Flood risk assessment method and flood risk assessment system

Abstract

To provide a flood disaster risk evaluation method and a flood disaster risk evaluation system that enable a flood disaster risk to be evaluated quantitatively.SOLUTION: In order to quantitatively evaluate a flood disaster risk in an evaluation target area, a computer derives a relational expression by a logistic regression analysis using, as an objective variable, an occurrence probability of a flood disaster based on past data indicating the presence / absence of a flood disaster in the past in a certain area including a place where a flood disaster occurred in the past and using a predetermined index indicating information in the certain area as an explanatory variable, thereby evaluating the flood disaster risk of the evaluation target area.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a flood hazard degree evaluation method and a flood hazard degree evaluation system.

Background Art

[0002] In recent years, due to concentrated heavy rain in urban areas, many human and material damages have occurred, and the need for flood hazard degree evaluation has been increasing. In response, local governments and the like have independently created and published flood hazard maps. On the other hand, companies having bases (such as factories and branch offices) scattered throughout the country sometimes compare the hazard degrees in order to determine the priorities when implementing flood countermeasures. However, since the conditions such as damage assumptions are different for each local government, it has been necessary to individually investigate the hazard degrees, which has been time-consuming. Therefore, methods for evaluating the hazard degrees with the same index in different regions have been proposed. For example, in Non-Patent Document 1, in order to simply evaluate the hazard degree of the land itself against floods (inland water flooding and external water flooding) with the same national index, a method for evaluating the flood hazard degree by using the Laplacian, which is often used as an index of terrain undulation, is disclosed.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the risk assessment method described in Patent Document 1, the elevation data obtained by dividing the area to be evaluated into a predetermined size mesh (with sides of 250m) was used to calculate the unevenness of the ground. Based on the calculated unevenness of the ground, a map showing the area to be evaluated was colored with colors corresponding to the level of flood risk to indicate the risk of flood damage. For example, areas with a high risk of flood damage were colored darker, while areas with a low risk of flood damage were not colored. A problem with this conventional assessment method was that the threshold for differentiating the colors indicating flood risk had no scientific meaning, so it could only provide a qualitative assessment.

[0005] This invention has been made in view of the above problems, and its objective is to provide a flood risk assessment method and a flood risk assessment system that can quantitatively evaluate the degree of flood risk. [Means for solving the problem]

[0006] The main invention for achieving the above objective is a method for evaluating flood risk, in which a computer derives a relational expression by logistic regression analysis, with the probability of flood occurrence based on past data indicating the presence or absence of past floods in a certain area as the dependent variable, and a predetermined index of the said area as the independent variable, wherein the past data is data indicating the presence or absence of past floods in each of a plurality of sample divisions obtained by dividing a sample area including locations where floods have occurred in the past into predetermined mesh sizes, and the predetermined index is information for each of the said sample divisions. As an indicator that serves as an explanatory variable in logistic regression analysis This is a flood risk assessment method characterized by including the Laplacian. Other features of the present invention will be made clearer by description in this specification and the accompanying drawings. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a flood risk assessment method and a flood risk assessment system that can quantitatively evaluate the degree of flood risk. [Brief explanation of the drawing]

[0008] [Figure 1]This is a block diagram of the flood risk assessment system. [Figure 2] This figure shows the evaluation area and the 50m and 250m size meshes. [Figure 3] This is a flowchart explaining the method for evaluating the probability of flooding. [Figure 4] This is a diagram to explain past data. [Figure 5] This figure shows the sample area and the 50m and 250m size meshes. [Figure 6] This diagram illustrates the sample segments included within the intermediate sample segment. [Figure 7] This is a diagram showing the database associated with the sample categories. [Figure 8] This figure shows the regression variables obtained by logistic regression analysis and the evaluation results of those regression variables. [Figure 9] This diagram illustrates the difference between simple regression analysis and logistic regression analysis. [Figure 10] This diagram illustrates the influence of adjacent evaluation categories. [Modes for carrying out the invention]

[0009] This specification and the accompanying drawings make it clear at least the following:

[0010] This flood risk assessment method is characterized by a computer deriving a relational equation through logistic regression analysis, using the probability of flood occurrence based on past data indicating the presence or absence of past floods in a certain area as the dependent variable, and a predetermined indicator of the said area as the independent variable.

