first, obtain various correlation parameters of Beichuan County determine Newmark displacement Dn based on Newmark displacement computation model.

Concrete steps are referring to embodiment 2 related contents, and result of calculation as shown in figure 12.

Subsequently, based on 617 actual slump points that occur in Beichuan and 617 non-slump points that generate at random outside the 200m of the actual slump point of Beichuan County scope, dependent variable assignment to slump point is 1, dependent variable assignment to non-slump point assignment is 0, utilize the various geographic factors that obtain and the Newmark displacement Dn that calculates etc. as known conditions, I-7 solves to binary Logistic regression equation, obtain each partial regression coefficient β in Beichuan County Logistic regression equation, as shown in Table 2-4.

Each partial regression coefficient β in table 2-4 Beichuan County Logistic regression equation

Variable	Coefficient	Standard error	Wald?χ 2	Degree of freedom	The level of signifiance	Exp(B)
							Constant	1.297	0.689	3.541	1	0.060	3.658
Ln (apart from the river distance)	-0.524	0.032	271.904	1	0.000	0.592
							Ln (apart from the zone of fracture distance)	-0.436	0.034	166.075	1	0.000	0.646
Ln (topographic relief degree)	1.134	0.133	72.381	1	0.000	3.107
							ln(Dn)	0.314	0.038	68.762	1	0.000	1.368

Partial regression coefficient substitution formula I-7 in table 2-4 is obtained to the probability of happening forecast model formula I-9 of the rear slump disaster of Beichuan County mountain area shake:

$P=\frac{\exp (1.297-0.524x_1-0.436x_2+1.134x_3+0.314x_4)}{1+\exp (1.297-0.524x_1-0.436x_2+1.134x_3+0.314x_4)}$ Formula I-9

In formula, P is the probability of happening on avalanche after the shake of arbitrary position, landslide in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

Figure 14 shows the ROC curve that is applied to Beichuan County of the rear slump disaster occurrence Probability Model I-9 of shake.By calculating ROC area under curve AUC, in Table 2-5, carry out quantificational expression model prediction success ratio, the result that obtains is that the success ratio that the prediction Beichuan slump disaster of the Logistic regression model formula I-9 of embodiment 3 foundation occurs reaches 84.5%, prediction effect is good, can be used in slump disaster probability of happening prediction after actual shake.

Table 2-5AUC statistical study

The size of avalanche after the shake of the probability model formula I-9 prediction that after Figure 15 had provided and shaken according to the prediction of embodiment 3 foundation, the slump disaster occurred, landslide disaster probability of happening and actual avalanche, a landslide disaster point distribution situation.As can be seen from this figure, the zone that after shake, slump disaster probability of happening is high is relatively concentrated, and mainly stream, zone of fracture, road are the ribbon distribution along the river.From the probabilistic forecasting result of slump disaster generation and the contrast effect of actual slump point position, the high-risk danger zone of slump disaster is more consistent with the distribution of actual slump disaster.

Embodiment 4

This embodiment verifies the prediction effect of the forecast model formula I-9 of embodiment 3 foundation to Wenchuan County by two indexs of ROC curve.

According to slump disaster occurrence Probability Model formula I-9 after the shake of inventive embodiments 3 foundation, the probability of happening forecasting process that the Wenchuan County regional earthquake is brought out to avalanche, landslide disaster mainly is divided into three parts: obtain geographic factor, geologic parameter and seismologic parameter; Calculate this regional Newmark displacement Dn and estimate applied disaster occurrence Probability Model.

Obtain the relevant geographic factor in Wenchuan County, geologic parameter, the related content of seismologic parameter and calculating Newmark displacement Dn specifically describes in embodiment 1, repeat no more here.By each correlation parameter substitution of resulting Wenchuan is set up in embodiment 3 forecast model formula I-9, can be used for the prediction to slump probability after the shake of Wenchuan County.

