Debris flow triggering type grading early warning method

Abstract

The invention discloses a debris flow trigger type grading early warning method. The method comprises the following steps: selecting an area rainfall threshold analysis area in which debris flow is widely distributed; debris flow data collection; selecting effective cumulative rainfall as a main factor for stimulating debris flow by rainfall; sub-region division is carried out on the region to be pre-warned, and the cumulative frequency of effective rainfall of the sub-regions is calculated as a pre-warning threshold value; taking the effective cumulative rainfall corresponding to the 50% cumulative frequency as the critical rainfall of early warning, and triggering blue early warning when the critical rainfall is reached; after blue early warning is triggered, the monitored rainfall is changed into minute-level sampling uploading frequency, and yellow early warning is carried out when the set rainfall intensity is reached; after yellow early warning is triggered, earth sound signal analysis is started, and whether debris flow breaks out or not is judged according to the fluctuation of the three-axis vibration amplitude. According to the method, the intelligent level of single-point debris flow rainfall monitoring can be optimized by analyzing regional rainfall early warning information, the early warning reliability is improved, and the debris flow disaster prevention consciousness of the masses in the risk area is enhanced.

Description

Technical Field

[0001] The invention relates to a multi-scale debris flow triggered graded early warning method based on remote sensing rainfall and ground professional monitoring, and belongs to the field of meteorological disasters and geological disasters. Background Art

[0002] The debris flow early warning system is divided into two scales: regional and single disaster point scale. Regionally, a certain number of rain gauges are deployed based on the debris flow prone or dangerous zoning in the region, and disaster early warning is carried out by monitoring rainfall and statistical thresholds (such as rainfall intensity-duration relationship, effective cumulative rainfall, etc.). For a single debris flow disaster point, the main focus is on monitoring key parameters and setting early warning thresholds.

[0003] In the prior art, there is a debris flow early warning technology based on regional rainfall monitoring. It is mainly based on the long-term debris flow events, and the rainfall intensity and the long-term data are fitted with a power function or the critical rainfall for the debris flow outbreak is statistically analyzed. The fitted curve or threshold is used as the meteorological judgment condition for whether the debris flow will break out or not to issue a debris flow early warning. The main problem with the debris flow early warning technology based on regional rainfall monitoring is that since debris flow is a low-probability event, the event samples in the same debris flow basin are limited. Therefore, this technology mainly reflects the risk of debris flow early warning at the regional scale. However, when the disaster-stricken objects (such as people, cars, and houses) are already in the warning area, this technology cannot provide accurate information such as the location of the debris flow and the time of the disaster, making it difficult to leave enough emergency evacuation time for the masses.

[0004] There are also single-disaster-point debris flow early warning methods based on rainfall monitoring in the prior art. This method is based on the theoretical rainfall calculation of debris flow outbreaks, and proposes monitoring and early warning rainfall thresholds, which are mainly used to guide disaster prevention and risk avoidance in single-disaster-body danger zones.

[0005] The main problem with the single-point debris flow early warning method based on rainfall monitoring is that single-point monitoring can only focus on a certain debris flow basin and cannot timely improve the safety awareness of people who have entered the risk area. Due to the lack of consideration of the regional environmental rainfall conditions, the high-frequency sampling of monitoring equipment under unnecessary circumstances will lead to waste of electricity, shorten the life of sensors, and easily cause problems such as insufficient power supply at critical moments. Summary of the invention

[0006] The purpose of the present invention is to overcome the above-mentioned problems existing in the prior art and to provide a debris flow triggered graded early warning method, which can optimize the intelligence level of single-point debris flow rainfall monitoring by analyzing regional rainfall early warning information, improve the reliability of early warning, and enhance the debris flow disaster prevention awareness of people in risk areas.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows: A debris flow triggered graded early warning method, characterized in that it comprises the following steps: a. Select the rainfall threshold analysis area in the area where debris flows are widely distributed; b. Debris flow data collection; c. Extraction of key rainfall characteristic factors: Select effective cumulative rainfall as the main factor that triggers debris flow and is regarded as the rainfall condition for debris flow; d. Regional division and warning threshold calculation: Divide the area to be warned into sub-regions, and for each sub-region, calculate the cumulative frequency of effective rainfall as the warning threshold; e. The effective cumulative rainfall corresponding to the 50% cumulative frequency is used as the critical rainfall for each sub-region warning. When the average rainfall in the sub-region reaches the critical rainfall, the sub-region blue warning is triggered; f. Triggering ground rainfall monitoring: For high-risk points in the region, when a blue warning is triggered in a sub-region, the monitored rainfall is immediately changed to a minute-level sampling and uploading frequency. When the set rainfall intensity is reached, a yellow warning is issued; g. Vibration signal coordinated warning: For high-risk points in the area, when the yellow warning of ground rainfall is triggered, the ground sound signal analysis is initiated to determine whether a mudslide will occur based on the fluctuation of the three-axis vibration amplitude.

