An unmanned aerial vehicle photogrammetry-based strong earthquake zone debris flow early warning method

Abstract

The application discloses a strong earthquake area debris flow early warning method based on unmanned aerial vehicle photogrammetry, and belongs to the technical field of debris flow prevention and control engineering, and is characterized by comprising the following steps: S1, determining basic parameters of a potential debris flow basin through a topographic map; S2, referring to hydrological manual data to obtain annual average rainfall of a debris flow basin formation area and a 10-minute rainfall variation coefficient of the debris flow basin formation area; S3, using unmanned aerial vehicle photogrammetry to investigate the average width of a channel of the debris flow basin formation area and the particle size of the debris flow basin formation area; S4, calculating a debris flow basin terrain factor; S5, calculating a debris flow basin geological factor; S6, calculating a rainfall factor inducing the debris flow; S7, calculating an occurrence index of the debris flow; and S8, judging the occurrence of the debris flow. The application can quantitatively and accurately judge the critical condition of the debris flow, and greatly improves the debris flow early warning accuracy.

Description

Technical Field

[0001] This invention relates to the field of debris flow prevention and control engineering technology, and in particular to a debris flow early warning method for strong earthquake zones based on unmanned aerial vehicle (UAV) photogrammetry. Background Technology

[0002] Domestic and international studies have shown that areas affected by strong earthquakes often experience multiple debris flows during post-earthquake rainfall, sometimes even in clusters. The scale of debris flows tends to be larger after strong earthquakes, and the inducing conditions are often more stringent, especially since debris flows can occur shortly after a strong earthquake even with minimal rainfall. This is because after a strong earthquake, numerous landslides and collapses occur in the debris flow basin, not only significantly increasing the amount of debris but also drastically reducing the particle size and narrowing the channel width. This significantly alters the original inducing conditions for debris flows, drastically lowering the critical conditions.

[0003] Currently, effective and accurate early warning methods for debris flows after strong earthquakes require extensive field investigations upstream of debris flow basins in strong earthquake zones. Only by clarifying the detailed information such as the debris source and channels of debris flows can we accurately and effectively warn of debris flow occurrences.

[0004] Chinese patent document CN109448325A, published on March 8, 2019, discloses a refined early warning method for debris flows based on one-hour rainfall. The method comprises the following steps: a) determining the area A, shape coefficient F, gully length L, and longitudinal slope J of the debris flow basin; b) obtaining the annual average rainfall R0 and the 10-minute rainfall variation coefficient Cv, and real-time monitoring of the preceding rainfall B and the rainfall I one hour before triggering the debris flow; c) determining the average gully width W and particle size D; d) calculating the topographic factor T; e) calculating the geological factor G; f) calculating the rainfall factor R; g) calculating the debris flow occurrence index P; and h) determining the occurrence of the debris flow.

[0005] The patent document discloses a refined debris flow early warning method based on one-hour rainfall. By studying the topographical and geological features and rainfall characteristics of the debris flow formation area, it uses the rainfall in the hour preceding the trigger as a key indicator, avoiding potential misjudgments in cases with shorter rainfall durations and no prior rainfall, thus achieving refined early warning. However, this method has a drawback: it is difficult to conduct on-site investigations in the upstream formation areas of debris flow basins in areas prone to strong earthquakes, as numerous landslides and collapses in these areas damage roads and make it difficult to reach the sites. If the debris flow source and channel conditions cannot be investigated before the rainy season, accurate debris flow early warnings cannot be made. In addition to the inaccessibility, the difficulty in conducting on-site investigations of the debris flow formation area's source particle size is also due to the large workload involved in manual field measurements, making it difficult to obtain parameters quickly and accurately. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention provides a debris flow early warning method for strong earthquake zones based on UAV photogrammetry. This invention studies the source and channel conditions of debris flows in strong earthquake zones and proposes to use UAV photogrammetry to obtain the source and channel conditions of debris flow formation areas. This method can quickly and accurately obtain parameters, thereby making a quantitative and accurate judgment on the critical conditions of debris flows, which greatly improves the accuracy of debris flow early warning.

[0007] This invention is achieved through the following technical solution:

[0008] A method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry, characterized by the following steps:

[0009] S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone.

[0010] S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time.

