A method for dynamic early warning of gully-type debris flow induced by shallow landslide

Abstract

The application discloses a kind of shallow landslide induced gully type debris flow dynamic early warning method, belong to landslide prevention engineering technical field, it is characterized in that, comprising the following steps: a, determine the area of debris flow basin;B, calculate the depth of grid soil layer;C, real-time monitoring or forecasting the real-time rainfall data of the position where debris flow is located;D, the stability coefficient Fs of slope stability state is output under different rainfall duration by TRIGRS model;E, calculate the landslide volume of forecast basin;F, the landslide source volume into channel in basin is calculated in combination with material source coefficient;G, calculate the occurrence index of debris flow, and judge the possibility of debris flow occurrence;The application can make quantitative judgment by TRIGRS physical model to judge the possibility of landslide, reduce error, improve the accuracy of debris flow judgment, guarantee early warning effect.

Description

Technical Field

[0001] This invention relates to the field of landslide prevention engineering technology, and in particular to a dynamic early warning method for gully-type debris flows induced by shallow landslides. Background Technology

[0002] Landslides and debris flows triggered by heavy rainfall in mountainous areas are sometimes not independent, but rather occur in combination. Valley-type debris flows induced by shallow landslides are characterized by their suddenness and destructive power.

[0003] The formation mechanism of shallow landslide-induced gully debris flows mainly involves prolonged and high-intensity rainfall leading to multiple shallow landslides within the watershed, resulting in the deposition of large amounts of sediment in the gullies. Subsequently, a sufficiently large flood triggers the landslide material deposited in the gullies, leading to debris flows. Domestic and international research indicates that areas prone to cluster landslides are more likely to experience shallow landslide-induced gully debris flow events, sometimes even clustered gully debris flows. Watersheds that would not normally experience gully debris flows may develop into them after multiple landslides; watersheds that might normally experience gully debris flows may become more dangerous due to multiple landslides; and watersheds that were originally debris flow gullies may experience larger-scale debris flows when multiple landslides provide sediment. Currently, there are no dynamic early warning methods for shallow landslide-induced gully debris flows, either domestically or internationally; only a limited number of quantitative assessment methods exist. The more loose material sources provided by landslides within a watershed, the greater the likelihood of gully debris flows occurring. The solid particles in shallow landslide-induced gully debris flows mainly originate from landslides occurring during rainfall; without a large number of shallow landslides, this type of debris flow will not occur.

[0004] Currently, there is a lack of quantitative early warning studies on shallow landslide-induced gully debris flows, particularly regarding the quantitative impact of the dynamic sediment supply from shallow landslides during rainfall on gully debris flows. Most studies rely on a combination of rainfall and topographic indicators to predict debris flow occurrence, which are indirect indicators. The primary influencing factor for shallow landslide-induced gully debris flows is the amount of sediment supplied. More importantly, most studies on the sediment supply for shallow landslide-induced gully debris flows need to be completed after the event, making it impossible to predict the spatial distribution of the landslide and the subsequent dynamic changes of the gully debris flow in advance.

[0005] Chinese patent document CN108776851A, published on November 9, 2018, discloses a method for determining the early warning threshold of shallow landslide disasters induced by rainstorms, characterized by the following steps:

[0006] Step 1: Collect topographic data of the study area, establish a high-precision digital elevation model (DEM), and divide the slope units based on the DEM;

[0007] Step 2: Based on the slope response mechanism to rainstorms, a large porosity coefficient is introduced. By combining the topographic humidity index of the steady-state hydrological model with the infinite slope model, and assuming that the slope is in a specific warning level state, the calculation formulas of the topographic humidity index and the slope stability index are combined to establish a stability analysis and early warning model based on slope units and critical infiltration.

[0008] Step 3: Set five warning levels, namely I, II, III, IV and V, with corresponding warning signals of red, orange, yellow, blue and green. Based on the meaning of the FS index, infer the slope stability state corresponding to the five levels and calculate the critical infiltration threshold for each warning level.

[0009] Step 4: By analyzing the response mechanism of the slope to heavy rain, establish a landslide rainfall-infiltration model based on the slope unit, calculate the relationship between the infiltration amount and the rainfall amount of each slope unit under specific rainfall conditions, and inversely deduce the critical rainfall amount of each slope unit through the critical infiltration amount.

