A multi-time-scale population spatialization method for geohazard vulnerability assessment

CN118644082BActive Publication Date: 2024-12-03HUNAN UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411092246.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-12-03
Estimated Expiration
2044-08-09

AI Technical Summary

Technical Problem

The existing technology lacks dynamic research with high time resolution in the vulnerability of landslide disasters, resulting in low efficiency of pre-disaster preparation and post-disaster rescue, large economic and population losses, and vulnerability of geological disasters in urban and rural planning and land use. Not fully considered.

Method used

Using a multi-time scale population spatialization method, the geological disaster vulnerability assessment method is adopted, and the population distribution in the study area is spatialized by multi-time scale spatialization, and the geological disaster vulnerability assessment unit is divided, vulnerability indicators are calculated, and the geological disaster vulnerability is constructed. Loss evaluation model, analyze the results of vulnerability evaluation under different time scales.

Benefits of technology

Identify high-risk vulnerable areas, make pre-disaster preparations in advance, optimize emergency resource allocation, improve rescue efficiency, reduce economic and population losses, guide urban and rural planning and land use, and reduce the impact of disasters.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118644082B_ABST
    Figure CN118644082B_ABST
Patent Text Reader

Abstract

The present invention belongs to the technical field of geological disaster risk assessment, and discloses a geological disaster vulnerability assessment method with multi-time scale population spatialization, including the following steps: S1, multi-time scale spatialization of population distribution in the study area; S2, division of geological disaster vulnerability assessment units; S3, calculation of various vulnerability indicators; S4, construction of geological disaster vulnerability assessment model; S5, analysis of vulnerability assessment results at different time scales. The present invention adopts the above-mentioned geological disaster vulnerability assessment method with multi-time scale population spatialization to identify high-risk vulnerable areas, carry out various pre-disaster preparations in advance, improve disaster defense capabilities, optimize emergency resource allocation, fully consider the vulnerability of geological disasters, improve rescue efficiency, reduce economic and population losses, guide urban and rural planning and land use, fully consider the vulnerability of disasters, reduce the impact of disasters, and reduce economic losses and population casualties caused by disasters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster risk assessment, and in particular to a geological disaster vulnerability assessment method with multi-time scale population spatialization. Background Art

[0002] Many areas have fragile geological environments and steep terrain, which are prone to geological disasters. In some areas, the soft mountains are prone to landslides, especially in heavy rainfall seasons, which can easily lead to landslides and mud-rock flows. In recent years, due to the development of karst and the influence of human engineering activities, landslide disasters have been intensifying, and geological disasters have occurred frequently, causing significant economic and population losses, and seriously hindering local economic development.

[0003] Geological hazard risk assessment based on multi-time scale population spatialization simulation results is of great significance to disaster management. It can capture the dynamic changes of population in the short and long term, such as seasonal migration, daily population mobility, and long-term population migration trends. Understanding these dynamic changes helps to more accurately assess population vulnerability. Multi-time scale population spatialization provides rich data support and new analytical perspectives for geological hazard vulnerability assessment, which can significantly improve the comprehensiveness and accuracy of assessment results, thereby providing a scientific basis for formulating effective prevention, response and recovery strategies.

[0004] Population spatialization refers to the distribution of demographic data summarized by administrative regions to a geographic grid of a specified size according to certain weights, so as to realize the transformation of the spatial expression of population from administrative units to geographic grids. According to existing research data, there are many methods for population spatialization simulation, and the land use type estimation method is the most widely used. However, there are few existing studies on landslide vulnerability alone, and most of them are static studies. The time scale of the dynamic research on geological hazard vulnerability is relatively large, and there is a lack of dynamic landslide vulnerability research with high time resolution. Therefore, the present invention proposes a multi-time scale population spatialization geological hazard vulnerability assessment method to provide a basis for pre-disaster preparation, post-disaster rescue, etc. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-time scale population spatialized geological disaster vulnerability assessment method, identify high-risk vulnerable areas, carry out various pre-disaster preparations in advance, improve disaster defense capabilities, optimize emergency resource allocation, fully consider the vulnerability of geological disasters, improve rescue efficiency, reduce economic and population losses, guide urban and rural planning and land use, fully consider the vulnerability of disasters, reduce the impact of disasters, and reduce economic losses and casualties caused by disasters.

[0006] To achieve the above object, the present invention provides a multi-time scale population spatialized geological hazard vulnerability assessment method, comprising the following steps:

[0007] S1. Spatialize the population distribution in the study area at multiple time scales;

[0008] S2. Divide geological hazard vulnerability assessment units;

[0009] S3. Calculate each vulnerability index;

[0010] S4. Construct a geological hazard vulnerability assessment model;

[0011] S5. Analysis of vulnerability assessment results at different time scales.

[0012] Preferably, in step S1, the population distribution in the study area is spatialized at multiple time scales, specifically divided into:

[0013] S11. Use the land use density method to simulate the short-term population dynamics of the population distribution in the study area;

[0014] S12. A refined population spatialization method based on zoning and identification of real residential areas is used, and a multivariate regression model is used to perform long-term population spatialization simulation.

[0015] Preferably, in step S11, the specific process of the short-period population dynamics simulation is as follows:

[0016] S111. Short-term population dynamics simulation, selecting land use type as an indicator, introducing building attraction rate, and obtaining the population spatial distribution in each time period, as shown below:

[0017]

[0018] Among them, P is the population, A is i is the area of ​​the i-th land use type, D i is the population density of the i-th category, k i is the attraction rate of the i-th land use type;

[0019] S112. Define residents' travel with high temporal resolution based on time geography;

[0020] S113, determining the building attraction rate of each land use type at different times;

[0021] S114. A mathematical model is established based on the building attraction rate in different time periods to simulate short-term population dynamics.

