Disaster scale estimation device and disaster scale estimation method

Abstract

To provide a disaster scale estimation device that can quickly and accurately grasp the extent of damage in the event of a disaster. [Solution] The disaster scale estimation device 1 comprises: an NDVI value calculation unit 121 that calculates the difference in NDVI values ​​before and after a disaster from ground surface data of the target area before and after the disaster; a pixel extraction unit 122 that extracts pixels from the ΔNDVI value calculated by the NDVI value calculation unit 121 that are greater than a preset first threshold as the sediment runoff estimation range A5; a collapsed sediment production amount estimation unit 123 that estimates the collapsed sediment production amount Vs from the area A of the sediment runoff estimation range A5 extracted by the pixel extraction unit 122; a driftwood production amount estimation unit 124 that estimates the driftwood production amount Vw from the collapsed sediment production amount Vs estimated by the collapsed sediment production amount estimation unit 123; a recovery period calculation unit 125 that calculates the recovery period T from the cumulative rainfall at the time of the disaster; and a disaster scale estimation unit 126 that estimates the disaster scale based on the collapsed sediment production amount Vs, the driftwood production amount Vw, and the recovery period T.

Description

[Technical Field]

[0001] The present invention relates to a disaster scale estimation device and a disaster scale estimation method for understanding the extent of damage during a disaster. [Background technology]

[0002] In recent years, due to extreme weather events, widespread landslides have become frequent in Japan. After a disaster, in order to promote efficient recovery, the forestry or erosion control management bodies must quickly grasp the extent of the damage and report it to the relevant organizations. Conventionally, a system for extracting landslide areas has been disclosed that detects areas where the Normalized Difference Vegetation Index (NDVI) is less than a threshold from among collapsed areas extracted by taking visible light images of the ground surface from above (see Patent Document 1). [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-140677 [Non-patent literature]

[0004] [Non-Patent Document 1] Simonett, DS (1967): Landslide distribution and earthquakes in the Bewani and Torricelli Mountains, New Guinea. In: Jennings, JN, Mabbutt, JA (Eds.), Landform Studies from Australia and New Guinea, Cambridge University Press, Cambridge, p.64-84 [Overview of the project] [Problems that the invention aims to solve]

[0005] The technology described in Patent Document 1 can extract the locations of slope collapses, but it could not obtain other information such as the scale of the collapses.

[0006] The present invention aims to provide a disaster scale estimation device and a disaster scale estimation method that, compared to the prior art, can not only extract locations of slope collapses but also quickly and accurately grasp the extent of damage in the event of a disaster. [Means for solving the problem]

[0007] The disaster scale estimation device according to the present invention is An NDVI value calculation unit calculates the ΔNDVI value, which represents the difference in NDVI values ​​before and after a disaster, from the ground surface data of the target area before and after the disaster, A pixel extraction unit extracts pixels from the ΔNDVI value calculated by the NDVI value calculation unit that are greater than a preset first threshold as the estimated sediment runoff range. A collapse-production sediment quantity estimation unit estimates the amount of collapsed sediment produced from the area of ​​the estimated sediment runoff range extracted by the pixel extraction unit, A driftwood production amount estimation unit estimates the amount of driftwood produced from the amount of landslide-produced sediment estimated by the aforementioned landslide-produced sediment production amount estimation unit, A recovery period calculation unit that calculates the recovery period from the cumulative rainfall during a disaster, A disaster scale estimation unit that estimates the scale of the disaster based on the amount of sediment produced by the collapse, the amount of driftwood produced, and the recovery period, It is equipped with. [Effects of the Invention]

[0008] Such disaster scale estimation devices allow for a rapid and accurate assessment of the damage situation during a disaster. [Brief explanation of the drawing]

[0009] [Figure 1] An example of an overall disaster scale estimation system using the disaster scale estimation device of this embodiment is shown. [Figure 2]An example of the disaster scale estimation device according to this embodiment is shown. [Figure 3] An example of the surface data acquired by the disaster scale estimation device according to this embodiment is shown. [Figure 4] An example of the pixels of the sediment outflow range calculated by the NDVI value calculation unit and the pixel extraction unit according to this embodiment is shown. [Figure 5] An example of a figure in which the region polygon according to this embodiment is superimposed on the polygon data of the probability rainfall is shown. [Figure 6] The past sediment disaster groups are shown. [Figure 7] An example of a flowchart of the disaster scale estimation method according to this embodiment is shown.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. In the following, the scope necessary for explaining the present invention is schematically shown, and mainly the scope necessary for explaining the relevant part of the present invention will be described, and the parts for which the explanation is omitted are assumed to be based on known techniques.