[0010] The patent document discloses a method for determining the early warning threshold of rainstorm-induced shallow landslides. This method incorporates slope units and rainstorm influence coefficients into the measurement and modeling of rainstorm-induced shallow landslides, providing fundamental data for landslide prevention research. However, it cannot make quantitative judgments, which can easily lead to errors and affect the accuracy of debris flow assessments. Summary of the Invention

[0011] To overcome the shortcomings of the prior art, this invention provides a dynamic early warning method for gully-type debris flows induced by shallow landslides. This invention uses the TRIGRS physical model to determine the probability of landslide occurrence, enabling quantitative judgment, reducing errors, improving the accuracy of debris flow judgment, and ensuring the effectiveness of early warning.

[0012] This invention is achieved through the following technical solution:

[0013] A dynamic early warning method for gully-type debris flows induced by shallow landslides, characterized by comprising the following steps:

[0014] a. Determine the potential debris flow basin area using topographic maps;

[0015] b. Determine the physical and mechanical properties of the soil and rock mass through field surveys and indoor experiments, investigate the soil thickness in the area where the debris flow basin is located, and calculate the depth of the grid soil layer using Equation 1.

[0016]

[0017] In the formula, Z i h represents the depth of the grid soil layer. iZ represents the grid elevation. max Z represents the maximum soil thickness in the study area. min h represents the minimum soil thickness in the study area. max h is the highest elevation in the study area min This represents the lowest elevation in the study area;

[0018] c. Real-time monitoring or forecasting of real-time rainfall data at the location of debris flows;

[0019] d. Input real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model. The TRIGRS model outputs the stability coefficient Fs, which characterizes the slope stability state under different rainfall durations. Inputting into the TRIGRS model means inputting through Equation 2.

[0020]

[0021] In the formula, Fs(Z, t) is the slope stability safety factor, Z is the soil depth of any grid, t is time, θ is the soil internal friction angle, c is the cohesion, and γ is the soil internal friction angle. s For soil bulk density, γ w Let Fs be the groundwater unit weight, a be the stability coefficient, and Ψ(Z,t) be the pressure head, which is a function of depth and time. It is a term related to the effect of gravity. It is a term related to cohesion. It is a term related to water pressure;

[0022] e. Calculate the predicted landslide volume in the watershed;

[0023] V n =A n h Formula 3

[0024] A z =A1+A2+A3+...+A n Formula 4

[0025] V z =A z h-type 5

[0026] In the formula, V z V represents the total volume of landslides within the watershed. n Let A be the landslide volume. n Let A be the area of ​​the nth landslide. Based on the fact that Fs < 1 represents the landslide area, A... z denoted as the total landslide area within the watershed, h as the average soil layer thickness in the study area, A1 as the area of ​​the first landslide, A2 as the area of ​​the second landslide, and A3 as the area of ​​the third landslide.

[0027] f. Combining the source coefficient, calculate the volume of landslide source material entering the gully within the watershed using Equation 6;

[0028] V Lz =(A1r1+A2r2+......+A n r n )h Formula 6

[0029] In the formula, V Lz Let r1 be the source volume of landslides entering the gully within the watershed, r2 be the source coefficient of the first landslide, and r3 be the source coefficient of the second landslide. n Let be the source coefficient of the nth landslide;

[0030] g. Calculate the occurrence index of debris flow using Equation 7, and determine the probability of debris flow occurrence;

[0031]

[0032] In the formula, P is the occurrence index of debris flow, and A is the area of ​​debris flow basin.

[0033] In step a, the debris flow basin area refers to the area of ​​all catchment areas other than the debris flow deposition area.

[0034] In step b, the physical and mechanical properties of the soil and rock include cohesion c, soil internal friction angle θ, and soil unit weight γ. s saturated hydraulic conductivity K S Diffusion rate D0, saturated soil moisture content θ s and residual soil moisture content θ r .

[0035] In step d, the stability state includes five categories: Fs≤0.85, 0.85<Fs≤1, 1<Fs≤1.15, 1.15<Fs≤1.3, and 1.3<Fs. Fs<1 is used as the classification criterion for landslide occurrence.

[0036] In step f, the source coefficient is determined by the ratio of the elevation difference H between the landslide and the gully to the distance L between the landslide and the gully. When H / L≥1, the source coefficient is 0.8; when 1>H / L≥0.5, the source coefficient is 0.5; when 0.5>H / L≥0.3, the source coefficient is 0.2; and when H / L<0.3, the source coefficient is 0.