[0022] Preferably, in step S12, the specific process of the long-period population dynamics simulation is as follows:

[0023] S121. Calculate the long-term population spatialization index as follows:

[0024]

[0025] Among them, EAHSI stands for Human Settlement Index, EVI max represents the annual maximum EVI, and DEM represents the elevation of the study area; Luojia nor Represents the standardized night light value, calculated as follows:

[0026]

[0027] Among them, Luojia min Indicates the minimum night light value, Luojia max Indicates the maximum night light value;

[0028] S122. Divide the study area into high economic zone and low economic zone for population spatial modeling, including:

[0029] (1) Establish a grid population estimation model for high economic zones;

[0030] (2) Establish a grid population estimation model for low economic zones;

[0031] S123. Model calibration and validation include: model calibration and error analysis of population spatialization simulation using absolute value of relative error and average relative error;

[0032] S124. Analysis of simulation results.

[0033] Preferably, in step S122, a high economic zone grid population estimation model is established, and the towns in the high economic zone are used as samples, night light data and building data are selected, and a regression model is established to calculate the building area and night light coefficient, as shown below:

[0034] P i =a 1 ·B i +a 2 ·L i +b 1 (4)

[0035] Among them, P i is the population of the ith township; B i is the building area of ​​the ith township, L i is the night light brightness value of the i-th township; a 1 is the coefficient of building area, a 2 is the coefficient of night light brightness; constant term b 1 is the intercept. According to the principle of no buildings and no population, b1 The value of is 0;

[0036] The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid population calculation model. The high economic zone grid population estimation model is as follows:

[0037] P ij =a 1 ·B ij +a 2 ·L ij (5)

[0038] Among them, P ij refers to the population of the ijth grid in the high economic zone, B ij is the building area of ​​the ijth grid, L ij is the night light brightness value of the ijth grid; a 1 is the building area (m 2 ), a 2 is the coefficient of night light brightness value;

[0039] A grid population estimation model for low economic zones was established. The towns in low economic zones were used as samples. The building area index, road network density and modified human settlement index were selected as population spatialization indicators. A population estimation model was established to calculate the coefficients of the building area index, road network density and modified human settlement index, as shown below:

[0040] P j =k 1 ·J j +k 2 ·RSD j +k 3 ·E j +b 2 (6)

[0041] Among them, P j is the population of the jth township; J j is the building density of the jth township, RSD j is the road network density of the jth township, E j is the modified human settlement index of the jth township; k 1 is the coefficient of the building area index, k 2 is the coefficient of road network density, k 3 is the coefficient of the modified human settlement index; the constant term b 2 is the intercept. According to the principle of no buildings and no population, b 2 The value of is 0;

[0042] The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid-scale population calculation model. The low-economic zone grid population estimation model is as follows:

[0043] P' ij =k 1 ·J ij +k 2 ·RSD ij +k 3 ·E ij (7)

[0044] Among them, P' ij is the population of the ijth grid in the low economic zone; J ij is the building density of the ijth grid, RSD ij is the road network density of the ijth grid, E ij is the revised human settlement index of the ijth grid; k 1 is the coefficient of the building area index, k 2 is the coefficient of road network density, k 3 is the coefficient of the modified human settlement index.

[0045] Preferably, in step S123, model calibration and verification specifically include:

[0046] (1) Model calibration: Calculate the relative errors of population statistics and population spatialization in different towns as follows:

[0047]

[0048] Among them, RE i is the relative error of the ith township, POP m To simulate the population, POP s To count the population;

[0049] (2) Error analysis: The absolute value of relative error and the average relative error are used to perform error analysis of population spatialization simulation, as shown below:

[0050]

[0051] Among them, MRE is the mean relative error, n is the number of towns for population spatial simulation;

[0052] Preferably, in step S2, regular grid units of 200m×200m are used to divide the geological hazard vulnerability assessment units.

[0053] Preferably, in step S3, the vulnerability assessment index is graded into population vulnerability, economic vulnerability and material vulnerability; wherein population vulnerability includes population density; economic vulnerability includes GDP density and cultivated land density; and material vulnerability includes building density and road density.

[0054] Preferably, the vulnerability evaluation index calculation method is as follows:

[0055] S31. Population density is obtained based on the results of population dynamics simulation as follows:

[0056]

[0057] Where P m is the population density, P i refers to the population of the ith grid, and A is the area of ​​the grid;

[0058] S32. GDP density refers to the ratio of regional gross domestic product to area, which represents the efficiency of economic activities and the density of land use per unit area of ​​the settlement. The greater the economic density, the higher the vulnerability after a landslide disaster.

[0059] S33, cultivated land density is obtained by obtaining the ratio of cultivated land area to grid area according to land use data, as shown below:

[0060]

[0061] Among them, G i represents the density of cultivated land in grid i, G ai represents the cultivated land area in grid i, and A represents the grid area;

[0062] S34. Building density is obtained based on the ratio of building area to area within the region, as shown below:

[0063]

[0064] Among them, J i represents the building density within grid i, B i represents the building area in grid i, and A represents the grid area;

[0065] S35. Road density is obtained based on the ratio of road length to area in the region, as shown below:

[0066]

[0067] Among them, RSD i represents the road density within grid i, N ri 、N nei 、N pei 、N cri、N tri They represent the lengths of railways, national roads, provincial roads, county roads and township roads in grid i respectively, and A represents the grid area.