[0011] In the emergency restoration after a sediment disaster, it is important to grasp the amount of sediment and driftwood flowing out in the affected area and basin, and this information can be one of the bases for setting the sand storage capacity of the sand control and mountain management facilities planned after the disaster. For example, when applying for the extremely severe disaster system, it is necessary to show the approximate business expenses (estimated amount) for restoration. These can be calculated from the height and volume of newly planned structures, etc., and whether the order of restoration business expenses meets the criteria for extremely severe disaster recognition is one of the judgment criteria. Furthermore, when applying for a sediment removal project, the amount of sediment flowing in and sediment deposited from collapses caused by the disaster is required. From the above, in the emergency restoration phase, if information such as the location and scope of the sediment disaster and the amount of sediment and driftwood flowing out can be obtained over a wide area, it is considered useful for grasping the outline after the disaster and can smoothly transition to subsequent permanent measures.

[0012] Figure 1 shows the overall disaster scale estimation system 100 using the disaster scale estimation device 1 of this embodiment.

[0013] The disaster scale estimation device 1 of this embodiment is used in a disaster scale estimation system 100. The disaster scale estimation system 100 comprises an aircraft 2 such as a satellite, aircraft, or drone that acquires ground surface data, a receiving device 3 that receives ground surface data transmitted from the aircraft 2, a database 4 that stores the ground surface data received by the receiving device 3, and a disaster scale estimation device 1 that estimates the scale of the disaster based on the ground surface data. The receiving device 3, the database 4, and the disaster scale estimation device 1 are connected by a wired or wireless network 5, etc. The disaster scale estimation device 1 may acquire ground surface data directly from the receiving device 3. Alternatively, the disaster scale estimation device 1 may store the ground surface data.

[0014] The disaster scale estimation system 100 uses ground data before and after a disaster, which is photographed from above by the aircraft 2 and received by the receiving device 3. Pre-disaster ground data can be taken from data previously photographed and stored in the database 4. Post-disaster ground data should preferably be acquired as soon as possible. Furthermore, post-disaster ground data should preferably be ground data from a satellite that can acquire data over a wide area periodically with each orbit, regardless of weather conditions.

[0015] Figure 2 shows the disaster scale estimation device 1 of this embodiment.

[0016] The disaster scale estimation device 1 is composed of, for example, a general-purpose or dedicated computer. The disaster scale estimation device 1 includes at least a control unit 10 for estimating the scale of the disaster. Preferably, the disaster scale estimation device 1 also includes a communication unit 13 connected to an external device such as a receiving device 3 or a database 4, a storage unit 14 for temporarily storing at least pre-disaster ground data 141 and post-disaster ground data 142, an input unit 15 such as a keyboard, and an output unit 16 such as a display.

[0017] The control unit 10 includes a ground data acquisition unit 11 and a calculation unit 12. The ground data acquisition unit 11 acquires ground data before and after a disaster from the receiving device 3 or database 4 via the communication unit 13. The acquired pre-disaster ground data 141 and post-disaster ground data 142 are stored in the storage unit 14.

[0018] Figure 3 shows an example of ground surface data acquired by the disaster scale estimation device 1 of this embodiment. Figure 4 shows an example of pixels representing the sediment runoff area calculated by the NDVI value calculation unit 121 and the pixel extraction unit 122 of this embodiment. The calculation unit 12 includes an NDVI value calculation unit 121 that calculates an NDVI value (NDVI: Normalized difference vegetation index) from ground surface data before and after the disaster, a pixel extraction unit 122 that extracts pixels that satisfy predetermined conditions from the ground surface data, a collapse production sediment volume estimation unit 123 that estimates the amount of collapsed sediment from the area A of the estimated sediment runoff range A5 after pixel extraction, a driftwood production volume estimation unit 124 that estimates the driftwood production volume Vw from the collapse production sediment volume Vs, a recovery period calculation unit 125 that calculates the recovery period T from the cumulative rainfall at the time of the disaster, and a disaster scale estimation unit 126 that estimates the scale of the disaster based on the collapse production sediment volume estimation unit 123, the driftwood production volume estimation unit 124, and the recovery period calculation unit 125.