[0008] In step a, the selection principles include: selecting a large-scale area covering a single disaster point to ensure that the debris flow basin in the area is developed, the historical debris flow event data is sufficient, the spatial distribution is relatively uniform, and the satellite rainfall data has sufficient temporal and spatial coverage.

[0009] In step b, the debris flow data samples must be within 30 years and have a clear outbreak date.

[0010] In the step c, based on regional research on rainstorm debris flows, the effective accumulated rainfall is selected as the main factor that triggers debris flows due to rainfall.

[0011] In step c, the effective cumulative rainfall ( R e ) is composed of the previous rainfall and the daily rainfall. After normalization based on the multi-year average rainfall, it can be regarded as the rainfall condition for debris flow, expressed as formula (1): (1) in R e is the effective accumulated rainfall; Before the mudslide n The effective previous precipitation of the day, P i For the i The accumulated precipitation of the day; P0 It is the daily precipitation on the day when the debris flow occurs, which can be obtained through short-term weather forecast in actual warning; K is the previous rainfall index, indicating that the previous rainfall has little impact on the formation of debris flow; K and n It can be set based on experience or determined experimentally; P anaual Represents the average rainfall over many years in a certain area.

[0012] In the step d, the area to be warned is divided into sub-areas according to the topography, geology and geomorphology of the region through the spatial superposition of the three elements.

[0013] In step d, referring to the hydrological frequency calculation method, the cumulative frequency of effective rainfall is calculated by the following formula: (2) F N is the cumulative frequency, which is also determined as the probability that a given value is less than the Nth significant rainfall amount in the M observation data sets.

[0014] Said F N The specific implementation method is: sort the M effective rainfalls of debris flows calculated in a sub-region from small to large, and the cumulative frequency corresponding to the effective rainfalls from the 1st to the Nth is F N They are expressed as: , ,…, .

[0015] In the step f, intelligent rainfall monitoring equipment is deployed before the flood season for the high-risk points in each sub-area, and the full network access and Beidou dual communication modes are adopted to connect with the background system in real time. Under normal weather conditions, the rainfall monitoring equipment uploads data to the background system once an hour. When the regional blue warning is triggered, the control signal is transmitted through the background system to the encrypted monitoring command to the ground rainfall monitoring equipment, and the monitored rainfall is then changed to a minute-level sampling and uploading frequency.

[0016] In the step f, a yellow warning is issued when the rainfall intensity exceeds 2 mm / h for 2 hours or more or exceeds 5 mm / h for a short time.

[0017] In step g, when the waveform changes suddenly, an orange warning is issued; when it lasts for more than 30 seconds, it can be confirmed as a debris flow, and a red warning is issued.

[0018] The advantages of adopting the present invention are: 1. The method of the present invention can organically combine regional early warning information and local monitoring means, refine early warning through a step-by-step triggering mechanism, reduce false alarm rate, improve early warning reliability, and provide graded early warning support for local disaster prevention and mitigation departments.