[0068] Preferably, in step S4, the evaluation model of geological hazard vulnerability is as follows:

[0069]

[0070] Among them, Y represents the value of geological hazard vulnerability, P i represents the value of the i-th indicator, W i Represents the weight of the i-th indicator.

[0071] Therefore, the present invention adopts the above-mentioned multi-time scale population spatialization geological disaster vulnerability assessment method to identify high-risk vulnerable areas, carry out various pre-disaster preparations in advance, improve disaster defense capabilities, optimize emergency resource allocation, fully consider the vulnerability of geological disasters, improve rescue efficiency, reduce economic and population losses, guide urban and rural planning and land use, fully consider the vulnerability degree of disasters, reduce the impact of disasters, and reduce the economic losses and casualties caused by disasters.

[0072] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flow chart of a multi-time-scale population spatialized geological hazard vulnerability assessment method of the present invention;

[0074] Figure 2 It is the technical roadmap for the short-period population dynamics simulation of the present invention;

[0075] Figure 3 is the population simulation result for time period 1;

[0076] Figure 4 is the population simulation result for time period 2;

[0077] Figure 5 is the population simulation result for time period 3;

[0078] Figure 6 is the population simulation result of time period 4;

[0079] Figure 7 is the population simulation result for time period 5;

[0080] Figure 8 is the population simulation result for time period 6;

[0081] Fig. 9 is the population simulation result for time period 7;

[0082] Fig.10 is the population simulation result on rest days;

[0083] Fig.11 is the weekday population simulation result;

[0084] Fig.12 It is the technical roadmap for the spatialization of the long-term population of the present invention;

[0085] Fig.13 It is the long-term population spatialization human settlement index distribution map of the present invention;

[0086] Fig.14 It is the result of long-period population spatialization in the embodiment of the present invention;

[0087] Fig.15 It is the calculation index of the geological disaster vulnerability of the present invention;

[0088] Fig.16 is the distribution of population vulnerability in time period 1;

[0089] Fig.17 is the distribution of population vulnerability in time period 2;

[0090] Fig.18 is the distribution of population vulnerability in time period 3;

[0091] Fig.19 is the distribution of population vulnerability in time period 4;

[0092] Fig. 20 is the distribution of population vulnerability in time period 5;

[0093] Fig.21 is the distribution of population vulnerability in time period 6;

[0094] Fig. 22 is the distribution of population vulnerability in time period 7;

[0095] Fig.23 is the distribution of population vulnerability on weekdays;

[0096] Fig.24 is the distribution of population vulnerability on rest days;

[0097] Fig.25 is the distribution of GDP vulnerability;

[0098] Fig.26 is the distribution of arable land vulnerability;

[0099] Fig. 27 is the distribution of building vulnerability;

[0100] Fig.28is the road vulnerability distribution;

[0101] Fig.29 is the distribution of landslide geological hazard vulnerability in time period 1;

[0102] Fig.30 is the distribution of landslide geological hazard vulnerability in time period 2;

[0103] Fig.31 is the distribution of landslide geological hazard vulnerability in time period 3;

[0104] Fig.32 is the distribution of landslide geological hazard vulnerability in time period 4;

[0105] Fig.33 is the distribution of landslide geological hazard vulnerability in time period 5;

[0106] Fig.34 is the distribution of landslide geological hazard vulnerability in time period 6;

[0107] Fig.35 is the distribution of landslide geological hazard vulnerability in time period 7;

[0108] Fig.36 It is the distribution of landslide geological hazard vulnerability on rest days;

[0109] Fig.37 is the distribution of landslide geological hazard vulnerability on weekdays;

[0110] Fig.38 It is the distribution of vulnerability to long-term landslide geological hazards. DETAILED DESCRIPTION

[0111] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0112] like Figure 1 As shown, the present invention provides a multi-time scale population spatialization geological disaster vulnerability assessment method, comprising the following steps:

[0113] S1. Spatialize the population distribution in the study area at multiple time scales;

[0114] S2. Divide geological hazard vulnerability assessment units;

[0115] S3. Calculate each vulnerability index;

[0116] S4. Construct a geological hazard vulnerability assessment model;

[0117] S5. Analysis of vulnerability assessment results at different time scales.

[0118] Example

[0119] S1. Spatialize the population distribution in the study area at multiple time scales.

[0120] S11. Use the land use density method to simulate the short-term population dynamics of the population distribution in the study area, such as Figure 2 shown.

[0121] S111. Short-term population dynamics simulation, selecting land use type as an indicator, introducing building attraction rate, and obtaining the population spatial distribution in each time period, as shown below:

[0122]

[0123] Among them, P is the population, A is i is the area of ​​the i-th land use type, D i is the population density of the i-th category, k i is the attraction rate of the i-th land use type.

[0124] S112. Define residents’ travel with high temporal resolution based on time geography.

[0125] The travel patterns of urban residents are the main research contents of behavioral geography and time geography. Behavioral geography believes that an individual's daily activities are always composed of habitual behaviors such as going to work, going home, shopping, etc., and temporary or accidental behaviors are only a minor part, such as traveling during holidays and going home for the New Year.

[0126] Daily activities are a mixture of choices and constraints. On the one hand, residents choose travel activities according to their own life and work needs; on the other hand, residents' choices are affected by external environmental factors, especially social and cultural factors. Time geography studies residents' daily activities on a time scale. By recording the time trajectory of individual activities, it synthesizes more abstract residents' travel patterns and incorporates these patterns into the entire social environment for investigation to discover general laws that conform to the entire society.