[0037] In step g, determining the probability of a debris flow specifically means that when P < 0.045, the probability of a debris flow is low; when 0.045 ≤ P < 0.09, the probability of a debris flow is moderate; and when 0.09 ≤ P, the probability of a debris flow is high.

[0038] The TRIGRS model described in this invention refers to a regional slope stability calculation model for transient rainfall infiltration.

[0039] The basic principle of this invention is as follows:

[0040] Shallow landslide-induced gully-type debris flows occur when prolonged rainfall causes multiple landslides within a watershed, accumulating material in the gully. Subsequent flash floods then initiate the flow of this loose solid material, forming a debris flow. If the probability of occurrence is too low (e.g., P < 0.045), the volume of shallow landslides entering the gully due to rainfall is insufficient, providing enough material for the debris flow and making its formation unlikely; conversely, the probability is high. The reasons for a low probability of debris flow occurrence are: firstly, the volume of material provided by landslides within the watershed is too small. This means that if the volume of landslides within the watershed is small, the material source is insufficient to form a debris flow, resulting mostly in flash floods; secondly, the volume of landslides entering the gully within the watershed is too small, resulting in insufficient material to form a debris flow.

[0041] The beneficial effects of this invention are mainly reflected in the following aspects:

[0042] 1. This invention: a) determines the potential debris flow basin area using topographic maps; b) determines the physical and mechanical properties of soil and rock through field surveys and laboratory experiments, investigates the soil thickness in the predicted debris flow basin area, and calculates the depth of the grid soil layer; c) monitors or forecasts real-time rainfall data at the location of debris flows; d) inputs real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model, and outputs the stability coefficient Fs characterizing the slope stability state under different rainfall durations; e) calculates the predicted landslide volume in the basin; f) calculates the landslide source volume entering the gully within the basin by combining the source coefficient; g) calculates the debris flow occurrence index and judges the probability of debris flow occurrence. Compared with existing technologies, judging the probability of landslide occurrence through the TRIGRS physical model can make quantitative judgments, reduce errors, improve the accuracy of debris flow judgment, and ensure the early warning effect.

[0043] 2. This invention addresses the fact that landslides on both sides of a gully may not necessarily move and accumulate in the gully to become debris flow sources. The H / L ratio can be used to determine the source of landslides entering the gully, providing a reasonable debris flow source calculation index, which helps improve the accuracy of debris flow judgment.

[0044] 3. This invention, by studying the relationship between the material source and water source of debris flows and combining the analysis of the two, can better reduce errors and improve the early warning effect.

[0045] 4. This invention, by studying the relationship between landslides and debris flow sources, gives the total debris flow source quantity of all landslides in the watershed. By studying the dynamic changes of debris flow sources, it is possible to more effectively judge the dynamic changes in debris flow development.

[0046] 5. This invention uses the amount of material source as a prediction indicator, which can more directly and effectively predict shallow landslide-type debris flows in gullies.

[0047] 6. This invention, by studying the influencing factors of shallow landslides inducing gully-type debris flows, provides calculations of source indicators that directly affect debris flows, thereby enabling more direct and effective early warning and judgment of debris flows. Attached Figure Description

[0048] The present invention will now be further described in detail with reference to the accompanying drawings and specific embodiments:

[0049] Figure 1 This is a diagram showing the relationship between the proportion of unstable slopes and rainfall in a specific embodiment of the present invention;

[0050] Figure 2 This is a time-by-time ratio trend chart in a specific embodiment of the present invention. Detailed Implementation

[0051] Example 1

[0052] A dynamic early warning method for gully-type debris flows induced by shallow landslides includes the following steps:

[0053] a. Determine the potential debris flow basin area using topographic maps;

[0054] b. Determine the physical and mechanical properties of the soil and rock mass through field surveys and indoor experiments, investigate the soil thickness in the area where the debris flow basin is located, and calculate the depth of the grid soil layer using Equation 1.

[0055]

[0056] In the formula, Z i h represents the depth of the grid soil layer. i Z represents the grid elevation. max Z represents the maximum soil thickness in the study area. min h represents the minimum soil thickness in the study area. max h is the highest elevation in the study area min This represents the lowest elevation in the study area;

[0057] c. Real-time monitoring or forecasting of real-time rainfall data at the location of debris flows;

[0058] d. Input real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model. The TRIGRS model outputs the stability coefficient Fs, which characterizes the slope stability state under different rainfall durations. Inputting into the TRIGRS model means inputting through Equation 2.