[0019] 2. The regional-scale warning information in the method of the present invention takes into account the influence of topographic, geological and geomorphic factors, forming a regional warning, making the traditional regional warning more refined in space. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is the regional zoning warning information map in Example 2; Figure 2 The daily effective accumulated rainfall monitoring chart based on GPM satellite monitoring in Example 2; Figure 3 It is the single-point high-precision rainfall monitoring map in Example 2; Figure 4 This is the geoacoustic equipment collaborative monitoring early warning map in Example 2. DETAILED DESCRIPTION

[0021] Example 1 A debris flow triggered graded early warning method, comprising: (I) Select rainfall threshold analysis areas in areas where debris flows are widely distributed. The selection principles include: selecting large-scale areas covering single disaster points, ensuring that the debris flow basin in the area is well developed, that historical debris flow event data is sufficient (no less than the number of records required for statistics), that the spatial distribution is relatively uniform, and that the satellite rainfall data has sufficient temporal and spatial coverage.

[0022] (II) Collection of debris flow data. Relevant data can be collected by consulting monographs, conducting on-site surveys and visiting local villagers, and collecting data from land departments. Debris flow data samples must be within 30 years and have a clear outbreak date (at least including: year, month, and day).

[0023] (III) Extraction of key rainfall characteristic factors. Based on the regional research on rainstorm debris flow, effective cumulative rainfall is selected as the main factor that triggers debris flow. R e ) is composed of the previous rainfall and the daily rainfall. After normalization based on the multi-year average rainfall, it can be regarded as the rainfall condition for debris flow, expressed as formula (1): (1) in R e is the effective precipitation; the first item on the right is the rainfall before the debris flow occurs. n The effective previous precipitation of the day, Pi For the i The accumulated precipitation of the day; P 0 It is the daily precipitation on the day when the debris flow occurs, which can be obtained through short-term weather forecast in actual warning; K It is the previous rainfall index, which indicates that the previous rainfall has little impact on the formation of debris flow. K and n It can be set based on experience or determined experimentally. anaual Represents the average rainfall over many years in a certain area.

[0024] (IV) Calculation of regional warning thresholds. Based on the topography, geology, and geomorphology of the region, the sub-regions are divided by spatial superposition of the three elements. For each sub-region, the cumulative frequency of effective rainfall is calculated by referring to the hydrological frequency calculation method, which is expressed as formula (2): (2) F N is the cumulative frequency, which is also determined as the probability that a given value is less than the Nth effective rainfall in the M observation data sets. Specific implementation method: sort the M debris flow effective rainfalls calculated in a sub-area from small to large, then the cumulative frequency corresponding to the 1st to the Nth effective rainfall F N They are expressed as: , ,…, .

[0025] (V) Generally speaking, the effective cumulative rainfall corresponding to the 50% cumulative frequency is used as the critical rainfall for regional warning. Once the average rainfall in the monitored area reaches this level, the regional blue warning (general) of the background system is triggered, warning people who are about to enter the area or are already in the area to observe and prevent possible debris flow disasters at any time.

[0026] (vi) Triggering ground rainfall monitoring. For high-risk areas in the region, intelligent rainfall monitoring equipment is deployed before the flood season, using full network access and Beidou dual communication modes to connect with the background system in real time. Under normal weather conditions, the rainfall monitoring equipment uploads data to the background system once an hour. When the regional warning is triggered, the control signal conveys the encrypted monitoring instructions to the ground rainfall monitoring equipment through the background system, and the monitored rainfall is immediately changed to a minute-level sampling upload frequency. Because the regional warning takes into account the previous monitored rainfall and predicted rainfall, in order to reduce the missed reporting rate, a yellow warning (heavy) is set when the rainfall intensity exceeds 2mm / h for 2 hours or more or exceeds 5mm / h for a short time. (VII) Vibration signal coordinated warning. For high-risk points in the region, ground sound monitoring equipment is deployed before the flood season, using full network access and real-time connection with the background system. Due to the huge amount of ground sound monitoring data, real-time data analysis is not performed on a daily basis. When the yellow warning of ground rainfall is triggered, the ground sound signal analysis is started, and the occurrence of debris flow is determined based on the fluctuation of the triaxial vibration amplitude. When the waveform suddenly changes, an orange warning (serious) is issued; when it is further sustained (more than 30s), it can be confirmed as a debris flow movement, and a red warning (particularly serious) is issued.

[0027] The monitoring system and equipment involved in the present invention adopt the existing technology.

