Mountain torrent disaster grading early warning method and system based on dynamic coupling of soil moisture content

Abstract

The invention discloses a mountain torrent disaster grading early warning method and system based on soil water content dynamic coupling, and belongs to the technical field of mountain torrent early warning. The method comprises the following steps: breaking through the limitation of a traditional static threshold by constructing a dynamic threshold model for multi-source soil humidity data fusion; a multi-stage early warning trigger mechanism is formed by combining a Kalman filtering real-time correction coefficient, and targeted information pushing and pre-arranged plan dynamic generation are achieved; the dynamic threshold model is constructed through a nonlinear regression model; the multi-stage early warning trigger mechanism comprises blue early warning: real-time rainfall is greater than or equal to 0.7 dynamic critical rainfall threshold; yellow early warning: R is greater than or equal to 0.85 dynamic critical rainfall threshold and the real-time soil water content is greater than or equal to 80% soil saturation water content; red early warning: R is greater than or equal to a dynamic critical rainfall threshold value and a soil liquefaction risk index gt; and 0.6. Through a soil water content dynamic coupling mechanism, three major pain points of data splitting, model stiffness and insufficient aging in mountain torrent early warning are solved.

Description

Technical Field

[0001] This invention belongs to the field of flash flood early warning technology, specifically relating to a method and system for graded early warning of flash flood disasters based on dynamic coupling of soil moisture content. Background Technology

[0002] Flash floods refer to sudden floods that occur in mountainous areas. Flash floods are characterized by their suddenness, concentrated water volume, high flow velocity, and strong destructive force. The water flow carries mud, sand, and even rocks, often causing localized flooding. The assessment of soil loss rate caused by flash floods requires comprehensive consideration of multiple factors, including rainfall intensity, slope, vegetation cover, and soil type.

[0003] Flash floods are highly destructive and sudden. Some regions are extremely prone to flash floods due to their unique meteorological, soil, and topographical characteristics. Failure to promptly assess and warn of the anticipated destructive impact of flash floods in these areas can lead to significant loss of life and property. Furthermore, the lack of accurate disaster warning and classification can result in the irrational allocation of disaster relief and mitigation resources, further exacerbating the disaster's impact. While existing technologies offer various methods for classifying and issuing early warnings for flash floods, they have their limitations and shortcomings, including: 1) Limitations of static thresholds: Traditional flash flood warnings use fixed critical rainfall thresholds, which do not take into account real-time changes in soil moisture content, leading to delayed warnings or false alarms.

[0004] 2) Severe data silos: The existing system does not integrate multi-source soil moisture data (such as satellite remote sensing and ground sensors), resulting in monitoring blind spots.

[0005] 3) Poor model adaptability: Early warning models based on empirical formulas (such as the water level-discharge inverse method) are difficult to adapt to the soil characteristics of different small watersheds. Summary of the Invention

[0006] The present invention aims to at least partially solve one of the technical problems in the aforementioned related technologies.

[0007] Therefore, the purpose of this invention is to provide a method and system for graded early warning of flash flood disasters based on dynamic coupling of soil moisture content, which solves the three major pain points of flash flood early warning: "data fragmentation, model rigidity, and insufficient timeliness" through the dynamic coupling mechanism of soil moisture content.

[0008] To solve the above-mentioned technical problems, the present invention is implemented as follows: This invention provides a method for graded early warning of flash floods based on dynamic coupling of soil moisture content, the method comprising: By constructing a dynamic threshold model that integrates multi-source soil moisture data, the limitations of traditional static thresholds are overcome; and by combining it with real-time correction coefficients using Kalman filtering, a multi-level early warning triggering mechanism is formed, enabling targeted information delivery and dynamic generation of contingency plans.

[0009] In addition, the flash flood disaster graded early warning method based on dynamic coupling of soil moisture content according to the present invention may also have the following additional technical features: In some of these implementations, the dynamic threshold model is constructed using a nonlinear regression model.

[0010] In some implementations, the input data of the dynamic threshold model includes real-time soil moisture content, soil saturation moisture content, previous rainfall, and soil moisture deficit; the output data is a dynamic critical rainfall threshold.

[0011] In some of these embodiments, the soil moisture deficit is calculated based on the soil saturation moisture content and the real-time soil moisture content.