[0127] Since residents’ travel has regularity in time, a day is divided according to the travel time of residents, and the high-time resolution of residents’ travel is defined. The travel pattern is as follows:

[0128] (1) For people who need to travel at regular times, such as students and office workers. For students, their travel pattern is generally "home-school-home"; for young people who need to commute to get off work, their travel pattern is generally "home-company-entertainment / home-company-entertainment / home-home". In terms of travel time, commuting and returning home are relatively concentrated, usually with morning peaks and evening peaks.

[0129] (2) For those who travel irregularly, such as retired elderly people, since they have more free time, they usually avoid rush hour. Their main travel pattern is “home-leisure / shopping-home”.

[0130] (3) For rural residents, their travel pattern is generally “home-farmland-home”, and their working hours are generally in the morning and afternoon. Their travel time is similar to that of residents who need to go to work.

[0131] According to the above travel patterns, residents are divided into two categories: those who need to travel regularly (to go to get off work or school) and those who travel freely (the elderly). The day is divided into seven stages according to commuting, going to work, lunch break, getting off work and night, as shown in Table 1.

[0132] Table 1 Time period division

[0133]

[0134] S113. Determine the building attraction rate of each land use type at different times.

[0135] The population estimation of land use type is mainly obtained by land use attraction rate method. Studies have shown that the spatial distribution of population is closely related to the spatial distribution of land use and the travel purpose of residents. Therefore, the attraction weights of various types of land use in the resident travel prediction model based on land use type are introduced to represent the proportion of population distribution of various land use types.

[0136] The portability of land attraction rights makes the attraction rights of various types of land use quite stable. There are two methods to determine the land use attraction rights. One is the direct method, that is, through field surveys, the number of residents' trips and travel purposes in each time period are obtained; the other is the indirect method, which is obtained based on existing research and the actual situation of the study area. The present invention adopts the indirect method, based on existing research and the actual situation of the study area, to obtain the attraction rate of various types of land use at each time.

[0137] The attraction rates of different land use types are different, and the attraction rates of the same land use type at different times are also different, reflecting that different land use types have different abilities to attract populations in different time periods. The value of the attraction rate is between 0 and 1. When the attraction rate is 0, it means that the population accommodated by this land use type is 0 during this time period; when the attraction rate is 1, it means that the population accommodated by this land use type has reached its rated population during this time period.

[0138] During the working hours, the number of people in the administrative office land has reached its quota, so its attraction rate is set to 1 during the working hours. During the lunch break, people will choose to eat at the workplace or go home to rest, and its attraction rate will decrease; after get off work, people will leave the workplace to go home or eat and shop. The nature of the educational and scientific research land is similar to that of the administrative office land. It mainly accommodates students, so its attraction rate is set to 1 during school hours and 0 after school. The population distribution of commercial and financial land during working hours is mainly employees, and there are employees and customers during the rest time period, so its attraction rate is set to 0.3 during the rest time period and 0.1 during the working time period, as shown in Table 2.

[0139] Table 2 Building attraction rate at different time periods within a day

[0140]

[0141]

[0142] For the population dynamics simulation within a week, the main research is on weekdays and weekends. Since the travel trajectories of residents are basically the same on weekdays, the basic travel trajectories of young people and students going to and from work on weekdays are "home-company / school-home"; while on weekends, their travel trajectories are generally "home-leisure / shopping-home". Therefore, the time period is mainly divided into weekdays and weekends, and the building attraction rate of each land use type at different times is shown in Table 3.

[0143] Table 3 Building attraction rate on weekdays and weekends

[0144] Land Type on Monday Tuesday Wednesday Thursday Friday Saturday Sunday Residential land 0.6 0.6 0.6 0.6 0.6 0.6 0.6 Administrative office space 1 1 1 1 1 0 0 Parks and green spaces 0 0 0 0 0 0.3 0.3 Education and scientific research land 1 1 1 1 1 0 0 Commercial and financial land 0.3 0.3 0.3 0.3 0.3 0.6 0.6 Medical and health land 0.8 0.8 0.8 0.8 0.8 1 1 Industrial land 0.8 0.8 0.8 0.8 0.8 0 0 arable land 0.3 0.3 0.3 0.3 0.3 0.2 0.2 Traffic land 0.3 0.3 0.3 0.3 0.3 0.3 0.3 Other land 0 0 0 0 0 0 0

[0145] S114. A mathematical model is established based on the building attraction rate in different time periods to simulate short-term population dynamics.

[0146] Taking a certain city as an example, according to the population dynamics simulation, the population is mainly distributed in urban areas. Figure 3 As shown in the figure, during the 7:00-8:00 am time period, the population is mainly concentrated in residential areas, and the population distribution in educational and scientific research land, administrative office land, etc. is relatively small. Figure 4 As shown in Figure 2, the population distribution in residential areas between 8:00 and 12:00 decreases, and the population begins to move to educational and scientific research land, administrative office land, and cultivated land. Figure 5 As shown in the figure, 12:00-14:00 is the lunch break time. During this time period, the number of people in the education and scientific research land and administrative office land decreases. Some people move to residential areas, and some people move to commercial areas. This is caused by differences in people's travel habits. Some people choose to eat nearby during lunch break, while others choose to go home to rest. Figure 6 As shown in the figure, 14:00-18:00 is the afternoon working time, and the population in residential areas decreases and moves to educational and scientific research land, administrative office land, etc. Figure 7 As shown in , 18:00-19:00 is the time when people get off work. At this time, the population in the residential area gradually increases, while the population in other areas gradually decreases. Among them, the population simulation results for the time period of 19:00-22:00 are as follows: Figure 8 As shown; the population simulation results for the 22:00-7:00 time period are as follows Fig. 9 As shown; the population simulation results on rest days are as follows Fig.10 As shown; the weekday population simulation results are as follows Fig.11 shown.