[0059]

[0060] In the formula, Fs(Z, t) is the slope stability safety factor, Z is the soil depth of any grid, t is time, θ is the soil internal friction angle, c is the cohesion, and γ is the soil internal friction angle. s For soil bulk density, γ w Let Fs be the groundwater unit weight, a be the stability coefficient, and Ψ(Z,t) be the pressure head, which is a function of depth and time. It is a term related to the effect of gravity. It is a term related to cohesion. It is a term related to water pressure;

[0061] e. Calculate the predicted landslide volume in the watershed;

[0062] V n =A n h Formula 3

[0063] A z =A1+A2+A3……+A n Formula 4

[0064] V z =A z h-type 5

[0065] In the formula, V z V represents the total volume of landslides within the watershed. n Let A be the landslide volume. n Let A be the area of ​​the nth landslide. Based on the fact that Fs < 1 represents the landslide area, A... z denoted as the total landslide area within the watershed, h as the average soil layer thickness in the study area, A1 as the area of ​​the first landslide, A2 as the area of ​​the second landslide, and A3 as the area of ​​the third landslide.

[0066] f. Combining the source coefficient, calculate the volume of landslide source material entering the gully within the watershed using Equation 6;

[0067] V Lz =(A1r1+A2r2+.....+A n r n )h Formula 6

[0068] In the formula, V LzLet r1 be the source volume of landslides entering the gully within the watershed, r2 be the source coefficient of the first landslide, and r3 be the source coefficient of the second landslide. n Let be the source coefficient of the nth landslide;

[0069] g. Calculate the occurrence index of debris flow using Equation 7, and determine the probability of debris flow occurrence;

[0070]

[0071] In the formula, P is the occurrence index of debris flow, and A is the area of ​​debris flow basin.

[0072] This embodiment is the most basic implementation method. Compared with the prior art, the TRIGRS physical model is used to determine the probability of landslide occurrence, which can make a quantitative judgment, reduce errors, improve the accuracy of debris flow judgment, and ensure the early warning effect.

[0073] Example 2

[0074] A dynamic early warning method for gully-type debris flows induced by shallow landslides includes the following steps:

[0075] a. Determine the potential debris flow basin area using topographic maps;

[0076] b. Determine the physical and mechanical properties of the soil and rock mass through field surveys and indoor experiments, investigate the soil thickness in the area where the debris flow basin is located, and calculate the depth of the grid soil layer using Equation 1.

[0077]

[0078] In the formula, Z i h represents the depth of the grid soil layer. i Z represents the grid elevation. max Z represents the maximum soil thickness in the study area. min h represents the minimum soil thickness in the study area. max h is the highest elevation in the study area min This represents the lowest elevation in the study area;

[0079] c. Real-time monitoring or forecasting of real-time rainfall data at the location of debris flows;

[0080] d. Input real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model. The TRIGRS model outputs the stability coefficient Fs, which characterizes the slope stability state under different rainfall durations. Inputting into the TRIGRS model means inputting through Equation 2.

[0081]

[0082] In the formula, Fs(Z, t) is the slope stability safety factor, Z is the soil depth of any grid, t is time, θ is the soil internal friction angle, c is the cohesion, and γ is the soil internal friction angle. s For soil bulk density, γ w Let Fs be the groundwater unit weight, a be the stability coefficient, and Ψ(Z,t) be the pressure head, which is a function of depth and time. It is a term related to the effect of gravity. It is a term related to cohesion. It is a term related to water pressure;

[0083] e. Calculate the predicted landslide volume in the watershed;

[0084] V n =A n h Formula 3

[0085] A z =A1+A2+A3+...+A n Formula 4

[0086] V z =A z h-type 5

[0087] In the formula, V z V represents the total volume of landslides within the watershed. n Let A be the landslide volume. n Let A be the area of ​​the nth landslide. Based on the fact that Fs < 1 represents the landslide area, A... z denoted as the total landslide area within the watershed, h as the average soil layer thickness in the study area, A1 as the area of ​​the first landslide, A2 as the area of ​​the second landslide, and A3 as the area of ​​the third landslide.