[0011] This flood risk assessment method allows for the deriving of a relational equation through logistic regression analysis, where the probability of flood occurrence based on past data is the dependent variable and predetermined indicators are the independent variables. This makes it possible to perform a quantitative assessment of the area for which flood risk assessment is desired.

[0012] Such a flood hazard degree evaluation method, wherein the past data is data indicating the presence or absence of past floods in each of a plurality of sample sections obtained by dividing a sample area including locations where floods occurred in the past into a predetermined mesh size, The predetermined index is characterized in that it is information for each of the sample sections.

[0013] According to such a flood hazard degree evaluation method, a relational expression is derived by logistic regression analysis using, as an objective variable, the probability of flood occurrence based on past data indicating the presence or absence of past floods in a plurality of sample sections, and using, as explanatory variables, predetermined indices in the sample sections. Therefore, it becomes possible to obtain the probability of flood occurrence for each mesh size identical to the sample sections.

[0014] Such a flood hazard degree evaluation method, wherein a computer calculates the probability of flood occurrence in each of the evaluation target sections based on the relational expression, and the predetermined indices indicating the information of each of a plurality of evaluation target sections obtained by dividing the evaluation target area into the predetermined mesh size.

[0015] According to such a flood hazard degree evaluation method, the probability of flood occurrence in each of the evaluation target sections is calculated based on the same indices as those of the sample sections, and the relational expression derived by logistic regression analysis, in a plurality of evaluation target sections obtained by dividing the evaluation target area into the same mesh size as the sample sections. Therefore, it becomes possible to calculate a more accurate probability of flood occurrence based on past cases for each evaluation target section.

[0016] Such a flood hazard degree evaluation method, wherein the predetermined index includes at least one of elevation, Laplacian, slope, geology, and ground conditions in the sample section or the evaluation target section.

[0017] According to such a flood hazard degree evaluation method, since at least one of the elevation, Laplacian, slope, geology, and ground conditions that can be factors of floods is included in the index, it becomes possible to calculate a more accurate flood occurrence probability.

[0018] Such a flood hazard degree evaluation method, wherein the predetermined index includes the predetermined index of the sample section adjacent to the target sample section and the predetermined index of the evaluation target section adjacent to the evaluation target section.

[0019] According to such a flood hazard degree evaluation method, since the target sample section is connected to the adjacent sample section, it becomes possible to derive a relational expression based on the information including the influence of the adjacent sample section. Therefore, it becomes possible to derive a relational expression capable of calculating a more accurate flood occurrence probability. Also, in the calculation according to this relational expression, since the index of the adjacent evaluation target section connected to the evaluation target section to be evaluated is included, it becomes possible to calculate the flood occurrence probability including the influence of the adjacent evaluation target section. Therefore, it becomes possible to calculate a more accurate flood occurrence probability.

[0020] Such a flood hazard degree evaluation method, wherein the occurrence probability of the flood is the occurrence probability of inundation.

[0021] According to such a flood hazard degree evaluation method, it becomes possible to quantitatively evaluate the occurrence probability of inundation.

[0022] Also, a storage unit that stores past data indicating the presence or absence of past floods in a certain area and a predetermined index of the certain area; An arithmetic processing unit that derives a relational expression between the past data and the predetermined index by logistic regression analysis, with the occurrence probability of floods based on the past data as the objective variable and the predetermined index as the explanatory variable; A flood hazard degree evaluation system characterized by comprising the above.

[0023] With this type of flood risk assessment system, a relational equation is derived through logistic regression analysis using the probability of flood occurrence based on past data as the dependent variable and predetermined indicators as independent variables. This makes it possible to provide a flood risk assessment system that can perform quantitative evaluations for areas where flood risk assessment is desired.