[0028] Table 1 Warning levels and monitoring methods in the present invention Example 2 In this embodiment, the feasibility of the method of the present invention is verified by taking the Guxianggou mudslide as an example. The specific steps are as follows: (I) Collect the large-area geology, geomorphology, and topographic zoning maps near the debris flow, and vectorize the zoning data. From the perspective of geomorphology, geology, and climate, it can be divided into high mountain and mountain original landforms, high and medium mountain and mountain original landforms, Gondwana and pro-Gondwana continental block groups, South China and pro-South China continental block groups, Qinghai-Tibet alpine climate, and eastern monsoon climate. Based on this, the spatial superposition of three major types of geographical elements was performed through ArcGIS software, forming five sub-zoning types composed of the superposition of these three types of elements.

[0029] (ii) By visiting the natural resources department, we collected information such as the date and geographical location of nearly a thousand debris flow events, and extracted the effective cumulative rainfall of each event (including the effective rainfall in the previous seven days and the effective rainfall on the day) from the satellite rainfall data GPM (global precipitation measurement mission) according to the longitude, latitude and time, referring to formula (1). We further used formula (2) to sort the effective cumulative rainfall of debris flow events of each sub-region type, and calculated the cumulative frequency corresponding to each value. Finally, we selected the characteristic rainfall with a cumulative frequency of 50% as the standard for regional blue warning. Considering that the Guxianggou Basin in this case is more affected by the high-altitude cold climate of the Qinghai-Tibet Plateau, the threshold of partition type 5 is used as a reference ( Figure 1 ).

[0030] (III) In the daily monitoring during the flood season, the first step is to carry out daily rainfall monitoring based on the GPM satellite (in actual business operations, the rainfall for the next day can be estimated from the weather forecast department). The effective accumulated rainfall in the region is calculated using formula (1), and the result is: Figure 2 When the ratio of regional effective cumulative rainfall to the multi-year average rainfall reaches or exceeds Figure 3 In this example, we can see that on July 9, the ratio reached 4.64%, exceeding the Figure 1 The warning threshold is given, which triggers a blue warning and initiates refined high-frequency monitoring on the ground.

[0031] (IV) As a famous debris flow basin in history, Guxianggou is generally recognized to be highly susceptible to debris flow, which will not be elaborated here. For this reason, high-precision tipping bucket rain gauges and geoacoustic monitoring equipment have been deployed in the debris flow basin. When the blue warning of the area is triggered, the rainfall monitoring and analysis function of Guxianggou is first activated, and the continuous rainfall intensity is calculated by high-frequency sampling and feedback. Since the regional warning has already included consideration of the previous water conditions, in order to reduce the underreporting rate of high-risk single points, the yellow warning is immediately activated when the continuous state of rainfall intensity in the single-point monitoring reaches the specified level. From the example analysis, it can be seen that there was continuous rainfall of a certain intensity at 11:30, 17:00 and 20:00 on July 9, which triggered the single-point warning threshold ( Figure 3 ).

[0032] (V) Under the premise that the dynamic changes of material sources in the upstream high-cold and high-altitude areas are unclear, rainfall warnings are highly empirical and more likely to be probabilistic predictions. In order to more accurately identify the occurrence of debris flows, monitoring information that can reflect the physical characteristics of debris flow movement is required. As described in the technical plan, after the local rainfall prompts the warning, the geoacoustic sensors deployed in the basin are immediately activated to carry out high-frequency, real-time waveform analysis. From this example analysis, we can see that ( Figure 4 ), at 21:29 on July 9, the vibration signal generated by the debris flow showed a significant change. At this time, it was considered that it might be a debris flow movement, landslide or rolling stone, and the warning level was raised from yellow to orange. Subsequently, the waveform showed continuous strong fluctuations, and the amplitude became larger and larger. At this time, considering the relatively long-duration movement characteristics of the debris flow, it can be clearly determined that a debris flow has occurred, and the warning level is raised to the highest level. The final example results show that the multi-scale debris flow cascade triggered graded warning method proposed in the present invention is feasible and can effectively provide step-by-step warnings for debris flow disasters.

