Earthquake signal-based glacier debris flow dynamic risk early warning method and system

Abstract

The invention provides a glacier debris flow dynamic risk early warning method and system based on a seismic signal, and belongs to the technical field of natural disaster early warning. Firstly, a glacier response chain of the seismic signal is constructed, and information such as a seismic signal propagation path, a glacier circle layer response sequence and a response characteristic transfer relation is acquired; analyzing risk evolution nodes in the response chain information, accurately determining key nodes of response triggering, energy accumulation, diffusion starting and the like to form a glacier debris flow risk evolution node set, and then linking the glacier debris flow risk evolution node set with the glacier region topographic features to obtain a glacier debris flow risk evolution node set; the method comprises the following steps: generating a dynamic risk evolution path capable of reflecting the dynamic development process of glacier debris flow, generating dynamic risk early warning content containing detailed early warning information based on the dynamic risk evolution path, and finally reversely associating field response data to a response chain construction process to realize optimization and updating of glacier response chain information of seismic signals. And the early warning capability for glacier debris flow disasters is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of natural disaster early warning technology, and more specifically, to a method and system for dynamic risk early warning of glacial debris flows based on seismic signals. Background Technology

[0002] Glacial debris flows, as a highly destructive natural disaster, pose a serious threat to human life, property, and the ecological environment in the areas surrounding glaciers. Currently, early warning systems for glacial debris flows primarily rely on traditional meteorological monitoring and geological surveys. Meteorological monitoring predicts the likelihood of debris flows by monitoring meteorological elements such as rainfall and temperature; however, the relationship between meteorological factors and the occurrence of glacial debris flows is not a simple linear one, and meteorological monitoring cannot accurately reflect the dynamic changes within the glacier. Geological surveys focus on static analysis of the topography and geological structure of the glacier area to assess the potential risk of debris flows; however, geological conditions are relatively stable, making it impossible to promptly capture changes in the internal structure of the glacier and the energy accumulation processes triggered by sudden factors such as earthquakes.

[0003] Earthquakes, as a powerful geological activity, generate seismic waves that significantly impact glaciers, potentially altering their internal stress states, causing ice fracturing, and accelerating glacial movement, thereby increasing the risk of glacial debris flows. However, existing early warning methods fail to fully utilize the complex response relationship between seismic signals and glaciers, lacking a systematic analysis and early warning mechanism for the dynamic process from seismic signal triggering to glacial debris flow formation. This results in insufficient accuracy and timeliness of early warnings, making it difficult to effectively prevent and mitigate the losses caused by glacial debris flow disasters. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for dynamic risk early warning of glacial debris flows based on seismic signals, the method comprising: A glacier response chain for seismic signals is constructed, and seismic signal glacier response chain information is generated. The seismic signal is a continuous seismic wave signal in and around the glacier region. The seismic signal glacier response chain information includes the seismic signal propagation path, the response sequence of each glacier layer, and the transmission relationship of response characteristics. The risk evolution nodes in the glacier response chain information of the seismic signal are analyzed to generate a set of glacier debris flow risk evolution nodes, which includes response triggering nodes, energy accumulation nodes, and diffusion initiation nodes. By linking the set of risk evolution nodes of glacial debris flows with the topographic features of glacial areas, a dynamic risk evolution path for glacial debris flows is generated; Based on the dynamic risk evolution path of glacial debris flow, generate dynamic risk warning content for glacial debris flow; By inversely linking the on-site response data of dynamic risk warning content to the glacier response chain construction process of seismic signals, the optimization and updating of glacier response chain information of seismic signals is completed.

[0005] Furthermore, embodiments of the present invention also provide a dynamic risk early warning system for glacial debris flows based on seismic signals, characterized in that it includes: A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described dynamic risk warning method for glacial debris flows based on seismic signals by executing the machine-executable instructions.

[0006] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of the seismic signal-based glacial debris flow dynamic risk early warning system reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the seismic signal-based glacial debris flow dynamic risk early warning system to execute the aforementioned seismic signal-based glacial debris flow dynamic risk early warning method.