[0147] The central area of ​​the city is relatively flat, which is the urban area of ​​the city. The simulation results of the present invention show that the population is mainly distributed in the urban area and its surrounding areas, and the rest of the places are relatively scattered, and the distribution is relatively consistent with the actual situation.

[0148] S12, based on the refined population spatialization method of zoning and identifying real residential areas, a multivariate regression model is used to simulate the long-term population spatialization, such as Fig.12 shown.

[0149] S121. Calculate long-term population spatialization indicators.

[0150] The Human Settlement Index (HSI) refers to the proportion of human settlements within a unit pixel, reflecting the density of population distribution in space. Fig.13 As shown in the figure, population density has a high correlation with the average altitude of counties (cities, districts). Since most of China's population is distributed in low-altitude areas, it is necessary to correct the HSI for elevation to solve the problem of overestimation of the human settlement index in unlit areas, as shown below:

[0151]

[0152] Among them, EAHSI stands for Human Settlement Index, EVI max represents the annual maximum EVI, and DEM represents the elevation of the study area; Luojia nor Represents the standardized night light value, calculated as follows:

[0153]

[0154] Among them, Luojia min Indicates the minimum night light value, Luojia max Indicates the maximum night light value.

[0155] S122. Divide the study area into high economic zone and low economic zone for population spatial modeling.

[0156] (1) Establish a grid population estimation model for high economic zones.

[0157] High-economic and high-population density areas are concentrated in urban areas. The terrain is relatively flat and the population distribution is less affected by the terrain. Night light data can better show the spatial distribution of the population. Therefore, towns in high-economic and high-population density areas are used as samples, and night light data and building data are selected to establish a regression model to calculate the building area and night light coefficient, as shown below:

[0158] P i =a 1 ·B i +a 2 ·L i +b 1 (4)

[0159] Among them, P i is the population of the ith township; B i is the building area of ​​the ith township, L i is the night light brightness value of the i-th township; a 1 is the coefficient of building area, a 2 is the coefficient of night light brightness; constant term b 1 is the intercept. According to the principle of no buildings and no population, b 1 The value of is 0.

[0160] The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid population calculation model. The high economic zone grid population estimation model is as follows:

[0161] P ij =a 1 ·B ij +a 2 ·L ij (5)

[0162] Among them, P ij refers to the population of the ijth grid in the high economic zone, B ij is the building area of ​​the ijth grid, L ij is the night light brightness value of the ijth grid; a 1 is the building area (m 2 ), a 2 The coefficient of night light brightness value.

[0163] (2) Establish a grid population estimation model for low-economic zones.

[0164] The light brightness and intensity values ​​in low-economic zones are both small, and the population distribution in the zone is greatly affected by factors such as terrain and vegetation coverage. It is inaccurate to use night light data to estimate population distribution. Therefore, the building area index, road network density and modified human settlement index are selected as population spatialization indicators. Taking the towns in low-economic zones as samples, a population estimation model is established to calculate the coefficients of the building area index, road network density and modified human settlement index, as shown below:

[0165] P j =k 1 ·J j +k 2 ·RSD j +k 3 ·E j +b 2 (6)

[0166] Among them, P j is the population of the jth township; J j is the building density of the jth township, RSD j is the road network density of the jth township, E j is the modified human settlement index of the jth township; k 1 is the coefficient of the building area index, k 2 is the coefficient of road network density, k 3 is the coefficient of the modified human settlement index; the constant term b 2 is the intercept. According to the principle of no buildings and no population, b 2 The value of is 0.

[0167] The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid-scale population calculation model. The low-economic zone grid population estimation model is as follows:

[0168] P' ij =k 1 ·J ij +k 2 ·RSD ij +k 3 ·E ij (7)

[0169] Among them, P' ij is the population of the ijth grid in the low economic zone; J ij is the building density of the ijth grid, RSD ij is the road network density of the ijth grid, E ij is the revised human settlement index of the ijth grid; k 1 is the coefficient of the building area index, k 2is the coefficient of road network density, k 3 is the coefficient of the modified human settlement index.

[0170] S123. Model calibration and verification.

[0171] (1) Model calibration: Calculate the relative errors of population statistics and population spatialization in different towns as follows:

[0172]

[0173] Among them, RE i is the relative error of the ith township, POP m To simulate the population, POP s To count the population, based on the zoning modeling, in order to reduce the population spatialization error, the towns and grids with large population spatialization errors are recalculated, and the regression modeling is re-performed to obtain relevant parameters, and then the grid population estimation simulation is performed based on the population spatialization model.

[0174] (2) Error analysis: The absolute value of relative error and the average relative error are used to perform error analysis of population spatialization simulation, as shown below:

[0175]

[0176] Among them, MRE is the mean relative error, and N is the number of towns for population spatial simulation.

[0177] S124. Analysis of simulation results.

[0178] Taking a certain city as an example, the population spatialization results in 2020 are as follows: Fig.14 As shown in the figure, its population is mainly distributed in the south-central area, where the population is densely distributed, and the population in other areas is sparse. The area with a dense population distribution is the urban area of ​​​​a certain city. The terrain is flat and suitable for people to carry out various production activities; while other areas are mostly medium and low mountains or low mountains, which brings certain difficulties to people's production activities. Therefore, the population distribution in this area is sparse, and its spatialization results are in line with reality. According to the error analysis formula, the error in the low economic zone is larger, and the error in the high economic zone is smaller. The average relative error of the population spatialization results of a certain city in 2020 is 17.76%.

[0179] S2. Divide the geological hazard vulnerability assessment units.