[0019] The NDVI value calculation unit 121 uses ground surface data to calculate the NDVI values ​​before and after the disaster. Subsequently, the NDVI value calculation unit 121 calculates the ΔNDVI value, which represents the difference between the NDVI value before the disaster and the NDVI value after the disaster.

[0020] The NDVI value is calculated using the following formula (1). NDVI = (NIR-RED) / (NIR+RED) (1) Here, NIR is the reflectance in the near-infrared region. RED is the reflectance in the red range. That is the case.

[0021] The NDVI value ranges from -1.0 to 1.0, and increases as the reflectance in the near-infrared region is greater than the reflectance in the red region. Therefore, pixels that capture forests (plants) will have a high NDVI value, while pixels that capture bare ground (soil) will have a low NDVI value. Furthermore, the NDVI value calculation unit 121 creates a ΔNDVI image by calculating the ΔNDVI value using the following equation (2). ΔNDVI = NDVI Pre-event - NDVI Post-event (2) Here, NDVI Pre-event The NDVI value before the disaster, NDVI Post-event This refers to the NDVI value after the disaster. That is the case.

[0022] The NDVI value calculation unit 121 calculates the difference value, ΔNDVI, which allows for the removal of existing bare ground, such as past landslide areas and reclaimed land, from the bare ground shown in the satellite data after the disaster. In other words, the ΔNDVI value will be large in areas where landslides or debris flows have occurred and newly created bare ground, and the ΔNDVI value will approach 0 in areas where there has been little change in the ground surface before and after the disaster.

[0023] The pixel extraction unit 122 uses the ΔNDVI value calculated by the NDVI value calculation unit 121 to distinguish between vegetation pixels in new bare ground A2 and vegetation pixels in areas other than new bare ground A1, based on a preset first threshold, and extracts pixels with a value greater than the first threshold as new bare ground A2. In this embodiment, new bare ground A2 is defined as an area where vegetation has newly changed to bare ground (assuming that soil runoff has occurred due to collapse or debris flow). The disaster scale estimation device 1 of this embodiment distinguishes areas where the ΔNDVI value is greater than or equal to the first threshold as new bare ground A2, and distinguishes all other areas as areas other than new bare ground A1. The pixel extraction unit 122 creates a binarized image based on the results of the discrimination using the first threshold. In this embodiment, the first threshold is preferably 0.1 or more and 0.3 or less. Furthermore, the first threshold is more preferably 0.15 or more and 0.25 or less.

[0024] Furthermore, the pixel extraction unit 122 is primarily intended for detecting areas with a high probability of slope failure or debris flow. Therefore, it determines areas with a slope gradient below a preset second threshold as deposition area A3 and extracts pixels with a gradient above the second threshold from the binarized image. In this embodiment, the second threshold is preferably between 10 degrees and 20 degrees. More preferably between 14 degrees and 16 degrees, and most preferably 15 degrees. Exclusion based on slope gradient also consequently excludes pixels along the riverbanks that originate from sediment deposited in the river. It is preferable to use the 10m DEM from the Geospatial Information Authority of Japan or similar for the slope gradient here.

[0025] Furthermore, the pixel extraction unit 122 interprets the ground surface data before and after the disaster, creates an exclusion area A4 of the area covered by clouds within the target region, and excludes the area that overlaps with the extracted pixels.

[0026] The pixel extraction unit 122 last extracts data with a resolution less than the spatial resolution of the ground data (for example, 100m when using Sentinel-2 satellite data as ground data). 2 Since pixels smaller than a certain size are considered unclear as sediment runoff, it is preferable to calculate the individual area A of the extracted sediment runoff estimation area A5 using a Geographic Information System (GIS) and exclude the exclusion area A4 which represents the smallest extracted area less than the spatial resolution of the ground surface data.

[0027] Furthermore, it is preferable to create contour lines from the Geospatial Information Authority of Japan's 10m DEM and overlay them to interpret all extracted estimated sediment runoff areas A5, and to check for any abnormal extracted data outside of mountainous areas. After that, it is preferable to calculate the area A of each estimated sediment runoff area A5 using a geometry tool or similar.

