A High-Precision Risk Assessment Method for Population and Economic Losses from Rainstorms Based on Downscaling in Complex Terrain

Abstract

This invention discloses a high-precision method for assessing population and economic losses from rainstorms based on downscaling of complex terrain. The method involves acquiring hourly precipitation observation data and calculating a climate field by overlaying topographic covariate factors. The differences between all hourly precipitation data and the climate field are calculated, and an anomaly field is created using a distance-based interpolation method. The climate field and the anomaly field are then overlaid, and the overlay result is corrected for discrepancies with the original hourly precipitation observation data to obtain hourly precipitation interpolation results. Rainstorm disaster events are identified based on the interpolation results. A multi-dimensional risk assessment method is used to calculate a rainstorm disaster risk index. This risk index is combined with population or economic data to calculate the final population and economic loss assessment result. This invention provides highly refined assessment results and has strong practical applicability.

Description

Technical Field

[0001] This invention relates to the field of rainstorm disaster loss risk assessment technology, specifically to a high-precision rainstorm population and economic loss risk assessment method based on downscaling of complex terrain. Background Technology

[0002] With the increasing severity of climate change and the growing frequency of rainstorm events, assessing the risk of rainstorm damage is imperative. Rainstorm disasters are characterized by long duration and sudden onset. Precipitation, population, and economy are the most important assessment criteria for rainstorm disaster losses. Complex terrain further amplifies the losses caused by rainstorm disasters and complicates post-disaster relief efforts. Current assessment techniques face the following challenges: ground-based meteorological stations cannot achieve full-area coverage of precipitation monitoring; there is a lack of precipitation monitoring data at different altitudes and with high spatial resolution, failing to meet the requirements for rainstorm disaster assessment and resulting in often low accuracy in current assessments; effective precipitation estimation methods are lacking for areas with limited observational data; and precipitation is affected by local altitude, making rainstorm disaster identification particularly difficult in complex terrain areas such as plateaus and mountains. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide a high-precision method for assessing the economic loss risk of rainstorm populations based on downscaling of complex terrain. On the one hand, it solves the problem that the lack of ground stations or insufficient accuracy of observation data makes it impossible to assess rainstorm disaster losses on a large scale and with high precision. On the other hand, it solves the problem that rainstorm disaster identification is difficult due to complex terrain and large changes in altitude.

[0004] Technical solution: The present invention provides a high-precision method for assessing the risk of economic loss due to rainstorms based on downscaling of complex terrain, comprising the following steps:

[0005] Step 1: Obtain hourly precipitation observation data and calculate the climate field by overlaying topographic covariate factors;

[0006] Step 2: Calculate the difference between all hourly precipitation data and the climate field, and create an anomaly field using a distance-based interpolation method;

[0007] Step 3: Overlay the climate field with the anomaly field, and correct the discrepancy between the overlay result and the original hourly precipitation observation data to obtain the hourly precipitation interpolation result;

[0008] Step 4: Identify rainstorm disaster events based on interpolation results; calculate the rainstorm disaster risk index using a multi-dimensional risk assessment method; combine the risk index with population or economic data to calculate the final population and economic loss assessment result.

[0009] Furthermore, in step 1, the topographic covariate factors include elevation, relative elevation difference, and proximity to water sources, which are used to reflect the influence of complex topography on the spatial distribution of precipitation in the climate field calculation.

[0010] Furthermore, in step 2, the anomaly field is generated using an interpolation method based on station distance. This method involves selecting observations from several known stations around the estimated point and combining them with distance weights to perform calculations, thereby achieving spatial interpolation of precipitation anomalies.

[0011] Furthermore, in step 3, the deviation correction uses the equidistant cumulative distribution function to correct the superposition results. By comparing the distribution differences between the observation data during the training period and the output results, and the output results during the correction period, the precipitation interpolation results are systematically corrected.

[0012] Furthermore, in step 4, the identification of rainstorm disaster events is based on the precipitation threshold standard within a specified time period, distinguishing between rainstorm, heavy rainstorm, and extremely heavy rainstorm levels.

[0013] Furthermore, in step 4, the multi-dimensional risk assessment method constructs the rainstorm disaster risk index as the product of hazard, exposure, and vulnerability. Hazard is derived by comprehensively considering the frequency of rainstorm disasters, cumulative precipitation, and duration. Exposure is constructed separately for population and economic losses, both incorporating elevation, relative elevation difference, and proximity to water sources, and introducing population density or economic intensity factors respectively. Vulnerability is derived by comprehensively considering the density of flood control dams and the density of emergency rescue points.