[0012] In some implementations, the multi-level early warning triggering mechanism includes: Blue alert: When the real-time rainfall R ≥ 0.7R critical Initiate risk assessment; Yellow alert: R ≥ 0.85R critical And S(t) ≥ 80% S max This triggers emergency preparedness. Red alert: R ≥ R critical Furthermore, the soil liquefaction risk index L>0.6; Among them, R critical S represents the dynamic critical rainfall threshold, and S(t) represents the real-time soil moisture content. max This indicates the saturated water content of the soil.

[0013] In some of these implementations, the multi-source soil moisture data includes: surface soil volumetric water content data obtained from SAR satellite data inversion, deformation-humidity coupled data obtained from BeiDou / GNSS data, and soil moisture content data obtained from ground sensor data.

[0014] In some of these implementations, the dynamic threshold model is represented as: R critical =α˙D(t) +β˙P(t-Δt) +γ˙log(S(t) / S max ) Where D(t) is the soil moisture deficit, P(t-Δt) is the previous rainfall, S(t) is the real-time soil moisture content, and S maxdenoted as soil saturation water content, and α, β, and γ are nonlinear regression coefficients.

[0015] In some of these implementations, when a red alert is met, A is improved. The algorithm dynamically generates evacuation routes.

[0016] In some of these implementations, deformation-humidity coupling data is obtained based on BeiDou / GNSS data by calculating deformation data using the surface deformation rate calculation formula, and then coupling it with humidity data to obtain the deformation-humidity coupling index. Soil moisture content data obtained from ground sensor data is obtained through ensemble Kalman filter calibration.

[0017] This invention also provides a graded early warning system for flash floods based on dynamic coupling of soil moisture content, used to implement the graded early warning method for flash floods based on dynamic coupling of soil moisture content as described in any of the above embodiments; the system includes: The data acquisition layer is used to realize multi-source soil moisture monitoring and meteorological data fusion. The meteorological data fusion includes: real-time access to meteorological radar precipitation forecasts and rain gauge measured data, and construction of a rainfall-soil moisture coupling matrix. The core algorithm layer is used to calculate the dynamic critical rainfall threshold based on the dynamic threshold model, and correct the model coefficients in real time. Then, based on the real-time rainfall, dynamic critical rainfall threshold, soil liquefaction risk index, real-time soil moisture content and soil saturated moisture content, the classification of flash flood disasters is determined. The early warning execution layer is used to realize the dynamic generation of contingency plans and the targeted information push.

[0018] Compared with the prior art, the present invention has at least the following beneficial effects: In this embodiment of the invention, the provided method for graded early warning of flash flood disaster based on dynamic coupling of soil moisture content constructs a dynamic early warning threshold model driven by real-time soil moisture, breaking through the limitations of traditional static thresholds. In this embodiment of the invention, the provided method for graded early warning of flash flood disaster based on dynamic coupling of soil moisture content realizes the fusion and assimilation of multi-source soil moisture data, thereby improving the accuracy of monitoring coverage; In this embodiment of the invention, a graded early warning method for flash flood disasters based on dynamic coupling of soil moisture content is provided, and a dynamic early warning algorithm that adapts to the characteristics of small watersheds is developed to suit different soil types and rainfall patterns.

[0019] The flash flood disaster classification and early warning system based on dynamic coupling of soil moisture content of the present invention is used to implement the aforementioned flash flood disaster classification and early warning method based on dynamic coupling of soil moisture content. Therefore, it has at least all the features and advantages of the aforementioned flash flood disaster classification and early warning method based on dynamic coupling of soil moisture content, which will not be repeated here. Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] Figure 1 This is a technical solution framework diagram disclosed in one embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and specific examples and application scenarios.

[0023] In some embodiments of the present invention, a technical framework for graded early warning of flash flood disasters is provided, including a data acquisition layer, a core algorithm layer, and an early warning execution layer.

[0024] The data acquisition layer primarily performs multi-source soil moisture monitoring and meteorological data fusion. Multi-source soil moisture monitoring integrates SAR satellite inversion of soil moisture content, BeiDou / GNSS surface deformation monitoring, and ground-based IoT sensors (such as capacitive soil moisture meters). Meteorological data fusion involves real-time access to meteorological radar precipitation forecasts and rain gauge data to construct a rainfall-soil moisture coupling matrix.