[0028] Although accuracy will decrease, the pixel extraction unit 122 may determine the new bare ground A2 as the estimated sediment runoff range A5 without excluding the deposition area A3 and the exclusion area A4. Alternatively, the new bare ground A2 may be determined as the estimated sediment runoff range A5 by excluding only the deposition area A3 or only the exclusion area A4.

[0029] The collapse production sediment volume estimation unit 123 estimates the amount of collapsed sediment from the area A of the sediment outflow estimation range A5 after the pixel extraction unit 122 has extracted the area. The sediment outflow estimation range A5 is divided into an occurrence area and a flow area according to the definition of sediment movement patterns, and can therefore be considered a new collapse area. For this reason, the amount of collapse production sediment Vs from within the sediment outflow estimation range A5, i.e., within the collapse area, can be estimated from the following equation (3) shown in Non-Patent Literature 1. Vs=αA γ (3) Here, A is the area of ​​the estimated region of soil erosion. α and γ are coefficients. That is the case.

[0030] The area A of the estimated sediment runoff range A5 can be easily calculated using GIS, but the problem lies in how to set the coefficients α and γ. The collapse production sediment volume estimation unit 123 of this embodiment proposes two models: the model in equation (4) in which the value of coefficient γ is estimated from the relationship with the value of coefficient α, and the general linear model in equation (5) in which the value of coefficient α is normalized by the slope gradient of the collapse site and the average annual rainfall, and the 48h cumulative rainfall is used as a variable. The collapse production sediment volume Vs is calculated from equation (6) obtained by substituting equations (4) and (5) into equation (3). γ = -0.153loge(α) + 1.053 (4) α= 0.686 - 1.747X1 + 0.013X2 (5) Vs = (α)A {-0.153loge(α)+1.053} (6) Here, X1 is the 48-hour cumulative rainfall normalized by the annual average rainfall. X2 is the average slope of the landslide area. That is the case.

[0031] The driftwood production estimation unit 124 estimates the driftwood production Vw from the amount of sediment produced by landslides Vs. The driftwood production Vw is estimated from equation (7), which is a relationship between the amount of sediment produced by landslides Vs and the amount of driftwood production Vw, statistically derived from driftwood disaster data over the past 40 years. Then, the estimated amount of sediment and driftwood V, which is the sum of Vs and Vw, is calculated for each sediment runoff area using equation (8). Vw = 0.018Vs (7) V = Vs + Vw (8)

[0032] The recovery period calculation unit 125 calculates the recovery period T of the cumulative rainfall for each target area in order to get an overview of the scale of the disaster from the perspective of rainfall, based on the estimated amount of sediment and driftwood V. The recovery period T is a representation method that estimates the amount of rainfall that occurs statistically once every x years (1 / x) using a probability distribution.

[0033] Figure 5 shows an example of a diagram in which the region polygon of this embodiment is overlaid on the polygon data of probabilistic rainfall. The recurrence period calculation unit 125 of this embodiment reads the National Institute for Land and Infrastructure Management data into ArcGIS Pro, processes it into spatial polygon data, and utilizes it. As a procedure, region polygons encompassing the individual sediment runoff estimation range A5 extracted from each target area are created as shown in Figure 3, and overlaid on the polygon data of probabilistic rainfall. Then, the average values ​​of the 1h, 24h, and 48h probability cumulative rainfall included in the region polygon of the target area for nine cases of 1 / 2, 1 / 3, 1 / 5, 1 / 10, 1 / 20, 1 / 30, 1 / 50, 1 / 100, and 1 / 200 year probabilities are calculated and organized using GIS. After performing these processes for all target areas, the recurrence period T is obtained from the functional form of equation (9). y = ax b (9) Here, y is the recovery period (years), x is the probability cumulative rainfall (mm), That is the case. Note that coefficients a and b are predetermined for each estimation point.

[0034] The disaster scale estimation unit 126 estimates the scale of the disaster based on the collapse production sediment volume estimation unit 123, the driftwood production volume estimation unit 124, and the recovery period calculation unit 125.

[0035] Figure 6 shows past landslide disaster groups. The areas enclosed by solid lines are past landslide disaster groups designated as severe disasters. In this embodiment, areas that were actually designated as severe disasters but had an extremely small number of samples for estimated sediment runoff ranges (only two) were classified unclearly. Although there is variability in the data plot, there is a tendency for the estimated sediment and driftwood amounts to increase as the maximum recurrence period Tm increases.