[0180] The division of landslide vulnerability assessment units is an important step in landslide vulnerability zoning. Whether the division of assessment units is reasonable determines the accuracy and reliability of landslide vulnerability assessment. Currently, the commonly used assessment units are: regular grid units, geomorphic units and administrative units.

[0181] Regular grid unit: Divide the study area into regular grids of equal size, which are called grids in ArcGIS. The size of the grid determines the accuracy of the assessment. The division method is relatively flexible and simple, and data processing is relatively convenient.

[0182] Geomorphic unit: The study area is divided into different units according to the topography and landforms. Currently, the commonly used geomorphic units include slope units, watershed units, etc.

[0183] Administrative unit: The study area is divided according to the boundaries of villages and towns, with each administrative district as a unit. This assessment unit is convenient for managers to inspect landslide hazards, but it does not take into account the differences between different regions as a whole and has low accuracy.

[0184] Taking all vulnerability assessment factors into consideration, in order to facilitate data processing and obtain landslide vulnerability assessment results with higher accuracy, the present invention adopts a regular grid unit of 200m×200m as the assessment unit for the study.

[0185] S3. Calculate each vulnerability index, such as Fig.15 shown.

[0186] Vulnerability assessment indicators are classified into demographic vulnerability, economic vulnerability and material vulnerability; among them, demographic vulnerability includes population density; economic vulnerability includes GDP density and cultivated land density; material vulnerability includes building density and road density.

[0187] S31, population density;

[0188] Population density refers to the number of people per unit area. Population density is related to the vulnerability of the area. The higher the population density, the greater the vulnerability to landslide disasters. Population density is obtained based on the results of population dynamics simulation, as shown below:

[0189]

[0190] Where P m is the population density, P i It refers to the population of the ith grid, and A is the area of ​​the grid.

[0191] Taking a certain city as an example, the distribution of population vulnerability is as follows: Fig.16 , Fig.17 , Fig.18 , Fig.19 , Fig. 20 , Fig.21 , Fig. 22 , Fig.23 and Fig.24 As shown, the areas where the population gathers are different in different time periods, but they are mainly concentrated in the urban areas of a certain city.

[0192] S32, GDP density;

[0193] GDP density refers to the ratio of gross domestic product to area in a region, which characterizes the efficiency of economic activities and the density of land use per unit area of ​​a settlement. The greater the economic density, the higher its vulnerability to landslide disasters. The GDP density data of the study area was obtained by cutting the downloaded GDP density data.

[0194] Taking a certain city as an example, the GDP density distribution is as follows: Fig.25 As shown in the figure, the GDP density distribution shows a clustering trend, and the GDP density is higher in the central area of ​​a certain city. The area with the highest GDP density can reach 322.79 million yuan, which is mainly distributed in the urban areas of a certain city, and the main towns are A Street, B Street and C Street.

[0195] S33, cultivated land density;

[0196] The density of cultivated land refers to the ratio of cultivated land area in a region to the regional area, and is one of the important indicators for evaluating agricultural production efficiency. The greater the density of cultivated land, the greater the possibility of damage when a disaster occurs, the greater the crop loss caused, and the higher the degree of vulnerability. The cultivated land area is obtained according to the land use data, and the ratio of cultivated land area in the grid to the grid area is calculated as follows:

[0197]

[0198] Among them, G i represents the density of cultivated land in grid i, G ai represents the cultivated land area in grid i, and A represents the grid area.

[0199] Taking a certain city as an example, the occurrence of landslide disasters will cause serious damage to food crops planted on cultivated land, resulting in huge economic losses. Using the data management tools in ArcGIS Pro, a grid size of 200 meters is constructed to obtain the density of cultivated land in each grid. Then, the vector data is converted into raster data to obtain the distribution of cultivated land in a certain city, such as Fig.26 The cultivated land of a certain city is mainly distributed in the central area, and is also distributed on both sides, but its distribution is relatively sparse. The central area of ​​a certain city is relatively flat, which meets the conditions for crop planting. The two sides are mainly medium and low mountains and low mountains, so there is less cultivated land.

[0200] S34, building density;

[0201] Building density refers to the ratio of building area to regional area, and is an important indicator for assessing building losses. The greater the building density in the area, the greater the loss may be in the event of a geological disaster, as shown below:

[0202]

[0203] Among them, J i represents the building density within grid i, B i represents the building area in grid i, and A represents the grid area.

[0204] Taking a certain city as an example, the building density distribution is as follows: Fig. 27 As shown, the buildings of City XX are mainly concentrated in the south-central part. This area is the urban area of ​​City XX, and the main administrative divisions are A Street, B Street, C Street, and D Street.

[0205] S35, road density;

[0206] Road density refers to the ratio of road length to area in a region, reflecting the density of roads. The greater the road density, the greater the loss when a geological disaster occurs. Due to the different traffic capacities of roads of different grades and the different investment funds for the construction of roads of different grades, it is necessary to convert the lengths of different types of roads into standardized road lengths, as shown below:

[0207]

[0208] Among them, RSD i represents the road density within grid i, N ri 、N nei 、N pei 、N cri 、N tri They represent the lengths of railways, national roads, provincial roads, county roads and township roads in grid i respectively, and A represents the grid area.

[0209] Road construction requires a large investment, and landslide disasters will cause significant damage to roads. Therefore, road density is used as an evaluation indicator of material vulnerability. Taking a certain city as an example, the road density distribution is as follows: Fig.28 As shown in the figure, by constructing a 200-meter fishing net, the length of the road in each fishing net is obtained. Since different types of roads have different construction costs, the losses caused by disasters are different. The higher the level of the road, the greater the loss. Therefore, the lengths of roads of different levels need to be converted according to the standard.