Table 2-6AUC statistical study

Figure 16 is the ROC check curve after the application of Wenchuan County for slump disaster probability of happening forecast model formula I-9 after shaking, and from tracing pattern, forecast model formula I-9 has predictive ability preferably to the probability of happening of the rear slump disaster of Wenchuan County shake.From AUC result of calculation, in Table 2-6, the forecast model formula I-9 of embodiment 3 is about 82.9% to the success rate prediction of the rear slump disaster of Wenchuan County shake, and prediction effect is good equally.

Embodiment 5

Below will, take slump disaster analysis after the shake of Mianzhu City, China Sichuan Province as example, illustrate according to the forecast model construction method of slump disaster occurs after earthquake of the present invention.

At first, obtain the various correlation parameters in Mianzhu City and determine Newmark displacement Dn based on Newmark displacement computation model, result of calculation as shown in figure 17.

Subsequently, based on 281 actual slump points that occur in Mianzhu City and 281 non-slump points that generate at random outside the 200m of the actual slump point in Mianzhu City scope, dependent variable assignment to slump point is 1, dependent variable assignment to non-slump point assignment is 0, utilize binary Logistic regression equation E, obtain each partial regression coefficient β in the Logistic of Mianzhu City regression equation, as shown in table 2-7.

Partial regression coefficient substitution formula I-7 in table 2-7 is obtained to the probability of happening forecast model formula I-10 of the rear slump disaster of mountain area, Mianzhu City shake:

Each partial regression coefficient β in the table Logistic of 2-7 Mianzhu City regression equation

Variable	Coefficient	Standard error	Wald?χ 2	Degree of freedom	The level of signifiance	Exp(B)
							Constant	-1.534	1.724	0.792	1	0.374	0.182
Ln (apart from the river distance)	-1.633	0.173	88.877	1	0.000	0.195
							Ln (apart from the zone of fracture distance)	-0.056	0.061	0.840	1	0.360	0.945
Ln (topographic relief degree)	2.722	0.321	72.151	1	0.000	15.218
							ln(Dn)	0.394	0.131	9.091	1	0.000	1.482

$P=\frac{\exp (-1.534-1.633x_1-0.056x_2+2.722x_3+0.394x_4)}{1+\exp (-1.534-1.633x_1-0.056x_2+2.722x_3+0.394x_4)}$ I-10

In formula, P is the probability of happening on avalanche after the shake of arbitrary position, landslide in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

Figure 18 shows the ROC curve that the rear slump disaster occurrence Probability Model formula I-10 of shake is applied to Mianzhu City.By calculating the ROC area under curve, be called for short AUC, in Table 2-8, carry out quantificational expression model prediction success ratio, the result that obtains is that the success ratio that prediction Mianzhu City slump disaster of the Logistic regression model of embodiment 5 foundation occurs reaches 92.6%, prediction effect is good, can be used in slump disaster probability of happening prediction after actual shake.

The size of avalanche after the shake of the probability model formula I-10 prediction that after Figure 19 had provided and shaken according to the prediction of embodiment 5 foundation, the slump disaster occurred, landslide disaster probability of happening and actual avalanche, a landslide disaster point distribution situation.As can be seen from this figure, the zone that after shake, slump disaster probability of happening is high is relatively concentrated, and mainly stream, zone of fracture, road are the ribbon distribution along the river.From the probabilistic forecasting result of slump disaster generation and the contrast effect of actual slump point position, the high-risk danger zone of slump disaster is more consistent with the distribution of actual slump disaster.