[0036] In addition, the data plots obtained by LiDAR terrain differential analysis indicated by the ○ mark are the result of using highly accurate data for both terrain and rainfall, and it is considered that the calculated recurrence period T and the estimated amount of sediment and driftwood are close to the true values. Looking at the overall data plots of the satellite data extraction analysis indicated by the × mark, since the trend does not deviate from the plot group of the LiDAR terrain differential analysis, the validity of the extraction results of the disaster scale estimation device 1 of the present embodiment can be confirmed.

[0037] If the maximum recurrence period Tm is 9.6 years (almost 10 years) and the estimated amount of sediment and driftwood from the collapsed area exceeds about 10,000 m 3 it is considered to correspond to the extremely severe disaster area of past sediment disasters (if it is 100 years, about 40,000 m 3 ). Note that the extremely severe disaster area does not necessarily have to exceed a maximum recurrence period Tm of 9.6 years (almost 10 years) and an estimated amount of sediment and driftwood from the collapsed area of about 10,000 m 3 . As another example, it may be estimated that it corresponds to the extremely severe disaster area when the maximum recurrence period is 9 years or more and the estimated amount of sediment and driftwood from the collapsed area is 9,000 m 3 or more. As described above, by calculating from the estimated sediment outflow range extracted from satellite data to the estimated amount of sediment and driftwood and plotting it on a relationship diagram with the separately obtained recurrence period T, it is possible to comprehensively determine whether it corresponds to the past extremely severe disaster area, and the series of these processes can be made possible within the emergency restoration phase by the disaster scale estimation device 1 of the present embodiment.

[0038] Thus, according to the disaster scale estimation device 1 of the present embodiment, the damage situation at the time of a disaster can be grasped quickly and accurately. And this information can be one of the bases for setting the sand storage capacity of the sediment control and mountain protection facilities planned after the disaster.

[0039] FIG. 7 shows an example of a flowchart of the disaster scale estimation method of the present embodiment.

[0040] First, in step 1, the NDVI value calculation unit 121 calculates the NDVI value for each pixel of the ground surface data before and after the disaster (ST1). Next, in step 2, the NDVI value calculation unit 121 calculates the ΔNDVI value, which represents the difference between the NDVI values ​​before and after the disaster (ST2). Steps 1 and 2 describe the ΔNDVI value calculation process.

[0041] Next, in step 3, the pixel extraction unit 122 extracts pixels from the ΔNDVI value calculated by the ΔNDVI value calculation unit that are equal to or greater than a preset first threshold as new bare ground A2 (ST3).

[0042] Next, in step 4, the pixel extraction unit 122 extracts pixels with a gradient equal to or greater than a preset second threshold as deposition area A3 (ST4). Subsequently, in step 5, the pixel extraction unit 122 excludes pixels smaller than the cloud area and the minimum extraction area as exclusion area A4 (ST5). Steps 3 to 5 describe the pixel extraction process.

[0043] Next, in step 6, the collapse-production sediment volume estimation unit 123 performs a collapse-production sediment volume estimation process (ST6) in which it estimates the collapse-production sediment volume Vs from the area A of the sediment outflow estimation range A5 extracted in the pixel extraction process.

[0044] Next, in step 7, the driftwood production estimation unit 124 executes a driftwood production estimation process (ST7) in which it estimates the amount of driftwood produced Vw from the amount of sediment produced by collapse that was estimated in the collapse production sediment production estimation process.

[0045] Next, in step 8, the recovery period calculation unit 125 performs a recovery period calculation process (ST8) to calculate the recovery period T from the accumulated rainfall during the disaster.

[0046] Next, in step 9, the disaster scale estimation unit 126 performs a disaster scale estimation process (ST9) to estimate the scale of the disaster based on the amount of sediment produced by the collapse Vs, the amount of driftwood produced Vw, and the recovery period T. In the disaster scale estimation process, Figure 6 is output, plotting the calculated maximum recovery period Tm and the estimated amount of sediment and driftwood V from the collapsed soil. Furthermore, it automatically determines whether the calculated values ​​fall within the areas of past severe disasters enclosed by the solid lines in Figure 6, and outputs the determination result.

[0047] Furthermore, before step 1, there may be a ground data acquisition step in which the ground data acquisition unit 11 acquires ground data, and a storage step in which the data acquired in the ground data acquisition step is stored in the storage unit 14 which stores the data before and after the disaster.