[0011] S3. Use drone photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area.

[0012] S4. Calculate the topographic factors of the debris flow basin using Equation 1;

[0013]

[0014] Where: T—topographic factor of debris flow basin;

[0015] F—Shape coefficient of the debris flow basin formation zone;

[0016] L—Length of the gully in the debris flow basin, in meters;

[0017] J—Longitudinal gradient of the gully bed in the debris flow basin formation area;

[0018] A – Area of ​​the debris flow basin, in m 2 ;

[0019] W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters;

[0020] S5. Calculate the geological factors of the debris flow basin using Equation 2;

[0021] G = D / D0 Equation 2

[0022] Where: G—geological factors of debris flow basin;

[0023] D—Particle size at the debris flow initiation point, mm;

[0024] D0—Particle size of coarse particles, D0 = 2 mm;

[0025] S6. Calculate the rainfall factor that induces debris flow using Equation 3;

[0026]

[0027] In the formula: R—the rainfall factor that induces debris flows;

[0028] R*—Rainfall induction index, mm;

[0029] B – Previous rainfall, mm;

[0030] I — Rainfall in the hour prior to triggering, in mm;

[0031] R0—Annual average rainfall in the debris flow basin formation area, mm;

[0032] C V —Coefficient of variation of rainfall over 10 minutes in the debris flow basin formation area;

[0033] S7. Calculate the occurrence index of debris flow using Equation 4;

[0034]

[0035] In the formula: P—an indicator of debris flow occurrence;

[0036] S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; and when P ≥ 0.33, the probability of debris flow is very high.

[0037] In step S1, the debris flow basin formation area refers to the area above the debris flow flow area and the debris flow deposition area.

[0038] Step S3 specifically includes:

[0039] S31. Conduct large-scale unmanned aerial vehicle (UAV) photogrammetry in areas affected by strong earthquakes to determine the locations of landslides and collapses in the investigated debris flow basins. Based on the concentrated distribution of landslide and collapse material sources caused by strong earthquakes, determine the debris flow initiation point in the debris flow formation area. Then, using topographic maps combined with UAV photogrammetry, measure the average longitudinal slope α of the channel between the safe point and the debris flow initiation point, as well as the elevation difference H0 between the safe point and the debris flow initiation point.

[0040] S32. In the safe zone downstream of the debris flow initiation point, set a safe drone flight altitude H. h =H0+50 and fly the drone for the second time along the gully to the debris flow initiation point in the upstream formation area of ​​the debris flow, measure the distance L from the safe point to the debris flow initiation point, start the preliminary photogrammetry, obtain the basic situation of the debris flow formation area through the preliminary photogrammetry, and then set the altitude for the third flight.

[0041] S33. Generate a digital orthophoto model and a digital surface model from the initial drone image data. Perform manual image measurement based on the digital model. Obtain the image grid based on the drone images. Estimate the average particle size D1 of the debris flow initiation point. Calculate the actual height Hz of the drone images.

[0042] S34. When Hz < H0 + 50, a third UAV flight photogrammetry is performed, with the altitude set at Hz. Based on the photogrammetry results and ArcGIS software, the average particle size D2 of 50 debris flow initiation points is measured. The local channel longitudinal slope β of the debris flow initiation point is measured based on the topographic map, and the minimum particle size D* is calculated.

[0043] S35. Based on the source particle size D3 of the debris flow, correct the particle size D at the initiation point of the debris flow.

[0044] In step S33, the actual altitude of the drone photography is calculated using Equation 5;

[0045] Hz=Ltanα+0.02D1×(H0+50-Ltanα) / B b Formula 5

[0046] Where: Hz—actual altitude of the drone photography, in meters;

[0047] B b —The ground resolution of the first UAV photogrammetry survey, generated by PCAS software, in mm.

[0048] In step S34, the minimum particle size D* is calculated using Equation 6;

[0049] D*=0.82D2-30 Equation 6

[0050] When D2≤50mm, the source particle size of debris flow is D3=D2.

[0051] In step S35, the particle size D at the debris flow initiation point is obtained by modifying Equation 7.

[0052] D = D³ / cosβ (Equation 7)

[0053] In the formula: β—the local longitudinal slope of the gully at the debris flow initiation point.