[0024] ===Implementation Method=== The flood risk assessment method and flood risk assessment system according to this embodiment will be explained with reference to the figures.

[0025] In this embodiment, the evaluation of the flood risk in a certain area (evaluation target area) will be explained using a method that evaluates the risk by calculating the probability of flooding as an example. In this embodiment, a certain evaluation target area is divided into meshes of a predetermined size, and the probability of flooding occurring in each of the divided sections (evaluation target sections) is calculated.

[0026] The probability of flooding in each evaluation target category is calculated by deriving a relational equation through logistic regression analysis, with the probability of flooding as the dependent variable and several indicators assumed to be factors of flooding as independent variables. This calculation is performed by a known information processing system 5, such as a computer, equipped with a storage unit 1, an arithmetic processing unit 2, an input unit 3, a display unit 4, etc., as shown in Figure 1. The storage unit 1 can store past data and explanatory variables that serve as predetermined indicators, which will be described later, by inputting them through the input unit 3. It also stores a program that performs the process of deriving a relational equation through logistic regression analysis based on the stored data and calculating the probability of flooding based on the derived relational equation and predetermined indicators for each evaluation target category. In other words, the information processing system 5 corresponds to the flood risk assessment system.

[0027] In this embodiment, the probability of flooding p is calculated for each evaluation target section R1a to R12y, which is obtained by dividing the evaluation target area RA into 50m meshes as shown in Figure 2. In Figure 2, the area enclosed by the frame is the evaluation target area RA, and the evaluation target area RA is divided by a 50m size mesh.

[0028] In Figure 2, the evaluation area RA is divided into evaluation intermediate sections R1 to R12 using 250m-sized meshes, and each of these intermediate sections R1 to R12 is further divided into evaluation sections R1a to R12y using 50m-sized meshes. The flooding probability p is calculated for each evaluation section R1a to R12y. Note that in Figure 2, the display of the 50m-sized meshes is partially omitted, but the entire evaluation area RA is divided into 50m meshes.

[0029] The calculation of the flood probability p is performed as shown in Figure 3. First, historical data is collected from a map of a certain area (sample area) SA that includes locations where flooding has occurred in the past, as shown in Figure 4, with the locations of flooding Q extracted and shown, along with data representing multiple indicators that are assumed to be factors in flooding (Step 1). The collected data is then compiled into a database and stored in the storage unit 1 (Step 2). From the stored database, each regression variable β in the logistic regression analysis equation shown in (Equation 1) is determined (Step 3). The determined regression variables β are substituted into (Equation 1) to derive a relational equation. Based on the derived relational equation and the data representing indicators of the area to be evaluated RA, the flood probability p for the area to be evaluated is calculated, and the flood risk is assessed (Step 4). This is described in detail below. In deriving the relationship here, the dependent variable, the probability of flooding p, is set to "1" representing a 100% flooding rate when flooding has occurred in the past, and to "0" representing a 0% flooding rate when no flooding has occurred in the past, based on past flooding records. In Figure 4, the flooded locations Q that have been flooded in the past are shaded.

[0030] log(p / (1-p)) = β0 + β1 * explanatory variable 1 + β2 * explanatory variable 2 + ... ∴p=1 / [1+exp{-(β0+β1*explanatory variable 1+β2*explanatory variable 2···)}] ...(Formula 1) p: Probability of flooding

[0031] The historical data is data associated with maps of existing areas (sample areas) SA that show locations where flooding has occurred in the past. For example, land history survey data published nationwide by the Ministry of Land, Infrastructure, Transport and Tourism (Land Classification Basic Survey (Land History Survey) by the Land Information Division, National Land Policy Bureau, Ministry of Land, Infrastructure, Transport and Tourism https: / / nlftp.mlit.go.jp / kokjo / inspect / landclassification / land / land_history_2011 / pdf_landform_03.html), and the Tokyo Metropolitan Government Bureau of Construction's "Past Flood Records ~Flood Performance Map~" https: / / www.kensetsu.metro.tokyo.lg.jp / jigyo / river / suishin / suigai_kiroku / kako.html can be used. As shown in Figure 5, the sample areas SA are divided into 50m size meshes, similar to the evaluation area RA.