[0048] Thus, the disaster scale estimation method of this embodiment allows for a rapid and accurate assessment of the damage situation at the time of a disaster. This information can then be used as one of the basis for setting the sediment storage capacity of erosion control and landslide prevention facilities planned after the disaster.

[0049] The disaster scale estimation device 1 of this embodiment may execute each step of the disaster scale estimation method controlled by the flowchart shown in Figure 7 using a computer control program. By being executed by a computer control program, the disaster scale estimation device 1 can automatically output the results, from acquiring ground surface data to estimating the disaster scale, without human visual inspection. Therefore, according to the control program of the disaster scale estimation device of this embodiment, the damage situation at the time of a disaster can be grasped quickly and accurately. This information can then be used as one of the grounds for setting the sediment storage capacity of erosion control and landslide prevention facilities planned after the disaster.

[0050] As described above, the disaster scale estimation device 1 of this embodiment includes: an NDVI value calculation unit 121 that calculates a ΔNDVI value indicating the difference in NDVI values ​​before and after a disaster from ground surface data of the target area before and after the disaster; a pixel extraction unit 122 that extracts pixels from the ΔNDVI value calculated by the NDVI value calculation unit 121 that are greater than a preset first threshold as the sediment runoff estimation range A5 (new bare ground A2); a collapsed sediment production amount estimation unit 123 that estimates the collapsed sediment production amount Vs from the area A of the sediment runoff estimation range A5 extracted by the pixel extraction unit 122; a driftwood production amount estimation unit 124 that estimates the driftwood production amount Vw from the collapsed sediment production amount Vs estimated by the collapsed sediment production amount estimation unit 123; a recovery period calculation unit 125 that calculates the recovery period T from the cumulative rainfall at the time of the disaster; and a disaster scale estimation unit 126 that estimates the disaster scale based on the collapsed sediment production amount Vs, the driftwood production amount Vw, and the recovery period T. Therefore, the disaster scale estimation device 1 of this embodiment can quickly and accurately grasp the extent of damage when a disaster occurs.

[0051] Furthermore, the pixel extraction unit 122 extracts pixels whose slope gradient is equal to or greater than a preset second threshold. Therefore, the disaster scale estimation device 1 of this embodiment excludes pixels in depositional areas A3, such as along the riverbanks, which are derived from sediment deposited in the river, by excluding them based on the slope gradient, and can accurately grasp the extent of damage when a disaster occurs.

[0052] Furthermore, the pixel extraction unit 122 interprets the ground surface data before and after the disaster, creates a section within the target area where clouds are present, and excludes the extracted pixels as exclusion area A4. Therefore, the disaster scale estimation device 1 of this embodiment can eliminate misidentification of clouds and accurately grasp the extent of damage at the time of the disaster.

[0053] Furthermore, the pixel extraction unit 122 excludes pixels representing the minimum extraction area less than the spatial resolution of the ground surface data as exclusion area A4. Therefore, the disaster scale estimation device 1 of this embodiment can exclude areas where it is unclear whether or not there is soil runoff, and can grasp the extent of damage at the time of a disaster with high accuracy.

[0054] Furthermore, the landslide production sediment volume estimation unit 123 estimates the amount of landslide production sediment Vs from within the estimated sediment runoff range using the following equation (3), and calculates the amount of landslide production sediment Vs from equation (6), which is obtained by substituting equation (4), which is derived from the relationship between the γ value and the α value, and equation (5), which uses the 48h cumulative rainfall normalized by the slope gradient of the landslide area and the average annual rainfall as variables, into equation (3). Vs=αA γ (3) γ = -0.153loge(α) + 1.053 (4) α= 0.686 - 1.747X1 + 0.013X2 (5) Vs = (α)A {-0.153loge(α)+1.053} (6) Here, Vs is the amount of sediment produced by landslides, statistically calculated from driftwood disaster data over the past 40 years. A is the area of ​​the estimated region of soil erosion. α and γ are coefficients. X1 is the 48-hour cumulative rainfall normalized by the annual average rainfall. X2 is the average slope of the landslide area. That is the case.

[0055] Therefore, the disaster scale estimation device 1 of this embodiment can estimate the amount of sediment produced by collapse Vs with high accuracy, and can grasp the damage situation at the time of a disaster with high accuracy.