[0054] The PCAS software mentioned in this invention refers to particle and crack image recognition and analysis software.

[0055] The ArcGIS software mentioned in this invention refers to geographic information system software.

[0056] The basic principle of this invention is as follows:

[0057] The reason for the sharp decrease in the critical rainfall conditions for debris flows in strong earthquake zones is the rapid increase in debris source, the sharp decrease in particle size, and the sharp narrowing of channels. After a strong earthquake, it is necessary to assess the watersheds where debris flows may occur as quickly as possible. However, due to road interruptions caused by landslides and collapses, a feasible method is to conduct drone photogrammetry surveys of the debris flow watershed. Because there are too many landslides and collapses in strong earthquake zones, and the area is too large, drones are the fastest way to comprehensively understand the landslide and collapse situation. Therefore, large drone flights over a wide area can provide a comprehensive overview. However, due to the high flight altitude, the particle size that can be resolved in the images is within the range of 0.5-1m, which is insufficient for the rapidly decreasing particle source in strong earthquake zones. A second flight is needed. The second flight uses a small drone to fly within a smaller watershed area. However, the vegetation in the smaller watershed is relatively good and at a higher altitude. To ensure the safety of the drone, a safe altitude needs to be set based on the elevation difference between the safe zone and the debris flow initiation point. If the set safe altitude is 50m, then in some small watersheds, the particle size of the debris flow source is too small, and the 50m altitude photogrammetry cannot achieve sufficient accuracy. Therefore, a third flight is required, that is, to reduce the flight altitude and measure again, so as to finally achieve the goal of high-precision measurement of the particle size of the debris flow source.

[0058] The beneficial effects of this invention are mainly reflected in the following aspects:

[0059] 1. This invention studies the source material and channel conditions of debris flows in areas of strong earthquakes, and proposes to use unmanned aerial vehicle (UAV) photogrammetry to obtain the source material and channel conditions of debris flow formation areas. This method can quickly and accurately obtain parameters, thereby making a quantitative and accurate judgment on the critical conditions of debris flows, which greatly improves the accuracy of debris flow early warning.

[0060] 2. This invention utilizes the characteristics of debris flow development, source material, and channel features in strong earthquake zones to obtain the source particle size and channel width characteristics of debris flows through photogrammetry using unmanned aerial vehicles (UAVs). This overcomes the transportation difficulties in strong earthquake zones and enables early warning of debris flows in these areas.

[0061] 3. This invention uses large-scale drone aerial photogrammetry to obtain information on landslide and collapse sources in a large area of ​​strong earthquake zones, identify debris flow basins with significant changes, and preliminarily determine debris flow basins with greater risks. This ensures that subsequent assessments will not miss important debris flow basins, nor will they waste human and material resources by blindly investigating all small watersheds in strong earthquake zones, thus improving early warning efficiency.

[0062] 4. This invention utilizes small drones to conduct targeted drone surveys of small areas and watersheds. A safe altitude is set to ensure the safety of the drones. The lower flight altitude of small drones allows for measurement accuracy that large drones cannot achieve. This enables photogrammetry to meet the special conditions of small particle size of debris flow sources in strong earthquake zones, resulting in higher and more accurate early warning.

[0063] 5. In this invention, the second flight uses a flight altitude of 50m, which can provide images for the measurement of larger particle sizes. However, in some strong earthquake zones, the particle size of debris flow sources is too small, so it is necessary to reduce the altitude and fly a third time to ensure that the accuracy of photogrammetry can meet the special conditions of strong earthquake zones. This allows the invention to obtain sufficient measurement accuracy and make debris flow early warning more accurate.

[0064] 6. This invention uses drones to overcome the difficulties of impassable roads in mountainous areas of strong earthquake zones, enabling rapid measurement of debris flow sources and channel widths, and rapid assessment and early warning of debris flows.

[0065] 7. This invention uses drone photogrammetry to warn of debris flows, which can detect potential debris flow areas and issue warnings in a short time, especially in mountainous areas where strong earthquakes occur during the rainy season, thus improving the disaster prevention and mitigation effect in strong earthquake zones. Detailed Implementation

[0066] Example 1

[0067] A method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry includes the following steps:

[0068] S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone.

[0069] S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time.

[0070] S3. Use drone photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area.