[0210] S4. Construct a geological hazard vulnerability assessment model.

[0211] The evaluation model of geological hazard vulnerability is as follows:

[0212]

[0213] Among them, Y represents the value of geological hazard vulnerability, P i represents the value of the i-th indicator, W iRepresents the weight of the i-th indicator.

[0214] S5. Analysis of vulnerability assessment results at different time scales.

[0215] S51. Short-term vulnerability assessment results.

[0216] Taking a certain city as an example, each evaluation factor is standardized, and each evaluation factor is weighted according to the landslide vulnerability evaluation model. The vulnerability evaluation results of a certain city are obtained using the grid calculator. The vulnerability evaluation results of the first time period are divided into four levels using the discontinuity method, namely low vulnerability area, medium vulnerability area, high vulnerability area and extremely high vulnerability area. The remaining time periods are divided according to the classification standard of the first time period. The short-term vulnerability evaluation results are as follows: Fig.29 , Fig.30 , Fig.31 , Fig.32 , Fig.33 , Fig.34 , Fig.35 , Fig.36 and Fig.37 shown.

[0217] The vulnerability distribution of a certain city in different time periods is generally consistent. During the rest period, the higher vulnerability areas and high vulnerability areas are mainly distributed in the south-central area of ​​the city, and the main administrative divisions are A Street, B Street, C Street, and D Street. The medium vulnerability areas are mainly distributed along the roads. During working hours, the medium vulnerability areas and high vulnerability areas are mainly distributed in the south-central area, and the medium vulnerability areas are also mainly distributed along the roads. The distribution of extremely high vulnerability areas is relatively small.

[0218] S52. Long-term vulnerability assessment results.

[0219] Taking a certain city as an example, each evaluation factor is standardized, and then weights are assigned to each evaluation factor according to the landslide vulnerability evaluation model, and the raster calculator is used to obtain the evaluation results of the landslide vulnerability of a certain city in 2020. The vulnerability results are divided into low vulnerability areas, medium vulnerability areas, high vulnerability areas and extremely high vulnerability areas using the natural break point classification method.

[0220] like Fig.38 As shown, the medium-vulnerable areas, high-vulnerable areas and extremely-high-vulnerable areas for landslide disasters in XX City in 2020 are mainly distributed in the south-central area, surrounding the central urban area of ​​XX City. Some medium-vulnerable areas are mainly distributed along roads, and the main administrative divisions for the distribution of high-vulnerable areas and extremely-high-vulnerable areas are A Street, B Street, C Street and D Street.

[0221] S53. Comparison of results at different time scales.

[0222] According to the above research on the vulnerability of landslide disasters in a certain city at different time scales, it is found that the distribution of low vulnerability areas, medium vulnerability areas, high vulnerability areas and extremely high vulnerability areas is roughly the same at each time scale. High vulnerability areas and extremely high vulnerability areas are mainly distributed in the central area of ​​a certain city. This area is the urban area of ​​a certain city, with high population density, GDP density and building density. When a landslide disaster occurs, it is very likely to cause heavy losses. The medium vulnerability area is mainly distributed in the central part of a certain city, and some of the medium vulnerability areas are distributed along the roads.

[0223] Comparing the short-term vulnerability to landslide disasters with the long-term vulnerability to landslide disasters, it was found that the distribution of high vulnerability areas and extremely high vulnerability areas are relatively similar, both mainly distributed in the south-central region; during working hours, due to the wide distribution of population, its high vulnerability areas and extremely high vulnerability areas are somewhat different from those in 2020, but compared with the short-term vulnerability distribution of landslide disasters, the distribution of extremely high vulnerability areas in 2020 is relatively more concentrated, mainly because the short-term vulnerability to landslide disasters takes into account the flow of population, and the different distribution of population in different time periods will affect the distribution of its vulnerability; in 2020, the high vulnerability areas of landslide disasters in a certain city are also distributed on roads compared with the short-term high vulnerability areas.

[0224] Therefore, the present invention adopts the above-mentioned multi-time scale population spatialization geological disaster vulnerability assessment method to identify high-risk vulnerable areas, carry out various pre-disaster preparations in advance, improve disaster defense capabilities, optimize emergency resource allocation, fully consider the vulnerability of geological disasters, improve rescue efficiency, reduce economic and population losses, guide urban and rural planning and land use, fully consider the vulnerability degree of disasters, reduce the impact of disasters, and reduce the economic losses and casualties caused by disasters.

[0225] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A multi-time scale population spatialization geological hazard vulnerability assessment method, characterized in that: The following steps are involved: S1. Multi-time-scale spatialization of population distribution in the study area. The specific process is as follows: S11. Use the land use density method to simulate the short-term population dynamics of the population distribution in the study area; S111. Short-term population dynamics simulation, selecting land use type as an indicator, introducing building attraction rate, and obtaining the population spatial distribution in each time period, as shown below: Among them, P is the population, A is i is the area of ​​the i-th land use type, D i is the population density of the i-th land use type, k i is the attraction rate of the i-th land use type; S112. Define residents' travel with high temporal resolution based on time geography; S113, determining the building attraction rate of each land use type at different times; S114. Establish a mathematical model through the attraction rate of buildings in different time periods to simulate the short-term population dynamics; S12, based on the refined population spatialization method of zoning and identifying real residential areas, using a multivariate regression model to simulate long-term population spatialization; S121. Calculate the long-term population spatialization index as follows: Among them, EAHSI stands for Human Settlement Index, EVI max represents the annual maximum EVI, and DEM represents the elevation of the study area; Luojia nor Represents the standardized night light value, calculated as follows: Among them, Luojia min Indicates the minimum night light value, Luojia max Indicates the maximum night light value; S122. Divide the study area into high economic zone and low economic zone for population spatial modeling, including: (1) Establish a grid population estimation model for high economic zones; (2) Establish a grid population estimation model for low economic zones; S123. Model calibration and validation include: model calibration and error analysis of population spatialization simulation using absolute value of relative error and average relative error; S124, simulation result analysis; S2. Divide geological hazard vulnerability assessment units; S3. Calculate each vulnerability index; S4. Construct a geological hazard vulnerability assessment model as follows: Among them, Y represents the value of geological hazard vulnerability, P j represents the value of the jth vulnerability assessment index, W j represents the weight of the jth vulnerability assessment indicator; S5. Analysis of vulnerability assessment results at different time scales.