[0007] Based on the above, by constructing a glacier response chain for seismic signals, information on the seismic signal-glacier response chain is generated, encompassing the seismic signal propagation path, the response sequence of each glacier sphere, and the transmission relationship of response characteristics. Risk evolution nodes within this information are analyzed, generating a set of glacier debris flow risk evolution nodes including response trigger nodes, energy accumulation nodes, and diffusion initiation nodes. This identifies key stages in the formation of glacier debris flows. By linking the risk evolution node set with the topographic features of the glacier region, a dynamic risk evolution path for glacier debris flows is generated. This comprehensively considers the impact of seismic activity and topographic factors on debris flow development. Based on the risk warning content generated from the dynamic risk evolution path, accurate early warnings of glacier debris flows can be provided. Finally, the field response data is back-linked to the response chain construction process to optimize and update the information, thereby continuously adapting to changes in glacier and seismic activity, improving the accuracy and reliability of the early warning method, and effectively enhancing the ability to prevent glacier debris flow disasters. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the execution flow of the dynamic risk early warning method for glacial debris flow based on seismic signals provided in an embodiment of the present invention.

[0009] Figure 2 This is a schematic diagram of exemplary hardware and software components of the dynamic risk early warning system for glacier debris flows based on seismic signals provided in an embodiment of the present invention. Detailed Implementation

[0010] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a dynamic risk warning method for glacial debris flows based on seismic signals, provided in one embodiment of the present invention. The following is a detailed description of this dynamic risk warning method for glacial debris flows based on seismic signals.

[0011] Step S110: Construct the glacier response chain of the seismic signal and generate seismic signal glacier response chain information. The seismic signal is a continuous seismic wave signal in and around the glacier region. The seismic signal glacier response chain information covers the seismic signal propagation path, the response sequence of each glacier layer, and the transmission relationship of response characteristics.

[0012] This embodiment focuses on a glacier region on the XX Plateau. This glacier region has a wide glacier coverage area and a complete glacier layer structure, and has historically experienced multiple glacier debris flow disasters triggered by earthquakes. All subsequent steps will be described in relation to this region.

[0013] Step S111: Continuously collect seismic signals from the glacier area and its surroundings, and organize the seismic signals in chronological order of collection time to generate a time-series seismic signal sequence containing seismic signals from multiple consecutive time periods. Each time period in the time-series seismic signal sequence is associated with a corresponding collection time identifier, which is used to establish a temporal correspondence with the environmental response data of the glacier area.

[0014] In this embodiment, multiple broadband seismometers can be deployed in and around the glacier area. The sampling frequency of the seismometers is set to a certain number of sampling points per second to ensure that subtle changes in the seismic signal can be captured. The acquisition process is continuous, and the acquired seismic signals are processed hourly as a time period. For example, the seismic signals from 00:00 to 01:00 each day are taken as the first time period, those from 01:00 to 02:00 as the second time period, and so on. Each time period's seismic signal is associated with an acquisition time identifier, such as "202405200000-0100", which indicates the signal acquired from 00:00 to 01:00 on May 20, 2024. This acquisition time identifier can accurately correspond to the environmental response data of the glacier area during that time period.

[0015] Step S112: Perform waveform analysis processing on the seismic signals of each time period in the time-series seismic signal sequence, extract waveform-related features such as waveform fluctuation shape, signal change amplitude, signal duration, and signal propagation speed of the seismic signals of each time period, generate a set of waveform features for each time period, and associate the set of waveform features for each time period with the corresponding time period identifier to distinguish the signal features of different time periods.

[0016] For each time-series seismic signal, waveform analysis is performed using seismic signal processing software. The waveform morphology is determined by analyzing the distribution of peaks and troughs, such as whether the waveform exhibits a single-peak, multi-peak, or irregular fluctuation pattern. The signal amplitude is obtained by calculating the difference between the maximum and minimum amplitudes of the seismic signal within that time period. The signal duration is the length of that time period, such as 1 hour in the previous example. The signal propagation speed is calculated by comparing the time difference between different seismographs receiving the same seismic wave and combining this with the known distance between the seismographs. These extracted features are integrated to form a set of seismic signal waveform features for each time period, and associated with a corresponding time period identifier, such as "202405200000-0100 waveform feature set".

[0017] Step S113: Collect glacier region layer structure data and generate glacier layer structure dataset. The glacier region layer structure data covers the distribution range, structural characteristics, and material composition of the glacier surface layer, glacier middle layer, glacier bottom layer, glacier surrounding strata, and glacier surrounding water layer.