[0056] Furthermore, the driftwood production estimation unit 124 calculates the amount for each area of ​​sediment runoff using the following equations (7) and (8). Vw = 0.018Vs (7) V = Vs + Vw (8) Here, Vs is the amount of sediment produced by landslides, statistically calculated from driftwood disaster data over the past 40 years. Vw is the volume of driftwood produced. V is the estimated amount of sediment and driftwood. That is the case.

[0057] Therefore, the disaster scale estimation device 1 of this embodiment can estimate the amount of driftwood produced Vw with high accuracy, and can accurately grasp the extent of damage when a disaster occurs.

[0058] Furthermore, the recurrence period calculation unit 125 creates region polygons that encompass the individual sediment runoff estimation ranges extracted from the target area, overlays them with the probability rainfall polygon data, estimates the probability cumulative rainfall for a predetermined time included within the region polygon of the target area for multiple recurrence probability years, calculates the average value for all of the target area, and obtains the recurrence period T from the functional form of equation (9). y = ax b (9) Here, y is the recovery period T (years), x is the probability cumulative rainfall (mm), That is the case. Note that coefficients a and b are predetermined for each target area.

[0059] Therefore, the disaster scale estimation device 1 of this embodiment can estimate the recovery period T of the target area with high accuracy and grasp the extent of damage at the time of the disaster with high accuracy.

[0060] Furthermore, the disaster scale estimation unit 126 estimates the scale of the disaster from a data plot of a graph on which the maximum recovery period Tm obtained from the recovery period calculation unit 125 is plotted on the horizontal axis, and the estimated amounts of sediment and driftwood obtained from the collapse production sediment volume estimation unit 123 and the driftwood production volume estimation unit 124 are plotted on the vertical axis.Therefore, the disaster scale estimation device 1 of this embodiment can grasp the damage situation at the time of a disaster with high accuracy.

[0061] Furthermore, the disaster scale estimation unit 126 assumes that the maximum recovery period Tm is 9 years or more, and the estimated amount of sediment and driftwood from the landslide area is 9,000 m³. 3 In the above cases, it is presumed that the area falls under the category of a severe disaster. Therefore, the disaster scale estimation device 1 of this embodiment can grasp the extent of damage at the time of a disaster with greater accuracy.

[0062] Furthermore, the disaster scale estimation device 1 includes a ground data acquisition unit 11 that acquires ground data, and a storage unit 14 that stores the data acquired by the ground data acquisition unit 11 before and after the disaster. Therefore, the disaster scale estimation device 1 of this embodiment can grasp the extent of damage at the time of a disaster more quickly.

[0063] It should be noted that the present invention is not limited by this embodiment. That is, while the description of the embodiment includes many specific details for illustrative purposes, those skilled in the art may make various variations and modifications to these details. [Explanation of symbols]

[0064] 1...Disaster scale estimation device, 121...NDVI value calculation unit, 122...Pixel extraction unit, 123...Collapse production sediment volume estimation unit, 124...Driftwood production volume estimation unit, 125...Recovery period calculation unit, 126...Disaster scale estimation unit

Claims

1. An NDVI value calculation unit calculates the ΔNDVI value, which represents the difference in NDVI values ​​before and after a disaster, from the ground surface data of the target area before and after the disaster. A pixel extraction unit extracts pixels from the ΔNDVI value calculated by the NDVI value calculation unit that are greater than a preset first threshold as the estimated soil runoff range. A collapse-production sediment quantity estimation unit estimates the amount of collapsed sediment produced from the area of ​​the estimated sediment runoff range extracted by the pixel extraction unit, A driftwood production amount estimation unit estimates the amount of driftwood produced from the amount of landslide-produced sediment estimated by the aforementioned landslide-produced sediment production amount estimation unit, A recovery period calculation unit that calculates the recovery period from the cumulative rainfall during a disaster, A disaster scale estimation unit that estimates the scale of the disaster based on the amount of sediment produced by the collapse, the amount of driftwood produced, and the recovery period, Equipped with, Disaster scale estimation device.

2. The aforementioned pixel extraction unit, Extract pixels where the slope gradient is equal to or greater than a pre-set second threshold. The disaster scale estimation device according to claim 1.