[0071] S4. Calculate the topographic factors of the debris flow basin using Equation 1;

[0072]

[0073] Where: T—topographic factor of debris flow basin;

[0074] F—Shape coefficient of the debris flow basin formation zone;

[0075] L—Length of the gully in the debris flow basin, in meters;

[0076] J—Longitudinal gradient of the gully bed in the debris flow basin formation area;

[0077] A – Area of ​​the debris flow basin, in m 2 ;

[0078] W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters;

[0079] S5. Calculate the geological factors of the debris flow basin using Equation 2;

[0080] G = D / D0 Equation 2

[0081] Where: G—geological factors of debris flow basin;

[0082] D—Particle size at the debris flow initiation point, mm;

[0083] D0—Particle size of coarse particles, D0 = 2 mm;

[0084] S6. Calculate the rainfall factor that induces debris flow using Equation 3;

[0085]

[0086] In the formula: R—the rainfall factor that induces debris flows;

[0087] R*—Rainfall induction index, mm;

[0088] B – Previous rainfall, mm;

[0089] I — Rainfall in the hour prior to triggering, in mm;

[0090] R0—Annual average rainfall in the debris flow basin formation area, mm;

[0091] C V —Coefficient of variation of rainfall over 10 minutes in the debris flow basin formation area;

[0092] S7. Calculate the occurrence index of debris flow using Equation 4;

[0093]

[0094] In the formula: P—an indicator of debris flow occurrence;

[0095] S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; and when P ≥ 0.33, the probability of debris flow is very high.

[0096] This embodiment is the most basic implementation method, which studies the source and channel conditions of debris flows in strong earthquake zones. It proposes to use UAV photogrammetry to obtain the source and channel conditions of debris flow formation areas, which can quickly and accurately obtain parameters, thereby making a quantitative and accurate judgment on the critical conditions of debris flows, and greatly improving the accuracy of debris flow early warning.

[0097] Example 2

[0098] A method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry includes the following steps:

[0099] S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone.

[0100] S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time.

[0101] S3. Use UAV photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area.

[0102] S4. Calculate the topographic factors of the debris flow basin using Equation 1;

[0103]

[0104] Where: T—topographic factor of debris flow basin;

[0105] F—Shape coefficient of the debris flow basin formation zone;

[0106] L—Length of the gully in the debris flow basin, in meters;

[0107] J—Longitudinal gradient of the gully bed in the debris flow basin formation area;

[0108] A – Area of ​​the debris flow basin, in m 2 ;

[0109] W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters;

[0110] S5. Calculate the geological factors of the debris flow basin using Equation 2;

[0111] G = D / D0 Equation 2

[0112] Where: G—geological factors of debris flow basin;

[0113] D—Particle size at the debris flow initiation point, mm;

[0114] D0—Particle size of coarse particles, D0 = 2 mm;

[0115] S6. Calculate the rainfall factor that induces debris flow using Equation 3;

[0116]

[0117] In the formula: R—the rainfall factor that induces debris flows;

[0118] R*—Rainfall induction index, mm;

[0119] B – Previous rainfall, mm;

[0120] I — Rainfall in the hour prior to triggering, in mm;

[0121] R0—Annual average rainfall in the debris flow basin formation area, mm;

[0122] C V —Coefficient of variation of rainfall over 10 minutes in the debris flow basin formation area;

[0123] S7. Calculate the occurrence index of debris flow using Equation 4;

[0124]

[0125] In the formula: P—an indicator of debris flow occurrence;

[0126] S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; and when P ≥ 0.33, the probability of debris flow is very high.

[0127] Furthermore, in step S1, the debris flow basin formation area refers to the area above the debris flow flow area and the debris flow deposition area.

[0128] This embodiment is a preferred implementation method. By using unmanned aerial vehicles (UAVs) to obtain the characteristics of debris flow development, source material, and channel features in strong earthquake zones through photogrammetry, the particle size of debris flow sources and channel width characteristics are obtained. This overcomes the transportation difficulties in strong earthquake zones and enables early warning of debris flows in strong earthquake zones.

[0129] Example 3

[0130] A method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry includes the following steps:

[0131] S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone.

[0132] S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time.

[0133] S3. Use UAV photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area.