[0018] In this embodiment, the glacier's layered structure data can be collected through various methods such as glacier drilling, remote sensing image interpretation, and geological surveys. The glacier's surface layer is located at the uppermost part of the glacier, is relatively thin, and generally consists of newly fallen snow and partially melted snow and ice, with a relatively loose structure. The glacier's middle layer lies below the surface layer and is the main body of the glacier. It is thicker and mainly consists of dense glacial ice, possibly containing a small amount of rock debris. The glacier's bottom layer is in contact with the glacier bedrock, has a more complex composition, including glacial ice, moraines, and rock debris scraped from the bedrock, and has poor structural stability. The glacier's perigee layer refers to the rock strata of the surrounding mountains, mainly composed of sedimentary or igneous rocks, and has a relatively stable structure. The glacier's perigee layer includes glacial lakes at the glacier's front, subglacial lakes within the glacier, and surrounding rivers, with varying distribution ranges and depths. After processing the above data, a glacier layered structure dataset is generated.

[0019] Step S114: Based on the signal propagation velocity characteristics in the time-series seismic signal sequence, and combined with the material composition and structural characteristics of each layer in the glacier stratum structure data, analyze the propagation path of the seismic signal in each glacier stratum and surrounding areas, determine the order in which the seismic signal enters each glacier stratum from the periphery of the glacier region, and generate seismic signal stratum propagation path data.

[0020] Signal propagation velocity features are extracted from the waveform feature sets of each time period in a time-series seismic signal sequence. Due to significant differences in material composition and structural characteristics among different spheres, seismic waves propagate at different speeds; for example, the propagation speed in glacier ice differs significantly from that in rock strata. Combining the distribution range, material composition (e.g., glacier ice, rock, water, etc.), and structural characteristics (e.g., density, porosity, etc.) of each sphere in the glacier stratum structure dataset, seismic wave propagation theory is used to simulate which spheres a seismic signal passes through after entering from the periphery of the glacier region (e.g., surrounding stratigraphic spheres), and the propagation path and direction within each sphere. For example, a seismic signal may first enter the surrounding stratigraphic spheres, then enter the glacier interior through the bottom sphere, and then propagate upwards to the middle and surface spheres, and may also propagate to the surrounding aquatic spheres. The order in which the seismic signal is received by each sphere is determined, and this information is recorded to generate seismic signal sphere propagation path data.

[0021] Step S115: Continuously monitor the state changes of each sphere in the glacier stratum structure data under the action of time-series seismic signal sequences, capture the state change information of each sphere under the action of seismic signals at different time periods, the state change information includes changes in structural integrity, changes in material movement state, and changes in spatial position displacement, generate seismic response data for each sphere, and each sphere's seismic response data is associated with the corresponding seismic signal time period identifier.

[0022] Various monitoring devices, such as strain gauges, displacement sensors, and GPS, were deployed in the glacier region to continuously monitor the state changes of each sphere under the influence of time-series seismic signals. Changes in structural integrity were assessed by monitoring the appearance of new cracks within the spheres and the expansion of existing cracks. Changes in material movement included monitoring changes in the flow velocity of glacier ice and the movement of rock debris. Spatial displacement was monitored by GPS and other devices, tracking the three-dimensional coordinate changes of key points in each sphere. These state changes were recorded for each sphere under seismic signals at different times. For example, after a seismic signal at a certain time, new cracks were detected in the surface glacier (change in structural integrity), the flow velocity in the middle glacier increased (change in material movement), and a small horizontal and vertical displacement occurred at a point in the bottom glacier (change in spatial displacement). This information was categorized and organized by sphere, and associated with the corresponding seismic signal time period identifiers to generate seismic response data for each sphere.

[0023] Step S116: Extract response features from the seismic response data of each sphere to generate a set of seismic response features for each sphere. The response features cover the time point of the response, the duration of the response, the form of the response change, and the range of the response. The set of seismic response features for each sphere is associated with the corresponding sphere identifier and time segment identifier.