[0088] f. Combining the source coefficient, calculate the volume of landslide source material entering the gully within the watershed using Equation 6;

[0089] V Lz =(A1r1+A2r2+.....+A n r n )h Formula 6

[0090] In the formula, V Lz Let r1 be the source volume of landslides entering the gully within the watershed, r2 be the source coefficient of the first landslide, and r3 be the source coefficient of the second landslide. n Let be the source coefficient of the nth landslide;

[0091] g. Calculate the occurrence index of debris flow using Equation 7, and determine the probability of debris flow occurrence;

[0092]

[0093] In the formula, P is the occurrence index of debris flow, and A is the area of ​​debris flow basin.

[0094] Preferably, in step a, the debris flow basin area refers to the area of ​​all catchment areas other than the debris flow deposition area.

[0095] In step b, the physical and mechanical properties of the soil and rock include cohesion c, soil internal friction angle θ, and soil unit weight γ. s saturated hydraulic conductivity K S Diffusion rate D0, saturated soil moisture content θ s and residual soil moisture content θ r .

[0096] In step d, the stability state includes five categories: Fs≤0.85, 0.85<Fs≤1, 1<Fs≤1.15, 1.15<Fs≤1.3, and 1.3<Fs. Fs<1 is used as the classification criterion for landslide occurrence.

[0097] This embodiment is a preferred implementation method. Landslides on both sides of the gully may not all move and accumulate in the gully to become debris flow sources. The H / L ratio can be used to determine the source of landslides entering the gully, and a reasonable debris flow source calculation index is given, which helps to improve the accuracy of debris flow judgment.

[0098] By studying the relationship between the material and water sources of debris flows and combining their analysis, errors can be reduced and the early warning effect can be improved.

[0099] Example 3

[0100] A dynamic early warning method for gully-type debris flows induced by shallow landslides includes the following steps:

[0101] a. Determine the potential debris flow basin area using topographic maps;

[0102] b. Determine the physical and mechanical properties of the soil and rock mass through field surveys and indoor experiments, investigate the soil thickness in the area where the debris flow basin is located, and calculate the depth of the grid soil layer using Equation 1.

[0103]

[0104] In the formula, Z i h represents the depth of the grid soil layer. i Z represents the grid elevation. max Z represents the maximum soil thickness in the study area. min h represents the minimum soil thickness in the study area. max h is the highest elevation in the study area min This represents the lowest elevation in the study area;

[0105] c. Real-time monitoring or forecasting of real-time rainfall data at the location of debris flows;

[0106] d. Input real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model. The TRIGRS model outputs the stability coefficient Fs, which characterizes the slope stability state under different rainfall durations. Inputting into the TRIGRS model means inputting through Equation 2.

[0107]

[0108] In the formula, Fs(Z, t) is the slope stability safety factor, Z is the soil depth of any grid, t is time, θ is the soil internal friction angle, c is the cohesion, and γ is the soil internal friction angle. s For soil bulk density, γ w Let Fs be the groundwater unit weight, a be the stability coefficient, and Ψ(Z,t) be the pressure head, which is a function of depth and time. It is a term related to the effect of gravity. It is a term related to cohesion. It is a term related to water pressure;

[0109] e. Calculate the predicted landslide volume in the watershed;

[0110] V n =A n h Formula 3

[0111] A z =A1+A2+A3+...+A n Formula 4

[0112] V z =A z h-type 5

[0113] In the formula, V z V represents the total volume of landslides within the watershed. n Let A be the landslide volume. n Let A be the area of ​​the nth landslide. Based on the fact that Fs < 1 represents the landslide area, A... z denoted as the total landslide area within the watershed, h as the average soil layer thickness in the study area, A1 as the area of ​​the first landslide, A2 as the area of ​​the second landslide, and A3 as the area of ​​the third landslide.

[0114] f. Combining the source coefficient, calculate the volume of landslide source material entering the gully within the watershed using Equation 6;

[0115] V Lz =(A1r1+A2r2+......+A n r n )h Formula 6

[0116] In the formula, V Lz Let r1 be the source volume of landslides entering the gully within the watershed, r2 be the source coefficient of the first landslide, and r3 be the source coefficient of the second landslide. n Let be the source coefficient of the nth landslide;

[0117] g. Calculate the occurrence index of debris flow using Equation 7, and determine the probability of debris flow occurrence;

[0118]

[0119] In the formula, P is the occurrence index of debris flow, and A is the area of ​​debris flow basin.