[0014] Furthermore, the frequency, cumulative precipitation, duration, population density, economic intensity, elevation, relative elevation difference, proximity to water sources, density of flood control dams, and density of emergency rescue points are all converted into indices with unified dimensions through data normalization methods before being included in the calculation.

[0015] Furthermore, in step 4, for the population exposure assessment, a population density factor is introduced based on the integration of elevation, relative elevation difference, and water source proximity factors; for the economic exposure assessment, an economic intensity factor is introduced based on the same topographic and water source factors.

[0016] Furthermore, in step 4, vulnerability is calculated by subtracting the product of the flood control dam density index and the emergency rescue point density index from a set baseline value. The higher the vulnerability value, the weaker the region's disaster prevention and mitigation capabilities.

[0017] Furthermore, in step 4, the final population and economic loss result is obtained by overlaying the independent risk indices of population and economy with the corresponding population or economic data, thereby achieving a refined separate assessment of the population and economic losses caused by rainstorm disasters.

[0018] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: By introducing topographic covariate factors such as elevation, relative elevation difference, and proximity to water sources into the climate field calculation, and combining thin-plate spline interpolation with equidistant cumulative distribution function deviation correction, this invention effectively solves the problems of sparse ground meteorological stations and significant altitude-dependent spatial distribution of precipitation in complex terrain areas. It achieves high-precision downscaling interpolation of hourly precipitation, providing a reliable precipitation data foundation for rainstorm disaster identification. This invention constructs a three-dimensional multiplicative risk index system of "hazard—exposure—vulnerability." Hazard integrates the frequency, cumulative precipitation, and duration of rainstorms; exposure is constructed separately for population and economy, incorporating topographic and water source factors; vulnerability integrates the density of flood control dams and the density of emergency rescue points. All indicators are normalized to a unified dimension, with a reasonable dimensional design and clear assessment logic, achieving a multi-dimensional systematic assessment of rainstorm disaster risk. This invention achieves separate assessment of population and economic risks, superimposing independent risk indices with corresponding population or economic data to obtain refined spatial assessment results. Meanwhile, all input data are publicly available observation and statistical data with clear data sources and complete methods and procedures, possessing strong operability and practical application value, and can provide a scientific basis for decision-making on rainstorm disaster prevention and mitigation in complex terrain areas. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0021] like Figure 1 As shown in the figure, this invention provides a high-precision method for assessing the risk of economic loss due to rainstorms based on downscaling of complex terrain, including the following steps:

[0022] (1) The average climate field is calculated by superimposing topographic covariate factors on the hourly precipitation observation data to obtain the climate field;

[0023] (2) Calculate the difference between all hourly precipitation data and the climate field to obtain the outlier interpolation results, which is to create the anomaly field;

[0024] (3) Interpolation results of superimposed climate field and anomaly field;

[0025] (4) Correct the deviation between the result obtained in step (3) and the hourly precipitation observation data to obtain the final result;

[0026] (5) Identify rainstorm disasters based on hourly precipitation interpolation results;

[0027] (6) Assess the risk index of rainstorm disasters using a multi-dimensional risk assessment method;

[0028] (7) Calculate population and economic losses by combining the rainstorm disaster index.

[0029] The calculation method is as follows:

[0030] (1) The calculation method for the superimposed topographic covariate factor climate field is as follows:

[0031] (1)

[0032] In the formula, It is the dependent variable concerning spatial location i; f( is an independent independent variable at spatial location i, dimension d). ) is about The positional smoothing function is estimated using the least squares method; Independent covariates at position p; For about The p-latitude coefficient; Let the random error be the independent variable with an expected value of 0. Let be the known local relative coefficient of variation used as the weight. The unknown error covariance; N is the number of spatial grid points to be interpolated, f( The least squares method is used for estimation, and the formula is as follows:

[0033] (2)

[0034] In the formula, The positive smoothness parameter is determined by minimizing the generalized cross-validation. f( The m-th order partial derivative of ). Let f(m) be the known local relative coefficient of variation used as the weight. In this embodiment, m=2, that is, the square integral of the second-order partial derivative is used as the smoothing penalty term. In this embodiment, f( The thin plate spline function is used, with smoothing parameters... Generalized cross-validation (GCV) is used for automatic selection, with a search range of log. 10 (ρ)∈[-5,5], with a search step size of 0.2, the search term is the one with the smallest GCV score. The value is the optimal smoothing parameter.

[0035] (2) The outlier interpolation method is as follows:

[0036] (3)

[0037] In the formula, for Estimated value; for The observed value; n is the number of surrounding stations used in the estimation, generally the 12 closest surrounding stations are selected; For the estimated point and each known sample point The distance to see; The distance is the degree of the distance, which is usually 2.