3. The aforementioned pixel extraction unit interprets the ground surface data before and after the disaster, creates a section representing the area covered by clouds within the target region, and excludes the area that overlaps with the extracted pixels. The disaster scale estimation device according to claim 1.

4. The aforementioned pixel extraction unit excludes pixels that represent the minimum extraction area less than the spatial resolution of the ground surface data. The disaster scale estimation device according to claim 1.

5. The aforementioned collapse production sediment volume estimation unit is: The amount of sediment produced by landslides Vs from within the estimated range of sediment runoff is estimated from the following equation (3): The amount of sediment produced by the landslide is calculated from equation (6), which is obtained by substituting equation (4), inferring the γ value from the relationship with the α value, and equation (5), which uses the 48h cumulative rainfall normalized by the slope gradient of the landslide area and the average annual rainfall as variables, into equation (3). The disaster scale estimation device according to claim 1. Vs=αA γ (3) γ = -0.153loge(α) + 1.053 (4) α= 0.686 - 1.747X1 + 0.013X2 (5) Vs= (α)A {-0.153loge(α)+1.053} (6) Here, Vs is the amount of sediment produced by landslides, statistically calculated from driftwood disaster data over the past 40 years. A is the area of ​​the estimated region of soil erosion. α and γ are coefficients. X1 is the 48-hour cumulative rainfall normalized by the annual average rainfall. X2 is the average slope of the landslide area. That is the case.

6. The driftwood production estimation unit calculates for each estimated sediment runoff area using the following equations (7) and (8): The disaster scale estimation device according to claim 1. Vw = 0.018Vs (7) V = Vs + Vw (8) Here, Vs is the amount of sediment produced by landslides, statistically calculated from driftwood disaster data over the past 40 years. Vw is the volume of driftwood produced. V is the estimated amount of sediment and driftwood. That is the case.

7. The aforementioned recovery period calculation unit, A region polygon is created that encompasses each of the estimated sediment runoff areas extracted from the aforementioned target area, and this is overlaid with the polygon data for probabilistic rainfall. The cumulative probability rainfall over a predetermined time within the area polygon of the target region is estimated using multiple recurrence probability years, and the average value is calculated for all of the target region. The recurrence period is determined from the functional form of equation (9). The disaster scale estimation device according to claim 1. y= ax b (9) Here, y is the recovery period (years), x is the probability cumulative rainfall (mm), That is the case. Note that coefficients a and b are predetermined for each target area.

8. The disaster scale estimation unit estimates the disaster scale from a data plot of a graph on which the maximum recovery period obtained from the recovery period calculation unit is plotted on the horizontal axis, and the estimated amounts of sediment and driftwood obtained from the collapse production sediment volume estimation unit and the driftwood production volume estimation unit are plotted on the vertical axis. The disaster scale estimation device according to claim 1.

9. The aforementioned disaster scale estimation unit is The aforementioned maximum recovery period is 9 years or more, and the estimated amount of sediment and driftwood from the estimated sediment runoff area is 9,000 m 3 In the above cases, it is presumed that the area falls under the category of a severe disaster. The disaster scale estimation device according to claim 8.

10. A ground data acquisition unit that acquires ground surface data, A storage unit that stores data acquired by the ground surface data acquisition unit before and after a disaster, Equipped with, The disaster scale estimation device according to claim 1.

11. A ΔNDVI value calculation process that calculates the ΔNDVI value, which shows the difference in NDVI values ​​before and after a disaster, from the ground surface data of the target area before and after the disaster, A pixel extraction step is performed to extract pixels from the ΔNDVI value calculated in the ΔNDVI value calculation step that are greater than a preset first threshold as the estimated range of soil runoff. A collapse-production sediment amount estimation step, which estimates the amount of collapse-production sediment from the area of ​​the estimated sediment runoff range extracted in the pixel extraction step, A driftwood production estimation step, which estimates the amount of driftwood produced from the amount of landslide-produced sediment estimated in the aforementioned landslide-produced sediment production estimation step, The recovery period calculation process involves calculating the recovery period from the cumulative rainfall during the disaster, and A disaster scale estimation process that estimates the scale of the disaster based on the amount of sediment produced by the collapse, the amount of driftwood produced, and the recovery period, Having, Disaster scale estimation method.

12. To cause a computer to perform each step of the disaster magnitude estimation method described in claim 11, A control program for disaster scale estimation methods.

Images