[0024] In-depth analysis of seismic response data for each sphere is conducted to extract response characteristics. The time point of response occurrence refers to the specific time when the state change begins in each sphere; the duration of response is the time elapsed from the onset of state change to basic stabilization; the morphology of response change refers to the specific manifestation of the state change, such as the expansion pattern of cracks and the trajectory of material movement; the range of response refers to the size of the spatial area affected by the state change. For example, for the response data of the glacier surface sphere under the action of the seismic signal during the period "202405200000-0100", the time point of response occurrence is extracted as a specific moment within that period, the duration of response is several minutes, the morphology of response change is radial expansion of cracks, and the range of response is an area of ​​one square meter. These response characteristics are integrated to form a set of seismic response characteristics for each sphere, and associated with the corresponding sphere identifier (e.g., "glacier surface sphere") and time period identifier (e.g., "202405200000-0100").

[0025] Step S117: Compare the response occurrence time points in the seismic response feature sets of each sphere with the signal arrival time in the seismic signal propagation path data of each sphere, divide the response delay period of each sphere to the seismic signal, and generate sphere response delay data. The sphere response delay data is used to characterize the time interval between the arrival of the seismic signal and the generation of the response in the sphere.

[0026] Step S1171: Extract the response occurrence time points corresponding to each sphere from the seismic response feature set of each sphere, classify and organize the response occurrence time points according to the sphere identifier, and generate a sphere response time association dataset.

[0027] The seismic response feature sets of each sphere are traversed, and the response occurrence time points of each sphere at different time periods are extracted. For example, the glacier surface sphere has response occurrence time points in multiple time periods, and these time points are associated with the sphere identifier "glacier surface sphere". This process is performed on all spheres, and then they are classified according to the sphere identifier. All response occurrence time points of the same sphere are grouped together to generate a sphere response time association dataset. Each record in this dataset contains the sphere identifier and the corresponding response occurrence time point.

[0028] Step S1172: Extract the time point of the earthquake signal arrival at each layer from the earthquake signal propagation path data, classify and organize the signal arrival time points according to the layer identifier, and generate a layer signal arrival time association dataset.

[0029] The seismic signal propagation path data records the arrival time of seismic signals in each sphere. The arrival time of the signal for each sphere is extracted, such as the arrival time of the signal for the strata surrounding a glacier, or the arrival time of the signal for the strata at the bottom of the glacier. These arrival times are then categorized and organized according to sphere identifiers, grouping arrival times for signals within the same sphere together to generate a sphere-based signal arrival time association dataset. Each record contains the sphere identifier and the corresponding arrival time.

[0030] Step S1173: Match the concentric response time association dataset with the concentric signal arrival time association dataset according to the concentric identifier, establish the correspondence between the response occurrence time point and the signal arrival time point of each concentric layer, and generate the concentric time corresponding dataset.

[0031] The layer response time association dataset and the layer signal arrival time association dataset are associated, with the matching condition being that the layer identifiers are the same. For each layer, the response occurrence time and signal arrival time within the same time period are found, and a one-to-one correspondence is established. For example, for the "glacier surface layer" in the time period "202405200000-0100", the signal arrival time and response occurrence time of this time period are found and associated with them. This generates a layer time correspondence dataset, where each record contains the layer identifier, signal arrival time, and response occurrence time.

[0032] Step S1174: For the time-corresponding data of each concentric circle, extract the response occurrence time point and the signal arrival time point, and calculate the time interval between the two time points. This time interval is used as the response delay duration of the concentric circle to the seismic signal.

[0033] From the time-correspondence dataset for each seismic sphere, the signal arrival time and response occurrence time are read sequentially. For each pair of time points, the difference between them is calculated: the response occurrence time minus the signal arrival time. The resulting time interval is the response delay duration of that sphere to the seismic signal. For example, if the signal arrival time for a certain sphere is t1 and the response occurrence time is t2, then the response delay duration is t2 - t1.

[0034] Step S1175: Record the response delay duration of each layer, associate it with the corresponding layer identifier, and generate preliminary layer response delay data.

[0035] The calculated response delay duration for each glacier layer is associated with its identifier, such as "glacier surface layer: response delay duration T1" and "glacier middle layer: response delay duration T2". This association information is then organized into a data table or list to generate preliminary layer response delay data.

[0036] Step S1176: Extract structural feature data of each layer from the glacier layer structure dataset; analyze the correlation between the material composition and structural integrity parameters of different layers and the response delay time, and generate structural impact analysis data.