[0025] The core algorithm layer primarily calculates the dynamic critical rainfall threshold based on soil moisture content, previous rainfall, and soil saturation moisture content. The steps of the dynamic threshold generation algorithm include: ① Calculate the soil moisture deficit: D(t) = S max -S(t), In the formula, S(t) is the real-time soil moisture content, S max This represents the saturated water content of the soil.

[0026] ② Calculate the dynamic critical rainfall threshold based on the nonlinear regression model: R critical =α˙D(t) +β˙P(t-Δt) +γ˙log(S(t) / Smax ) In the formula, P(t-Δt) represents the previous rainfall, and α, β, and γ are nonlinear regression coefficients.

[0027] ③ The nonlinear regression coefficients are corrected in real time using Kalman filtering.

[0028] Based on real-time rainfall, real-time soil moisture content, and R critical S max The system classifies mountain torrent disasters and obtains warning levels for subsequent warning implementation.

[0029] Early warning execution layer: Targeted information push: Based on GIS spatial analysis, automatically send multilingual voice alerts to residents in high-risk areas; Dynamic contingency plan generation: Based on historical disaster data, evacuation routes and material dispatch plans are generated.

[0030] In some embodiments of the present invention, a method for graded early warning of flash flood disasters based on dynamic coupling of soil moisture content is provided, the steps of which include: Step 1: Multi-source soil moisture data acquisition and fusion The targets of this step are: SAR satellite (Sentinel-1), BeiDou / GNSS ground station, and IoT sensors.

[0031] Table 1

[0032] This step, through the fusion of three-source data, can increase the monitoring coverage from 67% to 98.5% and achieve a spatiotemporal resolution of 100m / 10 minutes.

[0033] Step 2: Calculation of dynamic critical rainfall threshold The objects of this step are: real-time soil moisture content S(t) and historical rainfall P(t-Δt).

[0034] Processing flow: (1) Input data: S(t) = get_soil_moisture(), which is the function for the fused data in step 1; P(t-Δt)=get_rainfall_history(Δt=6h), which represents the cumulative rainfall over 6 hours.

[0035] (2) Calculate the water deficit: D(t) = S max -S(t), In the formula, S max Set according to soil type (0.45m for sandy soil) 3 / m3 0.32m of clay 3 / m 3 ).

[0036] (3) Dynamic threshold generation: R critical =α˙D(t) +β˙P(t-Δt) +γ˙ln(S(t) / S max ) Initial parameters: α=1.2, β=0.85, γ=0.15 (calibrated using historical flood data).

[0037] (4) Dynamic parameter correction: Kalman filter coefficients updated in real time: K = P _ prior ˙ H’ / ( H ˙ P _ prior ˙ H’ + R )

[0038] The advantages of this step include: (compared to traditional static thresholds) the critical rainfall prediction RMSE is reduced from 15.7 mm to 8.2 mm (a reduction of 47.8%); and the response time is reduced from ≥90 seconds to ≤30 seconds.

[0039] Step 3: Multi-level early warning triggering and emergency response The objects of this step are: real-time rainfall R(t) and soil liquefaction risk index L.

[0040] The multi-level early warning triggering mechanism for this step is as follows: Blue alert: When the real-time rainfall R ≥ 0.7R critical Initiate risk assessment; Yellow alert: R ≥ 0.85R critical And S(t)≥80% S max This triggers emergency preparedness. Red alert: R ≥ R critical Furthermore, the soil liquefaction risk index L>0.6.

[0041] The early warning execution mechanism includes: (1) Blue Alert (Risk Prediction): Update the GIS risk map and send advance reminders to grassroots personnel (via SMS / APP push).

[0042] (2) Yellow Alert (Emergency Preparedness): Start the material scheduling model: min st ≥ D j Preposition supplies to designated distribution points (response time ≤ 1 hour) (3) Red Alert (Immediate Action Required): Trigger BeiDou short message broadcast (supports voice broadcast in 12 languages) Dynamically generate evacuation routes (improved A) Algorithm): Cost(e) = +

[0043] The advantages of this step include: a 60% increase in evacuation efficiency due to the three-level early warning system, and an increase in multilingual broadcast coverage from 75% to 95%.