[0120] In step a, the debris flow basin area refers to the area of ​​all catchment areas other than the debris flow deposition area.

[0121] In step b, the physical and mechanical properties of the soil and rock include cohesion c, soil internal friction angle θ, and soil unit weight γ. s saturated hydraulic conductivity K S Diffusion rate D0, saturated soil moisture content θ s and residual soil moisture content θ r .

[0122] In step d, the stability state includes five categories: Fs≤0.85, 0.85<Fs≤1, 1<Fs≤1.15, 1.15<Fs≤1.3, and 1.3<Fs. Fs<1 is used as the classification criterion for landslide occurrence.

[0123] Preferably, in step f, the source coefficient is determined by the ratio of the elevation difference H between the landslide and the gully to the distance L between the landslide and the gully. When H / L≥1, the source coefficient is 0.8; when 1>H / L≥0.5, the source coefficient is 0.5; when 0.5>H / L≥0.3, the source coefficient is 0.2; and when H / L<0.3, the source coefficient is 0.

[0124] In step g, determining the probability of a debris flow specifically means that when P < 0.045, the probability of a debris flow is low; when 0.045 ≤ P < 0.09, the probability of a debris flow is moderate; and when 0.09 ≤ P, the probability of a debris flow is high.

[0125] This embodiment represents the optimal implementation method. By studying the relationship between landslides and debris flow sources, it provides the total debris flow source quantity for all landslides within the watershed. By studying the dynamic changes in debris flow sources, it is possible to more effectively determine the dynamic changes in debris flow development.

[0126] Using material source quantity as a prediction indicator allows for a more direct and effective prediction of shallow landslide-type debris flows in gullies.

[0127] By studying the influencing factors of shallow landslides inducing gully-type debris flows, we provide calculations of source indicators that directly affect debris flows, thereby enabling more direct and effective early warning and judgment of debris flows.

[0128] The invention will now be described with reference to specific examples:

[0129] On June 18, 2010, heavy rainfall in Shunchang County, Fujian Province, triggered a series of mudslides in Baozhuang Village, with mudslides occurring in 25 different watersheds.

[0130] First, the physical and mechanical properties, hydrological parameters, and soil thickness of the soil and rock in Baozhuang Village were obtained through indoor experiments and field surveys, as shown in Table 1.

[0131] Table 1

[0132]

[0133]

[0134] The maximum soil thickness Z in the study area max For 3m, Z min The minimum soil thickness in the study area is 0.1 m, h max The maximum elevation in the study area is 1308m. min The minimum elevation in the study area is 124m.

[0135] From June 14th to 18th, 2010, Shunchang County, Fujian Province, experienced four consecutive days of heavy rainfall, with an average rainfall of 335.7 mm across the county. The landslide mainly occurred between 14:00 and 19:00 on June 18th, which is also the period of debris flow outbreak. This actual rainfall event was selected as the rainfall condition for Baozhuang Village. (See [reference needed]). Figure 1 .

[0136] Because of a previous rainfall event, the stability of landslides within the watershed was analyzed under saturation conditions to obtain the slope stability at different rainfall times. The debris flow source mainly originates from landslides on both banks. First, the stability coefficient Fs < 1 output by the TRIGRS model was interpreted as the landslide extent. Then, the total volume of landslides entering the channel in each watershed was calculated hourly. Finally, the debris flow occurrence index was calculated to determine the dynamic probability of debris flow occurrence.

[0137] Table 2 shows the parameters of 25 debris flows, the calculated occurrence indicators of debris flows, and the actual occurrence of debris flows.

[0138] Table 2

[0139]

[0140]

[0141] In Table 2, when rainfall began 1 hour earlier (June 18th, 4:00 AM), 2 watersheds in the study area were considered highly likely to experience debris flows, while 23 were considered less likely; none of these watersheds experienced debris flows. When rainfall lasted 4 hours (7:00 AM), 2 watersheds in the study area were considered highly likely to experience debris flows, 2 were considered moderately likely, and 21 were considered less likely; none of these watersheds experienced debris flows. When rainfall lasted 8 hours (11:00 AM), 7 watersheds in the study area were considered highly likely to experience debris flows, 10 were considered moderately likely, and 8 were considered less likely; none of these watersheds experienced debris flows. When rainfall lasted 12 hours (3:00 PM), 13 watersheds in the study area were considered highly likely to experience debris flows, and 12 were considered moderately likely; a cluster of debris flows occurred.