[0032] In Figure 5, as with the evaluation area RA in Figure 3, the sample is divided into intermediate sections S1 to S12 using a 250m mesh, and each of these intermediate sections S1 to S12 is further divided into sample sections S1a to S12y using a 50m mesh.

[0033] The following explanation will primarily use the 25 sample divisions S1a to S1y included in the intermediate sample division S1 as examples. First, the presence or absence of past flooding in each sample division S1a to S1y is examined from past data, and the flooding probability p associated with each sample division S1a to S1y is stored in the memory unit 1.

[0034] For example, as shown in Figure 6, sample section S1a does not include any areas that have been flooded in the past, so the flooding probability p for sample section S1a is set to "0" and stored in the storage unit 1 in association with the location information of sample section S1a. Also, sample section S1d occupies a larger area where flooding has occurred in the past. For this reason, the flooding probability p for sample section S1d is set to "1" and stored in the storage unit 1 in association with the location information of sample section S1d.

[0035] The probability of flooding occurring p in each sample category S1a to S12y is divided into "1" (flooding occurred) and "0" (no flooding occurred) based on the proportion of the area occupied by areas that have been flooded in the past relative to the total area of ​​a single sample category. The threshold is set appropriately based on the proportion of the area occupied by flooded areas, such as 30%, 50%, or 80%. In this case, setting a smaller threshold will result in a relationship formula that calculates a more conservative probability of flooding. Here, a proportion of the area occupied by flooded areas of 50% is considered "1" (flooding occurred), and anything less than 50% is considered "0" (no flooding occurred).

[0036] The explanatory variables are indicators that are assumed to be factors in flooding, such as elevation, Laplacian, slope, and ground conditions in each sample section S1a to S1y, as well as the slope of sample sections adjacent to the target sample section. In this embodiment, the following 10 indicators are used as explanatory variables 1 to 10.

[0037] Explanatory variable 1 is the elevation data for each sample segment S1a to S12y, which is divided by a 50m mesh, and explanatory variable 2 is the elevation data for the intermediate sample segments S1 to S12, which are divided by a 250m mesh that includes each sample segment S1a to S12y. For example, the elevation data for sample segments S1a to S1y consists of the elevation data for each sample segment S1a to S1y, which is divided by a 50m mesh (explanatory variable 1), and the elevation data for the intermediate sample segment S1, which includes each sample segment S1a to S1y (explanatory variable 2).

[0038] Elevation data for each sample category S1a to S12y can be obtained from sources such as the Geospatial Information Authority of Japan's Digital Elevation Model (https: / / fgd.gsi.go.jp / download / ref_dem.html) and the National Land Numerical Information Elevation / Slope 5th-order Mesh Data (https: / / nlftp.mlit.go.jp / ksj / gml / datalist / KsjTmplt-G04-d.html).

[0039] Explanatory variable 3 is the Laplacian of each sample segment S1a to S12y, and explanatory variable 4 is the Laplacian of the intermediate sample segments S1 to S12 that each sample segment S1a to S12y contains. The Laplacian is a value that quantifies the variation in unevenness in a certain region, and is calculated, for example, based on the elevation data published nationwide by the Ministry of Land, Infrastructure, Transport and Tourism, as mentioned above.

[0040] For example, explanatory variable 3 for sample division S1a is the Laplacian of a 50m mesh calculated based on the elevation data of sample division S1a, which is divided into 50m meshes, and explanatory variable 4 for sample division S1a is the Laplacian of a 250m mesh calculated based on the elevation data of intermediate sample division S1, which includes sample division S1a.