[0134] S4. Calculate the topographic factors of the debris flow basin using Equation 1;

[0135]

[0136] Where: T—topographic factor of debris flow basin;

[0137] F—Shape coefficient of the debris flow basin formation zone;

[0138] L—Length of the gully in the debris flow basin, in meters;

[0139] J—Longitudinal gradient of the gully bed in the debris flow basin formation area;

[0140] A – Area of ​​the debris flow basin, in m 2 ;

[0141] W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters;

[0142] S5. Calculate the geological factors of the debris flow basin using Equation 2;

[0143] G = D / D0 Equation 2

[0144] Where: G—geological factors of debris flow basin;

[0145] D—Particle size at the debris flow initiation point, mm;

[0146] D0—Particle size of coarse particles, D0 = 2 mm;

[0147] S6. Calculate the rainfall factor that induces debris flow using Equation 3;

[0148]

[0149] In the formula: R—the rainfall factor that induces debris flows;

[0150] R*—Rainfall induction index, mm;

[0151] B – Previous rainfall, mm;

[0152] I — Rainfall in the hour prior to triggering, in mm;

[0153] R0—Annual average rainfall in the debris flow basin formation area, mm;

[0154] C V —Coefficient of variation of rainfall over 10 minutes in the debris flow basin formation area;

[0155] S7. Calculate the occurrence index of debris flow using Equation 4;

[0156]

[0157] In the formula: P—an indicator of debris flow occurrence;

[0158] S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; and when P ≥ 0.33, the probability of debris flow is very high.

[0159] In step S1, the debris flow basin formation area refers to the area above the debris flow flow area and the debris flow deposition area.

[0160] Furthermore, step S3 specifically includes:

[0161] S31. Conduct large-scale unmanned aerial vehicle (UAV) photogrammetry in areas affected by strong earthquakes to determine the locations of landslides and collapses in the investigated debris flow basins. Based on the concentrated distribution points of landslide and collapse material sources caused by strong earthquakes, determine the debris flow initiation points in the debris flow formation areas. Then, using topographic maps combined with UAV photogrammetry, measure the average longitudinal slope α of the channel between the safe point and the debris flow initiation point, as well as the elevation difference H0 between the safe point and the debris flow initiation point.

[0162] S32. In the safe zone downstream of the debris flow initiation point, set a safe drone flight altitude H. h=H0+50 and fly the drone for the second time along the gully to the debris flow initiation point in the upstream formation area of ​​the debris flow, measure the distance L from the safe point to the debris flow initiation point, start the preliminary photogrammetry, obtain the basic situation of the debris flow formation area through the preliminary photogrammetry, and then set the altitude for the third flight.

[0163] S33. Generate a digital orthophoto model and a digital surface model from the initial drone image data. Perform manual image measurement based on the digital model. Obtain the image grid based on the drone images. Estimate the average particle size D1 of the debris flow initiation point. Calculate the actual height Hz of the drone images.

[0164] S34. When Hz < H0 + 50, a third UAV flight photogrammetry is performed, with the altitude set at Hz. Based on the photogrammetry results and ArcGIS software, the average particle size D2 of 50 debris flow initiation points is measured. The local channel longitudinal slope β of the debris flow initiation point is measured based on the topographic map, and the minimum particle size D* is calculated.

[0165] S35. Based on the source particle size D3 of the debris flow, correct the particle size D at the initiation point of the debris flow.

[0166] This embodiment is another preferred implementation method. By using large-scale drone aerial photogrammetry, the landslide and collapse material sources in a large area of ​​strong earthquake zone are obtained, debris flow basins with significant changes are identified, and debris flow basins with greater risks are preliminarily identified. This ensures that subsequent judgments will not miss important debris flow basins, nor will they waste human and material resources by blindly investigating all small watersheds in the strong earthquake zone, thus improving the efficiency of early warning.

[0167] By using small drones to conduct targeted drone surveys of small areas and watersheds, and setting a safe altitude to ensure the safety of the drones, the lower flight altitude of small drones can achieve measurement accuracy that large drones cannot obtain. This allows photogrammetry to meet the special conditions of small particle size of debris flow sources in strong earthquake zones, resulting in higher accuracy and more precise early warning.