[0142] The landslides and debris flows in Baozhuang Village occurred concentratedly after 14:00 on June 18th. Therefore, by 14:00, if there were no incorrectly predicted debris flow areas, the false alarm rate would be 0; if there were no missed reports, the missed report rate would also be 0. Figure 2 As can be seen, the accuracy rate is greater than 72%, the false alarm rate is less than or equal to 28%, and the false negative rate is 0%.

[0143] As can be seen from the above examples, following the principle of no missed reports and few false reports, the method described in this invention has a high accuracy in predicting shallow landslide-type gully debris flows, and can provide a basis for dynamic early warning of shallow landslide-type gully debris flows.

Claims

1. A dynamic early warning method for gully-type debris flows induced by shallow landslides, characterized in that, Includes the following steps: a. Determine the potential debris flow basin area using topographic maps; b. Determine the physical and mechanical properties of the soil and rock mass through field surveys and indoor experiments, investigate the soil thickness in the area where the debris flow basin is located, and calculate the depth of the grid soil layer using Equation 1. Formula 1 In the formula, Z i h represents the depth of the grid soil layer. i Z represents the grid elevation. max Z represents the maximum soil thickness in the study area. min h represents the minimum soil thickness in the study area. max h is the highest elevation in the study area min This represents the lowest elevation in the study area; c. Real-time monitoring or forecasting of real-time rainfall data at the location of debris flows; d. Input real-time rainfall data, physical and mechanical properties of soil and rock, and hydrological parameters into the TRIGRS model. The TRIGRS model calculates the pressure head of each grid cell under different rainfall durations, and outputs the stability coefficient characterizing the slope stability state using Equation 2. ; Formula 2 In the formula, For any grid, the soil depth is... For time, The internal friction angle of the soil. For cohesion, For soil bulk density, The density of groundwater Slope; The pressure head is a function of depth and time. It is a term related to the effect of gravity. It is a term related to cohesion. It is a term related to water pressure; e. A grid area less than 1 is used as the landslide area to calculate the predicted landslide volume in the watershed. Formula 3 In the formula, The volume of the landslide. Let h be the area of ​​the nth landslide, and h be the average thickness of the soil layer in the study area. f. Combining the source coefficient, calculate the volume of landslide source material entering the gully within the watershed using Equation 6; Formula 6 In the formula, This represents the volume of landslide source material entering the gully within the watershed. The source coefficient for the first landslide. The source coefficient for the second landslide. Let be the source coefficient of the nth landslide; This represents the area of ​​the first landslide. This represents the area of ​​the second landslide; g. Calculate the occurrence index of debris flow using Equation 7, and determine the probability of debris flow occurrence; Formula 7 In the formula, As an indicator of debris flow occurrence, The area of ​​the debris flow basin; In step f, the source coefficient is determined by the height difference between the landslide and the gully. Distance between landslide and gully The ratio is determined when When ≥1, the source coefficient is taken as 0.8; when 1> When ≥0.5, the source coefficient is taken as 0.5; when 0.5> When ≥0.3, the source coefficient is taken as 0.2; when When <0.3, the source coefficient is taken as 0.

2. The method for dynamic early warning of gully-type debris flows induced by shallow landslides according to claim 1, characterized in that: In step a, the debris flow basin area refers to the area of ​​all catchment areas other than the debris flow deposition area.

3. The method for dynamic early warning of gully-type debris flows induced by shallow landslides according to claim 1, characterized in that: In step b, the physical and mechanical properties of the soil and rock include cohesion. soil internal friction angle Soil bulk density saturated hydraulic conductivity K S Diffusion rate D0, saturated soil moisture content and residual soil moisture content .

4. The method for dynamic early warning of gully-type debris flows induced by shallow landslides according to claim 1, characterized in that: In step d, the stability states include five categories, namely: ≤0.85, 0.85 < ≤1, 1< ≤1.15, 1.15< ≤1.3 and 1.3< ,Will <1 is used as a classification criterion for landslide occurrence.

5. The method for dynamic early warning of gully-type debris flows induced by shallow landslides according to claim 1, characterized in that: In step g, determining the likelihood of a debris flow specifically refers to when... When the value is <0.045, the probability of debris flow is low; when 0.045 ≤ When the value is <0.09, the probability of a debris flow is moderate; when 0.09 ≤ At that time, the likelihood of a mudslide is high.

Images