Claims

1. A debris flow triggered graded early warning method, characterized in that: The steps include: a. Select the rainfall threshold analysis area in the area where debris flows are widely distributed; b. Debris flow data collection; c. Extraction of key rainfall characteristic factors: Select effective cumulative rainfall as the main factor that triggers debris flow and is regarded as the rainfall condition for debris flow; d. Regional division and warning threshold calculation: Divide the area to be warned into sub-regions, and for each sub-region, calculate the cumulative frequency of effective rainfall as the warning threshold; e. The effective cumulative rainfall corresponding to the 50% cumulative frequency is used as the critical rainfall for each sub-region warning. When the average rainfall in the sub-region reaches the critical rainfall, the sub-region blue warning is triggered; f. Triggering ground rainfall monitoring: For high-risk points in the region, when a blue warning is triggered in a sub-region, the monitored rainfall is immediately changed to a minute-level sampling and uploading frequency. When the set rainfall intensity is reached, a yellow warning is issued; g. Vibration signal coordinated warning: For high-risk points in the area, when the yellow warning of ground rainfall is triggered, the ground sound signal analysis is initiated to determine whether a mudslide will occur based on the fluctuation of the three-axis vibration amplitude.

2. The debris flow triggered graded early warning method according to claim 1, characterized in that: In step a, the selection principles include: selecting a large-scale area covering a single disaster point to ensure that the debris flow basin in the area is developed, the historical debris flow event data is sufficient, the spatial distribution is relatively uniform, and the satellite rainfall data has sufficient temporal and spatial coverage.

3. The debris flow triggered graded early warning method according to claim 2 is characterized by: In step b, the debris flow data samples must be within 30 years and have a clear outbreak date.

4. The debris flow triggered graded early warning method according to claim 3 is characterized by: In the step c, based on regional research on rainstorm debris flows, the effective accumulated rainfall is selected as the main factor that triggers debris flows due to rainfall.

5. The debris flow triggered graded early warning method according to claim 4, characterized in that: In step c, the effective cumulative rainfall ( R e ) is composed of the previous rainfall and the daily rainfall. After normalization based on the multi-year average rainfall, it can be regarded as the rainfall condition for debris flow, expressed as formula (1): (1) in R e is the effective accumulated rainfall; Before the mudslide n The effective previous precipitation of the day, P i For the i The accumulated precipitation for the day; P 0 It is the daily precipitation on the day when the debris flow occurs, which can be obtained through short-term weather forecast in actual warning; K is the previous rainfall index, indicating that the previous rainfall has little impact on the formation of debris flow; K and n It can be set based on experience or determined experimentally; P anaual Represents the average rainfall over many years in a certain area.

6. The debris flow triggered graded early warning method according to claim 5, characterized in that: In the step d, the area to be warned is divided into sub-areas according to the topography, geology and geomorphology of the region through the spatial superposition of the three elements.

7. The debris flow triggered graded early warning method according to claim 6, characterized in that: In step d, referring to the hydrological frequency calculation method, the cumulative frequency of effective rainfall is calculated by the following formula: (2) F N is the cumulative frequency, which is also determined as the probability that a given value is less than the Nth significant rainfall amount in the M observation data sets.

8. The debris flow triggered graded early warning method according to claim 7, characterized in that: Said F N The specific implementation method is: sort the M effective rainfalls of debris flows calculated in a sub-region from small to large, and the cumulative frequency corresponding to the effective rainfalls from the 1st to the Nth is F N They are expressed as: , ,…, .

9. The debris flow triggered graded early warning method according to claim 8, characterized in that: In the step f, intelligent rainfall monitoring equipment is deployed before the flood season for the high-risk points in each sub-area, and the full network access and Beidou dual communication modes are adopted to connect with the background system in real time. Under normal weather conditions, the rainfall monitoring equipment uploads data to the background system once an hour. When the regional blue warning is triggered, the control signal is transmitted through the background system to the encrypted monitoring command to the ground rainfall monitoring equipment, and the monitored rainfall is then changed to a minute-level sampling and uploading frequency.

10. The debris flow triggered graded early warning method according to claim 9, characterized in that: In the step f, a yellow warning is issued when the rainfall intensity exceeds 2 mm / h for 2 hours or more or exceeds 5 mm / h for a short time; in the step g, an orange warning is issued when the waveform mutates; and when the waveform continues for more than 30 seconds, it can be confirmed as a debris flow, and a red warning is issued.

Images