[0168] The second flight, at an altitude of 50m, can provide images for the measurement of larger particle sizes. However, in some strong earthquake zones, the particle size of debris flow sources is too small, so it is necessary to lower the altitude and fly a third time to ensure that the accuracy of photogrammetry can meet the special conditions of strong earthquake zones. This allows the invention to obtain sufficient measurement accuracy and make debris flow early warning more accurate.

[0169] Example 4

[0170] A method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry includes the following steps:

[0171] S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone.

[0172] S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time.

[0173] S3. Use drone photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area.

[0174] S4. Calculate the topographic factors of the debris flow basin using Equation 1;

[0175]

[0176] Where: T—topographic factor of debris flow basin;

[0177] F—Shape coefficient of the debris flow basin formation zone;

[0178] L—Length of the gully in the debris flow basin, in meters;

[0179] J—Longitudinal gradient of the gully bed in the debris flow basin formation area;

[0180] A – Area of ​​the debris flow basin, in m 2 ;

[0181] W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters;

[0182] S5. Calculate the geological factors of the debris flow basin using Equation 2;

[0183] G = D / D0 Equation 2

[0184] Where: G—geological factors of debris flow basin;

[0185] D—Particle size at the debris flow initiation point, mm;

[0186] D0—Particle size of coarse particles, D0 = 2 mm;

[0187] S6. Calculate the rainfall factor that induces debris flow using Equation 3;

[0188]

[0189] In the formula: R—the rainfall factor that induces debris flows;

[0190] R*—Rainfall induction index, mm;

[0191] B – Previous rainfall, mm;

[0192] I — Rainfall in the hour prior to triggering, in mm;

[0193] R0—Annual average rainfall in the debris flow basin formation area, mm;

[0194] C V —Coefficient of variation of rainfall over 10 minutes in the debris flow basin formation area;

[0195] S7. Calculate the occurrence index of debris flow using Equation 4;

[0196]

[0197] In the formula: P—an indicator of debris flow occurrence;

[0198] S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; and when P ≥ 0.33, the probability of debris flow is very high.

[0199] In step S1, the debris flow basin formation area refers to the area above the debris flow flow area and the debris flow deposition area.

[0200] Step S3 specifically includes:

[0201] S31. Conduct large-scale unmanned aerial vehicle (UAV) photogrammetry in areas affected by strong earthquakes to determine the locations of landslides and collapses in the investigated debris flow basins. Based on the concentrated distribution points of landslide and collapse material sources caused by strong earthquakes, determine the debris flow initiation points in the debris flow formation areas. Then, using topographic maps combined with UAV photogrammetry, measure the average longitudinal slope α of the channel between the safe point and the debris flow initiation point, as well as the elevation difference H0 between the safe point and the debris flow initiation point.

[0202] S32. In the safe zone downstream of the debris flow initiation point, set a safe drone flight altitude H. h =H0+50 and fly the drone for the second time along the gully to the debris flow initiation point in the upstream formation area of ​​the debris flow, measure the distance L from the safe point to the debris flow initiation point, start the preliminary photogrammetry, obtain the basic situation of the debris flow formation area through the preliminary photogrammetry, and then set the altitude for the third flight.

[0203] S33. Generate a digital orthophoto model and a digital surface model from the initial drone image data. Perform manual image measurement based on the digital model. Obtain the image grid based on the drone images. Estimate the average particle size D1 of the debris flow initiation point. Calculate the actual height Hz of the drone images.

[0204] S34. When Hz < H0 + 50, a third UAV flight photogrammetry is performed, with the altitude set at Hz. Based on the photogrammetry results and ArcGIS software, the average particle size D2 of 50 debris flow initiation points is measured. The local channel longitudinal slope β of the debris flow initiation point is measured based on the topographic map, and the minimum particle size D* is calculated.

[0205] S35. Based on the source particle size D3 of the debris flow, correct the particle size D at the initiation point of the debris flow.

[0206] In step S33, the actual altitude of the drone photography is calculated using Equation 5;

[0207] Hz=Ltanα+0.02D1×(H0+50-Ltanα) / B b Formula 5

[0208] Where: Hz—actual altitude of the drone photography, in meters;

[0209] B b —The ground resolution of the first UAV photogrammetry survey, generated by PCAS software, in mm.