[0041] Explanatory variable 5 is the slope data for the intermediate sample segments S1 to S12, which contain each sample segment S1a to S12y. For example, the slope data for sample segments S1a to S1y is the slope data for the intermediate sample segment S1, which contains each of the sample segments S1a to S1y.

[0042] For each sample category S1a to S12y, slope data can be obtained from sources such as the National Land Numerical Information Elevation and Slope 5th-order Mesh Data (https: / / nlftp.mlit.go.jp / ksj / gml / datalist / KsjTmplt-G04-d.html).

[0043] Explanatory variable 6 is data on the ground conditions for each sample section S1a to S12y. For example, average S-wave velocity data down to 30m underground estimated from the microtopography of the National Research Institute for Earth Science and Disaster Resilience (J-SHIS) surface ground (https: / / www.j-shis.bosai.go.jp / map / ), or the Geospatial Information Authority of Japan's land condition maps (https: / / www.gsi.go.jp / bousaichiri / lc_index.html) can be used. For example, explanatory variable 6 for sample section S1a is the average S-wave velocity data down to 30m underground for sample section S1a.

[0044] Explanatory variables 7-10 are slope data for sample divisions S1a-S12y adjacent to the target sample division S1a-S12y. For explanatory variables 7-10, for example, the above-mentioned National Land Numerical Information Elevation / Slope 5th-order mesh data (https: / / nlftp.mlit.go.jp / ksj / gml / datalist / KsjTmplt-G04-d.html) can be used. For the slope data of adjacent sample divisions S1a-S12y, the slope data of sample divisions S1a-S12y divided by 50m mesh and the slope data of intermediate sample divisions S1-S12 which include adjacent sample divisions S1a-S12y are used.

[0045] Explanatory variables 7-10 are, for example, when the target sample segment is sample segment S1b, the slope data of the adjacent sample segment upstream of sample segment S1b, for example, sample segment S1a, becomes explanatory variable 7, and the slope data of the intermediate sample segment S1, which includes sample segment S1a, becomes explanatory variable 8. The slope data of the adjacent sample segment downstream of the target sample segment S1b, for example, sample segment S1c, becomes explanatory variable 9, and the slope data of the intermediate sample segment S1, which includes sample segment S1c, becomes explanatory variable 10.

[0046] The explanatory variables 1 to 10 described above are associated with the flooding probability p for each sample category S1a to S1y and stored in the storage unit 1 in a database as shown in Figure 7.

[0047] Based on the database stored in the memory unit 1, the probability of flooding occurring p for each sample category S1a to S1y is used as the dependent variable, and explanatory variables 1 to 10 are used. The calculation processing unit 2 then performs logistic regression analysis using, for example, a computer spreadsheet application, to derive the relationship equation.

[0048] Logistic regression analysis allows us to determine the regression variables β0 to β10 for each explanatory variable 1 to 10 in the relational equation. Based on the database shown in Figure 7, the logistic regression analysis was performed, and the regression variables β0 to β10 were determined as shown in Figure 8.

[0049] For the obtained regression variables β0 to β10, the significance probability is calculated and evaluated, and explanatory variables deemed to have no explanatory power are excluded. For example, the significance probability when the regression variable β is "0" is calculated and the obtained regression variables β0 to β10 are evaluated. In the example shown in Figure 8, the significance probability of explanatory variable 7, the upstream slope (50m mesh), is 0.56, and although the probability of flooding occurrence and the upstream slope (50m mesh) of explanatory variable 7 are unrelated [β=0], there is a 56% probability that β7=1.1 occurred by chance. Therefore, explanatory variable 7 is judged to have no explanatory power and is excluded from the explanatory variables. The regression variables β1 to β6 and β8 to β10 obtained in this way are then applied to (Equation 1) to derive a relational expression (an expression obtained by substituting the value of β into Equation 1) for determining the probability of flooding occurrence p of the desired evaluation target area RA.