[0044] Table 2

[0045] This invention establishes a real-time feedback loop of rainfall-soil moisture-surface deformation through a dynamic coupling mechanism (data update frequency ≤ 10 minutes); the multi-level response model of this invention quantifies the mapping relationship between risk level and emergency measures (blue / yellow / red warnings correspond to 12 / 24 / 48-hour emergency responses); cross-protocol communication is compatible with BeiDou short message, 4G / 5G, and LoRa multi-mode transmission (covering areas without network coverage).

[0046] Any part of this invention not described in detail can be referred to in the prior art or in the art known to those skilled in the art. This embodiment does not limit such part and will not describe it in detail here.

[0047] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for graded early warning of flash flood disasters based on dynamic coupling of soil moisture content, characterized in that, The method includes: By constructing a dynamic threshold model that integrates multi-source soil moisture data, the limitations of traditional static thresholds are overcome; and by combining it with real-time correction coefficients using Kalman filtering, a multi-level early warning triggering mechanism is formed, enabling targeted information delivery and dynamic generation of contingency plans.

2. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 1, characterized in that, The dynamic threshold model is constructed using a nonlinear regression model.

3. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 1, characterized in that, The input data of the dynamic threshold model includes real-time soil moisture content, soil saturation moisture content, previous rainfall, and soil moisture deficit; the output data is the dynamic critical rainfall threshold.

4. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 3, characterized in that, The soil moisture deficit is calculated based on the soil saturation moisture content and the real-time soil moisture content.

5. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 1, characterized in that, The multi-level early warning triggering mechanism includes: Blue alert: When the real-time rainfall R ≥ 0.7R critical Initiate risk assessment; Yellow alert: R ≥ 0.85R critical And S(t) ≥ 80% S max This triggers emergency preparedness. Red alert: R ≥ R critical Furthermore, the soil liquefaction risk index L > 0.6; Among them, R critical S represents the dynamic critical rainfall threshold, and S(t) represents the real-time soil moisture content. max This indicates the saturated water content of the soil.

6. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 1, characterized in that, Multi-source soil moisture data includes: topsoil volumetric water content data obtained from SAR satellite data inversion, deformation-humidity coupled data obtained from BeiDou / GNSS data, and soil moisture content data obtained from ground sensor data.

7. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 1, characterized in that, The dynamic threshold model is expressed as follows: R critical =α˙D(t) +β˙P(t-Δt) +γ˙log(S(t) / S max ) Where D(t) is the soil moisture deficit, P(t-Δt) is the previous rainfall, S(t) is the real-time soil moisture content, and S max denoted as soil saturation water content, and α, β, and γ are nonlinear regression coefficients.

8. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 5, characterized in that, When a red alert is met, improvements can be made. The algorithm dynamically generates evacuation routes.

9. The method for graded early warning of flash floods based on dynamic coupling of soil moisture content according to claim 6, characterized in that, The method for obtaining deformation-humidity coupled data based on BeiDou / GNSS data is to calculate deformation data using the surface deformation rate calculation formula, and then couple it with humidity data to obtain the deformation-humidity coupling index. Soil moisture content data obtained from ground sensor data is obtained through ensemble Kalman filter calibration.

10. A graded early warning system for flash floods based on dynamic coupling of soil moisture content, characterized in that, The content is used to implement the mountain flood disaster graded early warning method based on dynamic coupling of soil moisture content as described in any one of claims 1 to 9; The system includes: The data acquisition layer is used to realize multi-source soil moisture monitoring and meteorological data fusion. The meteorological data fusion includes: real-time access to meteorological radar precipitation forecasts and rain gauge measured data, and construction of a rainfall-soil moisture coupling matrix. The core algorithm layer is used to calculate the dynamic critical rainfall threshold based on the dynamic threshold model, and correct the model coefficients in real time. Then, based on the real-time rainfall, dynamic critical rainfall threshold, soil liquefaction risk index, real-time soil moisture content and soil saturated moisture content, the classification of flash flood disasters is determined. The early warning execution layer is used to realize the dynamic generation of contingency plans and the targeted information push.

Images