[0210] In step S34, the minimum particle size D* is calculated using Equation 6;

[0211] D*=0.82D2-30 Equation 6

[0212] When D2≤50mm, the source particle size of debris flow is D3=D2.

[0213] The average particle size above the minimum particle size D* is obtained based on photogrammetry results and ArcGIS software measurements, which is the source particle size D3 of the debris flow; that is, the particle size measured by ArcGIS software is the particle size after removing fine particles smaller than the minimum particle size D*.

[0214] In step S35, the particle size D at the debris flow initiation point is obtained by modifying Equation 7.

[0215] D = D³ / cosβ (Equation 7)

[0216] In the formula: β—the local longitudinal slope of the gully at the debris flow initiation point.

[0217] This embodiment represents the optimal implementation method. By using drones, the difficulties of impassable roads in mountainous areas of strong earthquake zones can be overcome. Drones can quickly measure debris flow sources and channel widths, enabling rapid assessment and early warning of debris flows.

[0218] Using drone photogrammetry to warn of debris flows can quickly identify potential debris flow areas and issue warnings, especially in mountainous areas prone to strong earthquakes during the rainy season, thus improving disaster prevention and mitigation in earthquake-prone areas.

[0219] The embodiments of the present invention will be described in detail below with reference to specific examples:

[0220] No mudslides occurred in Luojingou and Liangchahegou, the areas affected by the devastating Luding earthquake on September 5, 2022, during the two rainstorms that occurred on June 20 and July 31, 2013, prior to the earthquake. Following the Luding earthquake, large unmanned aerial vehicles (UAVs) conducted extensive UAV photogrammetry surveys of the affected area, discovering large amounts of landslide debris in both Luojingou and Liangchahegou. However, the debris was small in size, making the measurements too inaccurate for large UAVs. Subsequently, smaller UAVs were used to conduct UAV photogrammetry surveys in the two smaller watersheds.

[0221] First, the area A, shape coefficient F, channel length L, and longitudinal slope J of the debris flow formation zone in each debris flow basin were measured using topographic maps. The average channel width W and particle size D at the debris flow initiation point were measured twice using drones, and the topographic factors T and G were calculated. The annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow formation zone were obtained by consulting hydrological handbooks. The antecedent rainfall and the rainfall one hour before initiation were obtained based on actual monitoring, and the rainfall index R* and rainfall factor R were calculated. Finally, the occurrence index of debris flow was calculated.

[0222] Based on historical rainfall data from two rainstorm events, "June 20, 2013" and "July 31, 2013", the debris flow parameters and debris flow occurrence index P of the two watersheds after the strong earthquake were analyzed to determine the probability of debris flow occurrence, as shown in Tables 1 and 2. Table 1 is a table of topographic and geological factors of debris flow after the strong earthquake, and Table 2 is a table of debris flow warning values ​​in the strong earthquake zone.

[0223] Table 1

[0224] Luojingou 3220000 0.287 0.192 84 1.43 0.956 42 Two-forked River Ditch 13440000 0.361 0.187 151.3 4.4 0.995 75.65

[0225] Table 2

[0226] 2013.6.20 40.1 32.6 447.6 0.901 0.216 Luojingou 2013.6.20 40.1 32.6 447.6 0.901 0.174 Two-forked River Ditch 2013.7.13 59.6 37.4 527.1 1.061 0.254 Luojingou 2013.7.13 59.6 37.4 527.1 1.061 0.205 Two-forked River Ditch

[0227] When P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; when P ≥ 0.33, the probability of debris flow is very high.

[0228] Table 2 shows that in Luojingou, following the Luding earthquake on September 5, 2022, the probability of debris flows occurring under the conditions of two heavy rainfall events, "June 20, 2013" and "July 31, 2013," was medium and high, respectively. Similarly, in Liangchahegou, the probability of debris flows occurring under the same conditions was low and medium. This indicates that the critical point for debris flows in the basin decreased after a strong earthquake, and the probability of debris flows increased in both cases.

[0229] In summary, the method of this invention can provide rapid and accurate early warning of debris flows in areas prone to severe earthquakes.