[0050] Based on the derived relation (the equation obtained by substituting the value of β into Equation 1), the calculation processing unit 2 calculates the probability of flooding occurring in the evaluation target sections R1a to R12y, which are divided by a 50m-sized mesh in the evaluation target area RA shown in Figure 2.

[0051] Specifically, explanatory variables 1 to 10 for evaluation target categories R1a to R12y, which are divided using meshes of the same size as the sample categories (50m mesh), are acquired based on various data in the same way as for the sample categories. A database of evaluation target areas RA is created by associating these variables with each evaluation target category R1a to R12y, and stored in storage unit 1.

[0052] The values ​​from the database of the evaluation target area RA are substituted into the relational expression derived by the calculation processing unit 2 (the expression obtained by substituting the value of β into Equation 1), and the probability of flooding occurring p in each evaluation target category R1a to R12y is calculated. In other words, the risk of flooding occurring in each evaluation target category R1a to R12y is quantitatively evaluated by the probability of flooding occurring p.

[0053] According to the flood risk assessment method of this embodiment, the flood risk p for each evaluation target category R1a to R12y is calculated based on a relational equation (equation obtained by substituting the value of β into equation 1) derived by logistic regression analysis with the probability of flood occurrence p based on past data as the dependent variable and predetermined indicators as explanatory variables 1 to 10, and explanatory variables 1 to 10 consisting of the same indicator for multiple evaluation target categories R1a to R12y, obtained by dividing the evaluation target area RA into multiple evaluation target categories R1a to R12y with the same mesh size as the sample categories S1a to S12y. As a result, a more accurate flood risk p based on past cases is calculated. Therefore, it becomes possible to quantitatively evaluate the flood risk for each evaluation target category R1a to R12y.

[0054] Furthermore, since the relationship is derived using logistic regression analysis, the range of the flood probability p does not extend from -∞ to ∞, as shown in the simple regression analysis result in Graph B of Figure 9. Instead, the flood probability can be determined in the range of 0 to 1, as shown in Graph C of Figure 9. Therefore, it becomes possible to quantitatively evaluate the probability of flooding.

[0055] Furthermore, by including at least one of the following indicators—elevation, Laplacian, slope, geology, and ground conditions—in the sample divisions S1a-S12y or the evaluation divisions R1a-R12y, which can be factors in flood damage, as an explanatory variable in the logistic regression analysis, it becomes possible to calculate a more accurate probability of flood occurrence p.

[0056] Furthermore, since the target sample divisions S1a to S12y are connected to adjacent sample divisions S1a to S12y, it is possible to derive a relational equation (the equation obtained by substituting the value of β into Equation 1) based on information that includes the influence of adjacent sample divisions S1a to S12y. Therefore, it is possible to derive a relational equation (the equation obtained by substituting the value of β into Equation 1) that allows for the calculation of a more accurate flood occurrence probability p.

[0057] Furthermore, the calculation of the flood probability p for the desired evaluation area RA using this relation (the equation obtained by substituting the value of β into Equation 1) includes the indicators of adjacent evaluation areas R1a to R12y that are connected to the evaluation area R1a to R12y being evaluated. Therefore, it is possible to calculate the flood probability that includes the influence of adjacent evaluation areas R1a to R12y. For this reason, it is possible to calculate a more accurate flood probability p.

[0058] For example, as shown in Figure 10(a), if the slope of adjacent evaluation areas upstream is large and the slope of adjacent evaluation areas downstream is small, the likelihood of flooding is higher. Conversely, as shown in Figure 10(b), if the slope of adjacent evaluation areas upstream is small and the slope of adjacent evaluation areas downstream is large, flooding is considered less likely. Therefore, by using the slope data of adjacent evaluation areas upstream and the slope data of adjacent evaluation areas downstream as indicators, the accuracy of the calculated flood occurrence probability can be improved.

[0059] Furthermore, the logistic regression analysis used to derive the aforementioned relationship, and the calculation of the probability of flooding based on the relationship and the database, are performed by an information processing system 5 such as a computer, making it possible to easily and quickly obtain a more accurate probability of flooding.