2. The multi-time scale population spatialization geological hazard vulnerability assessment method according to claim 1 is characterized in that: In step S122, a high economic zone grid population estimation model is established. The towns in the high economic zone are used as samples, night light data and building data are selected, and a regression model is established to calculate the building area and night light coefficient, as shown below: P x =a1·B x +a2·L x +b1 (4); Among them, P x is the population of the xth township; B x is the building area of ​​the xth township, L x is the night light brightness value of the x-th township; a1 is the coefficient of building area, a2 is the coefficient of night light brightness value; the constant term b1 is the intercept, according to the principle of no building no population, the value of b1 is 0; The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid population calculation model. The high economic zone grid population estimation model is as follows: P uv =a1·B uv +a2·L uv (5) Among them, P uv Refers to the population of the uvth grid in the high economic zone, B uv is the building area of ​​the uvth grid, L uv is the night light brightness value of the uvth grid; a1 is the coefficient of the building area, and a2 is the coefficient of the night light brightness value; A grid population estimation model for low economic zones was established. The towns in low economic zones were used as samples. The building area index, road network density and modified human settlement index were selected as population spatialization indicators. A population estimation model was established to calculate the coefficients of the building area index, road network density and modified human settlement index, as shown below: P x =k1·J x +k2·RSD x +k3·E x +b2 (6); Among them, P x is the population of the xth township; J x is the building density of the xth township, RSD x is the road network density of the xth township, E x is the revised human settlement index of the xth township; k1 is the coefficient of the building area index, k2 is the coefficient of the road network density, and k3 is the coefficient of the revised human settlement index; the constant term b2 is the intercept, and according to the principle of no building, no population, the value of b2 is 0; The grid population has the same model as the township population. Substituting the calculated coefficients into the model, we can obtain a 200m×200m grid-scale population calculation model. The low-economic zone grid population estimation model is as follows: P′ uv =k1·J up +k2·RSD uv +k3·E uv (7); Among them, P' uv is the population of the uvth grid in the low economic zone; J uv is the building density of the uvth grid, RSD uv is the road network density of the uvth grid, E uv is the corrected human settlement index of the uvth grid; k1 is the coefficient of the building area index, k2 is the coefficient of the road network density, and k3 is the coefficient of the corrected human settlement index.

3. The multi-time scale population spatialization geological hazard vulnerability assessment method according to claim 1 is characterized in that: In step S123, the model calibration and verification specifically includes: (1) Model calibration: Calculate the relative errors of population statistics and population spatialization in different towns as follows: Among them, RE x is the relative error of the xth township, POP m To simulate the population, POP s To count the population; (2) Error analysis: The absolute value of relative error and the average relative error are used to perform error analysis of population spatialization simulation, as shown below: Among them, MRE is the mean relative error, RE x is the relative error of the xth township.

4. The multi-time scale population spatialization geological hazard vulnerability assessment method according to claim 1 is characterized by: In step S2, regular grid units of 200m×200m are used to divide the geological hazard vulnerability assessment units.

5. The multi-time scale population spatialization geological hazard vulnerability assessment method according to claim 1 is characterized by: In step S3, the vulnerability assessment index is graded into population vulnerability, economic vulnerability and material vulnerability; wherein population vulnerability includes population density; economic vulnerability includes GDP density and cultivated land density; material vulnerability includes building density and road density.

6. The multi-time-scale population spatialization geological hazard vulnerability assessment method according to claim 5 is characterized in that: The calculation method of vulnerability assessment index is as follows: S31. Population density is obtained based on the results of population dynamics simulation as follows: Where P m is the population density, P uv is the population of the uvth grid, and A is the grid area; S32. GDP density refers to the ratio of regional gross domestic product to area, which represents the efficiency of economic activities and the density of land use per unit area of ​​the settlement. The greater the economic density, the higher the vulnerability after a landslide disaster. S33, cultivated land density is obtained by obtaining the ratio of cultivated land area to grid area according to land use data, as shown below: Among them, G uv Represents the density of cultivated land within the grid uv, G auv represents the cultivated land area within the grid uv, and A represents the grid area; S34. Building density is obtained based on the ratio of building area to area within the region, as shown below: Among them, J uv Indicates the building density within the grid uv, B uv represents the building area within the grid uv, and A represents the grid area; S35. Road density is obtained based on the ratio of road length to area in the region, as shown below: Among them, RSD uv Represents the road density within the grid uv, N ruv 、N neuv 、N peuv 、N cruv 、N truv They represent the lengths of railways, national roads, provincial roads, county roads and township roads within the grid uv respectively, and A represents the grid area.

Citation Information

Patent Citations

  • Urban flood disaster dynamic assessment method based on multi-source data

    CN112819340A

  • Flood disaster loss assessment method based on spatial information grid

    CN115310806A