Claims

1. A method for early warning of debris flows in strong earthquake zones based on unmanned aerial vehicle (UAV) photogrammetry, characterized in that, Includes the following steps: S1. Determine the basic parameters of potential debris flow basins through topographic maps, including the area of ​​the debris flow basin formation zone, the shape coefficient of the debris flow basin formation zone, the length of the gully in the debris flow basin formation zone, and the longitudinal slope of the gully bed in the debris flow basin formation zone. S2. Consult hydrological manuals to obtain the annual average rainfall and the 10-minute variation coefficient of rainfall in the debris flow basin formation area, and monitor or forecast the previous rainfall and the rainfall 1 hour before the triggering of the debris flow basin formation area in real time. S3. Use drone photogrammetry to investigate the average width of channels and particle size in the debris flow basin formation area. S4. Calculate the topographic factors of the debris flow basin using Equation 1; Formula 1 Where: T—topographic factor of debris flow basin; F—Shape coefficient of the debris flow basin formation zone; L—Length of the gully in the debris flow basin, in meters; J—Longitudinal gradient of the gully bed in the debris flow basin formation area; A – Area of ​​the debris flow basin, in m 2 ; W—Average width of channels in the debris flow basin formation area, measured by UAV, in meters; S5. Calculate the geological factors of the debris flow basin using Equation 2; Formula 2 Where: G—geological factors of debris flow basin; D—Particle size at the debris flow initiation point, mm; S6. Calculate the rainfall factor that induces debris flow using Equation 3; Formula 3 In the formula: R—the rainfall factor that induces debris flows; S7. Calculate the occurrence index of debris flow using Equation 4; Formula 4 In the formula: P—an indicator of debris flow occurrence; S8. To determine the occurrence of debris flow, when P < 0.19, the probability of debris flow is low; when 0.24 > P ≥ 0.19, the probability of debris flow is moderate; when 0.33 > P ≥ 0.24, the probability of debris flow is high; when P ≥ 0.33, the probability of debris flow is very high. Step S3 specifically includes: S31. Conduct large-scale unmanned aerial vehicle (UAV) photogrammetry in areas affected by strong earthquakes to determine the locations of landslides and collapses in the surveyed debris flow basins. Based on the concentrated distribution points of landslide and collapse debris sources caused by strong earthquakes, determine the debris flow initiation points in the debris flow formation areas. Then, using topographic maps combined with UAV photogrammetry, measure the average longitudinal slope of the gullies between safe points and debris flow initiation points. and the elevation difference between the safe point and the debris flow initiation point ; S32. In the safe zone downstream of the debris flow initiation point, set a safe flight altitude for the drone. The drone then flew a second time along the gully to the debris flow initiation point in the upstream formation area of ​​the debris flow, measured the distance L from the safe point to the debris flow initiation point, and began preliminary photogrammetry to obtain the basic situation of the debris flow formation area. Then, the drone set the altitude for the third flight. S33. Generate a digital orthophoto model and a digital surface model from the preliminary UAV image data. Perform manual image measurement based on the digital model. Obtain the image grid based on the UAV images and estimate the average particle size of the debris flow initiation point. Calculate the actual altitude (Hz) of the drone photography; S34, when Then, a third UAV aerial photogrammetry was conducted, with the altitude set at Hz. Based on the photogrammetry results and ArcGIS software, the average particle size of 50 debris flow initiation points was measured. The local longitudinal slope of the gully at the debris flow initiation point was measured based on the topographic map. Calculate the minimum particle size ; S35. Based on the source particle size of debris flow Correcting the particle size at the debris flow initiation point .

2. The method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry according to claim 1, characterized in that: In step S1, the debris flow basin formation area refers to the area above the debris flow flow area and the debris flow deposition area.

3. The method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry according to claim 1, characterized in that: In step S33, the actual altitude of the drone photography is calculated using Equation 5; Formula 5 Where: Hz—actual altitude of the drone photography, in meters; The ground resolution of the first UAV photogrammetry survey was generated in mm by PCAS software.

4. The method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry according to claim 1, characterized in that: In step S34, the minimum particle size Calculated using Equation 6; Formula 6 5. The method for early warning of debris flows in strong earthquake zones based on UAV photogrammetry according to claim 1, characterized in that: In step S35, the particle size D at the debris flow initiation point is obtained by modifying Equation 7. Formula 7 In the formula: —The longitudinal slope of the local gully at the point where the debris flow originates.

Images