Table 2-8AUC statistical study

Should be appreciated that the above detailed description of technical scheme of the present invention being carried out by preferred embodiment is illustrative and not restrictive.Those of ordinary skill in the art is reading on the basis of instructions of the present invention and can modify to the technical scheme that each embodiment puts down in writing, or part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (10)

1. the Forecasting Methodology of slump disaster position occurs after an earthquake, and the method comprises the following steps:

Obtain a plurality of fundamental geological parameters in space to be predicted, a plurality of geologic parameters, seismologic parameter, the geographic position of the geographic position of at least one known slump point and a plurality of not slump points;

Based on the Newmark displacement model, calculate Newmark displacement D _n

With at least a portion in a plurality of fundamental geological correlation parameters and Newmark displacement D _nAs the independent variable factor, build the binary Logistic regression equation of slump hazard prediction model;

By the described geographical correlation parameter of described known slump point and ,Ge position, described not slump point geographic position and the Newmark displacement D that calculates _nAs the known conditions of described binary Logistic regression equation, calculate in described binary Logistic regression equation the partial regression coefficient of Variable Factors separately;

The partial regression coefficient that utilization calculates builds the slump hazard prediction model in this space to be predicted.

2. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 1, it is characterized in that, described fundamental geological parameter comprises one or more in the gradient, topographic relief degree, zone of fracture position, position, river; Described geologic parameter comprises one or more in rock mass completeness, rock-mass quality, rock mass physical mathematic(al) parameter and rock group intensity; Described seismologic parameter comprises earthquake magnitude, one or more in source depth and epicenter coordinate.

3. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 1, it is characterized in that, the binary Logistic regression equation of described forecast model is as follows:

$P=\frac{\exp ({\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+\&CenterDot;\&CenterDot;\&CenterDot;+{\mathrm{\&beta;}}_4{x}_4)}{1+\exp ({\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+\&CenterDot;\&CenterDot;\&CenterDot;+{\mathrm{\&beta;}}_4{x}_4)}$

Wherein, P is the probability of happening of the rear slump of arbitrary position shake in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n).

4. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 1, it is characterized in that, described non-slump point utilizes create random points instrument random generation the outside actual slump point in ArcGIS.

5. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 4, it is characterized in that, described non-slump point generates at random in the scope outside actual slump point 200m.

6. the Forecasting Methodology of slump disaster position occurring after earthquake as claimed in claim 1, it is characterized in that, according to following formula, calculates Newmark displacement D _n,

log?D _n＝1.521log?I _a-1.993log?A _c-1.546±0.375

Wherein, I _aFor Arias intensity, m/s; A _cFor critical acceleration.

7. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 6, it is characterized in that,

$\log I_a=M-2\log \sqrt{r^2{+7.5}^2}-3.99\&PlusMinus;0.5$

Wherein, M is moment magnitude, and r is hypocentral distance.

8. the method for slump disaster position occurring after prediction earthquake as claimed in claim 7, it is characterized in that, using the center line in maximum earthquake intensity zone as linear focus, calculates hypocentral distance.

9. the Forecasting Methodology of slump disaster position occurs after earthquake as claimed in claim 6, it is characterized in that, the method is applicable to the slump hazard prediction of sliding mass thickness less than 6m.

10. the Forecasting Methodology of slump disaster position occurs after a Southwest China earthquake, comprising:

Obtain a plurality of fundamental geological parameters in space to be predicted, a plurality of geologic parameters, seismologic parameter, the geographic position of the geographic position of at least one known slump point and a plurality of not slump points;

Based on the Newmark displacement model, calculate Newmark displacement D _n

Utilize the probability of generation slump disaster in position to be predicted in following formula calculating prediction space,

$P=\frac{\exp ({\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+\&CenterDot;\&CenterDot;\&CenterDot;+{\mathrm{\&beta;}}_4{x}_4)}{1+\exp ({\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+\&CenterDot;\&CenterDot;\&CenterDot;+{\mathrm{\&beta;}}_4{x}_4)}$

P is the probability of happening of the rear slump of arbitrary position shake in space to be predicted;

x ₁..., x ₄Be respectively ln (this position is far from the river distance), ln (this position is apart from the zone of fracture distance), ln (topographic relief degree) and ln (D _n),

Wherein,

β ₀=2.438～1.297；

β ₁=-0.798～-0.524；

β ₂=-0.431～-0.436；

β ₃=1.272～1.134；

β ₄=0.318～0.314。

Images