[0038] (3) The method for correcting the deviation is as follows:

[0039] (4) (5)

[0040] In the formula, Input data for climate element variables, The results are the corrections for climate element variables. It is the cumulative distribution function with equal intervals. This is the inverse operation of the interval cumulative distribution function. For observational data during training, The output during training. The output results during the correction period. In this embodiment, the training period and correction period involved in the correction of the equidistant cumulative distribution function deviation are divided according to the following rules: the training period is used to establish the statistical relationship between the observed data and the output data, and the correction period is the target period to be corrected. The two do not overlap in time and the training period is earlier than the correction period. According to the availability of data and actual needs, the training period can be determined in one of the following ways: (1) Fixed-duration training period - take N consecutive years (recommended N=5) or N consecutive days (recommended N=30 days) before the correction period as the training period; (2) Sliding training period - slide forward period by period with a fixed-length time window (recommended 60 days), and each correction period corresponds to a dynamically updated training period; (3) Mixed training period - composed of recent data (recommended 30 days) before the correction period and historical data of the same period (recommended 15 days before and after the same day of the previous year); (4) Proportional division - divide the entire historical data into training period (recommended first 60%) and correction period (recommended last 40%) in chronological order. The optimal training period length can be determined by comparing the correction effects (such as root mean square error, correlation coefficient, etc.) at different lengths. Equal-interval cumulative distribution function The formula is as follows:

[0041] (6)

[0042] In the formula, For all less than or equal to The sum of the probabilities of the numerical values ​​occurring.

[0043] (4) The criteria for identifying rainstorm disasters are as follows (Table 1): In this embodiment, the identification of rainstorm disaster events is carried out in accordance with the national standard "Precipitation Grades" (GB / T 28592-2012).

[0044] Table 1. Criteria for Identifying Rainstorm Disasters

[0045]

[0046] (5) The multi-dimensional risk index assessment method for rainstorm disasters is as follows: (Flood control dam data comes from the China Surface Water Bodies, Dams, Reservoirs and Lakes Data Set released by the National Earth System Science Data Center; Emergency rescue point data refers to the Emergency Shelter Information Table released by the Emergency Management Bureaus of various provinces and cities; In this embodiment, water system or water source point data comes from the water system layer (including line and surface elements such as rivers, lakes and reservoirs) in the 1:250,000 National Basic Geographic Database provided by the National Geographic Information Center, or from the China 1:250,000 Third-Level Watershed Division Water System Distribution Data Set released by the National Earth System Science Data Center. After data acquisition, the data is uniformly...) Converted to the WGS84 geographic coordinate system, the vector data was rasterized and matched with the evaluation grid. The Euclidean distance from each grid point to the nearest water system element was calculated as the water source proximity index. The emergency rescue point data came from the emergency shelter information released by the emergency management bureaus of various provinces and cities, the national emergency shelter comprehensive management and service system, or the public data open platform of various provinces (such as the "provincial shelter" dataset released by the Shandong Public Data Open Network). After acquisition, the name, type and spatial location information of each emergency rescue point were extracted. The number of emergency rescue points was counted by evaluation grid or administrative division, the emergency rescue point density (points / km²) was calculated, and normalized to the emergency rescue point density index according to formula (12). ).

[0047] (7)

[0048] In the formula, This refers to a risk index for rainstorm disasters. The higher the index, the greater the degree to which a certain area is affected by rainstorm disasters. Represents the risk of rainstorm disasters. This represents the degree to which a local population or economy is exposed to rainstorm disasters. It represents the vulnerability of a local population or economy to rainstorm disasters; the greater the vulnerability, the weaker the population or economy's disaster prevention capabilities. , and The calculation method is as follows:

[0049] (8)

[0050] In the formula, An index representing the frequency of rainstorm disasters in a given year. The cumulative precipitation index representing the amount of rainfall that caused a rainstorm disaster in a given year. An index representing the duration of rainstorm disasters.

[0051] The calculation method is as follows:

[0052] Targeting the population:

[0053] (9)

[0054] In the formula, Represents the exposure level to rainstorm disasters in a given year. Represents population index, Represents the elevation index. Represents the relative height difference index. This represents the proximity index to the water source; the higher the index, the closer the location is to the water source.

[0055] Regarding the economy:

[0056] (10)

[0057] In the formula, Represents the exposure level to rainstorm disasters in a given year. Represents an economic index. Represents the elevation index. Represents the relative height difference index. This represents the proximity index to the water source; the higher the index, the closer the location is to the water source.