[0060] ===Regarding other embodiments=== The embodiments described above are provided to facilitate understanding of the present invention and are not intended to limit its interpretation. The present invention can be modified and improved without departing from its spirit, and it goes without saying that equivalents thereof are included. In particular, embodiments described below are also included in the present invention.

[0061] In the above embodiment, an example was described in which the probability of flooding occurs is calculated using historical data that shows the locations of flooding in an area that has previously experienced flooding, as an evaluation of the risk of flooding. However, this is not the only example. For instance, as historical data for flooding, the probability of landslides may be calculated using historical data that shows the locations of landslides in an area that has previously experienced landslides, and the risk of flooding may be evaluated by calculating this probability.

[0062] In the embodiments described above, examples were explained in which land history survey data, elevation / slope 5th-order mesh data from the National Land Numerical Information System, and J-SHIS surface ground data from the National Research Institute for Earth Science and Disaster Resilience were used as data for calculating the probability of flood occurrence. However, the examples are not limited to these; other data may be used as long as they are standardized nationwide and have a certain degree of reliability. [Explanation of symbols]

[0063] 1 storage section, 2. Processing Unit, 3. Input section, 4 Display section, 5. Information processing systems, Q: Locations where flooding occurred, Areas of RA evaluation, R1~R12 Intermediate evaluation category, R1a~R12y Evaluation target categories, SA sample area, S1~S12 Sample Intermediate Section, S1a~S12y Sample classification, Area with M50 mesh size of 50m, M250, area with a mesh size of 250m,

Claims

1. A method for evaluating flood risk, in which the probability of flood occurrence based on past data showing the presence or absence of past floods in a certain area is used as the dependent variable, and a predetermined indicator of the said area is used as the independent variable, and a relationship is derived by logistic regression analysis, and the relationship is derived by a computer. The aforementioned historical data is data indicating whether or not there have been past floods in each of the multiple sample divisions, which are obtained by dividing a sample area containing locations where floods have occurred in the past into predetermined mesh sizes. A flood risk assessment method characterized in that the predetermined indicator includes the Laplacian as an indicator that serves as an explanatory variable in a logistic regression analysis, which is information for each of the aforementioned sample categories.

2. A flood risk assessment method according to claim 1, A flood risk assessment method characterized in that a computer calculates the probability of flood occurrence in each of the multiple evaluation target categories obtained by dividing the evaluation target area into predetermined mesh sizes, based on the aforementioned relational expression and the predetermined index indicating information for each of the evaluation target categories.

3. A flood risk assessment method according to claim 2, A flood risk assessment method characterized in that the predetermined indicator includes at least one of elevation, slope, geology, and ground conditions in the sample category or the evaluation target category.

4. A flood risk assessment method according to claim 2 or 3, A flood risk assessment method characterized in that the predetermined indicators include the predetermined indicators of the sample category adjacent to the target sample category and the predetermined indicators of the evaluation target category adjacent to the evaluation target category.

5. A flood risk assessment method according to any one of claims 2 to 4, A flood risk assessment method characterized in that the probability of flood occurrence is the probability of inundation.

6. A storage unit that stores historical data indicating whether or not there have been past floods in a certain area, and a predetermined indicator for the said area. A calculation processing unit that uses the probability of flood occurrence based on the aforementioned past data as the dependent variable and the aforementioned predetermined indicator as the independent variable, and derives a relationship between the aforementioned past data and the aforementioned predetermined indicator by performing logistic regression analysis, It has, The aforementioned historical data is data indicating whether or not there have been past floods in each of the multiple sample divisions, which are obtained by dividing a sample area containing locations where floods have occurred in the past into predetermined mesh sizes. A flood risk assessment system characterized in that the predetermined indicator includes the Laplacian as an indicator that serves as an explanatory variable in a logistic regression analysis, which is information for each of the aforementioned sample categories.

Images