[0058] The calculation method is as follows:

[0059] (11)

[0060] In the formula, Represents the vulnerability to rainstorm disasters in a given year. Represents the density index of flood control dams. This represents the density index of emergency rescue points.

[0061] , , , , , , , , and The data were analyzed by frequency of rainstorm disasters (times / year), cumulative rainfall from rainstorm disasters (mm / year), and population density (people / km²). 2GDP per unit area (GDP / km²) 2 ), Elevation (m), Relative elevation difference (m), Proximity to water source, Flood control dam density (units / km) 2 ) and density of emergency rescue points (number / km) 2 The calculation is performed using a data normalization method, which is as follows:

[0062] (12)

[0063] In the formula, The index representing the j-th indicator item in the i-th region. ∈[0,1], The original data representing the j-th indicator for the i-th region. This represents the minimum value of the j-th indicator. This represents the maximum value of the j-th indicator.

[0064] relative elevation difference Proximity to water source The calculation method is as follows:

[0065]

[0066] (13)

[0067] In the formula, This represents the maximum elevation within a given grid point. This represents the minimum elevation within the same grid point. Let be the distance from the i-th grid point to the nearest water system or water source. Calculate the distances from all grid points to the nearest water system or water source, where the maximum value is... .

[0068] (7) The final loss result is obtained by multiplying the multidimensional risk index of the corresponding population or economy with the population or economy.

Claims

1. A high-precision method for assessing the risk of economic loss due to rainstorms based on downscaling in complex terrain, characterized in that, Includes the following steps: Step 1: Obtain hourly precipitation observation data and calculate the climate field by overlaying topographic covariate factors; Step 2: Calculate the difference between all hourly precipitation data and the climate field, and create an anomaly field using a distance-based interpolation method; Step 3: Overlay the climate field with the anomaly field, and correct the discrepancy between the overlay result and the original hourly precipitation observation data to obtain the hourly precipitation interpolation result; Step 4: Identify rainstorm disaster events based on interpolation results; calculate the rainstorm disaster risk index using a multi-dimensional risk assessment method; combine the risk index with population or economic data to calculate the final population and economic loss assessment result.

2. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 1, the topographic covariate factors include elevation, relative elevation difference, and proximity to water sources, which are used to reflect the influence of complex topography on the spatial distribution of precipitation in climate field calculations.

3. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 2, the anomaly field is generated using an interpolation method based on station distance. This method involves selecting observations from several known stations around the estimated point and combining them with distance weights to perform calculations, thereby achieving spatial interpolation of precipitation anomalies.

4. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 3, the deviation correction uses the equidistant cumulative distribution function to correct the superposition results. By comparing the distribution differences between the observation data during the training period and the output results, and the output results during the correction period, the precipitation interpolation results are systematically corrected.

5. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 4, the identification of rainstorm disaster events is based on the precipitation threshold standard within a specified time period, distinguishing between rainstorm, heavy rainstorm, and extremely heavy rainstorm levels.

6. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 4, the multi-dimensional risk assessment method constructs the rainstorm disaster risk index as the product of hazard, exposure, and vulnerability. Hazard is derived by combining the frequency of rainstorm disasters, cumulative precipitation, and duration. Exposure is constructed separately for population and economic losses, both incorporating elevation, relative elevation difference, and proximity to water sources, and introducing population density or economic intensity factors respectively. Vulnerability is derived by combining the density of flood control dams and the density of emergency rescue points.

7. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 6, is characterized in that... The frequency, cumulative precipitation, duration, population density, economic intensity, elevation, relative elevation difference, proximity to water sources, density of flood control dams, and density of emergency rescue points are all converted into indices with uniform dimensions through data normalization methods before being used in the calculation.

8. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 7, is characterized in that... In step 4, for the population exposure assessment, a population density factor is introduced based on the integration of elevation, relative elevation difference and water source proximity factors; for the economic exposure assessment, an economic intensity factor is introduced based on the same topographic and water source factors.

9. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 6, is characterized in that... In step 4, vulnerability is calculated by subtracting the product of the flood control dam density index and the emergency rescue point density index from the set baseline value. The higher the vulnerability value, the weaker the regional disaster prevention and mitigation capabilities.

10. The method for high-precision assessment of population and economic loss risk from rainstorms based on downscaling of complex terrain, as described in claim 1, is characterized in that... In step 4, the final population and economic loss results are obtained by overlaying the independent risk indices of population and economy with the corresponding population or economic data, thereby achieving a refined separate assessment of the population and economic losses caused by rainstorm disasters.

Images