[0037] Structural characteristic data for each glacier sphere was obtained from a glacier sphere structure dataset, including material composition (such as glacial ice purity, rock type, and water salinity) and structural integrity parameters (such as fracture density, porosity, and compressive strength). The effects of these factors on response delay time were analyzed. For example, spheres with homogeneous material composition and high structural integrity may have shorter response delay times because seismic waves propagate and trigger responses more rapidly within them; conversely, spheres with complex material composition and low structural integrity may have longer response delay times. Through statistical analysis and correlation studies of a large amount of data, the degree and patterns of correlation between material composition, structural integrity parameters, and response delay time were determined, generating structural impact analysis data.

[0038] Step S1177: Based on the structural influence analysis data, for the concentric circles in the preliminary concentric circle response delay data where the correlation between structural features and delay duration does not conform to the preset rules, re-verify the accuracy of the response occurrence time and signal arrival time, and correct the calculation results of the response delay duration; integrate all concentric circle corrected response delay duration, concentric circle identifier and structural feature data to generate concentric circle response delay data.

[0039] Based on the correlation patterns derived from the structural impact analysis data, the preliminary layer response delay data are examined. If the structural characteristics of a certain layer indicate that its response delay should be short, but the preliminary calculation result is longer, or vice versa, it indicates that there may be errors in the response occurrence time or signal arrival time of that layer. In this case, it is necessary to review the original monitoring data and seismic signal records to verify and correct the response occurrence time and signal arrival time of that layer. For example, the error may be due to a monitoring equipment malfunction causing an incorrect recording of the response occurrence time, or there may have been a calculation deviation when analyzing the seismic signal arrival time. After correction, the response delay duration is recalculated, and the corrected response delay durations of all layers, the corresponding layer identifiers, and structural characteristic data are integrated to generate the final layer response delay data.

[0040] Step S118: Based on the layer response delay data and the seismic signal layer propagation path data, determine the response order of each layer of the glacier to the seismic signal, integrate the transmission relationship between the response characteristics of each layer, and generate layer response transmission relationship data.

[0041] Step S1181: Extract the response delay duration of each layer from the layer response delay data, combine it with the time point of the earthquake signal arriving at each layer in the earthquake signal propagation path data, calculate the absolute time of the actual response of each layer, and generate the layer response absolute time data.

[0042] The seismic signal propagation path data for each seismic layer contains the response delay duration of that layer, while the data on the seismic signal's arrival time in each layer contains the arrival time of the seismic signal. For each layer, adding the response delay duration to its arrival time yields the absolute time of the actual response generated in that layer. For example, if the arrival time of a certain layer is t_arrival and the response delay duration is t_delay, then its actual absolute response time t_delay = t_arrival + t_delay. By calculating the actual absolute response times of all layers and associating them with their corresponding layer identifiers, the absolute response time data for each layer is generated.

[0043] Step S1182: Sort each sphere according to the absolute time sequence in the absolute time data of the sphere response, determine the response order of each glacier sphere to the seismic signal, and generate sphere response order data.

[0044] The absolute times in the absolute time data of the glacier's response are sorted in ascending order; the smaller the absolute time, the earlier the response occurred. Based on the sorting result, the identifiers of each glacier sphere are listed sequentially to obtain the order of their response to the seismic signal. For example, the sorting result might be: glacier periphery strata, glacier basal layer, glacier middle layer, glacier surface layer, and glacier periphery water layer. This indicates that under the influence of the seismic signal, each sphere responded sequentially. This response order is recorded to generate glacier response sequence data.

[0045] Step S1183: Extract the seismic response feature set of each sphere from the seismic signal glacier response chain information, organize the response features of each sphere according to the order in the sphere response sequence data, and generate an ordered sphere response feature set.

[0046] From the previously generated seismic response data for each sphere, we extract seismic response feature sets for each sphere. These sets include features such as the time of response occurrence, duration of response, morphology of response changes, and extent of response. Then, according to the chronological order of responses determined in the sphere response sequence data, these response feature sets are rearranged. For example, the response feature sets are arranged sequentially according to the following order: periglacial stratigraphy, glacial subsurface, glacial middle strata, glacial surface, and periglacial water bodies, generating an ordered sphere response feature set.

[0047] Step S1184: Analyze the response features of two adjacent layers in the ordered layer response feature set, extract the change pattern and generation range in the response feature of the previous layer, and identify the similarity between the response feature and the change pattern and generation range in the response feature of the next layer.

[0048] Within the ordered strata response feature set, two adjacent strata are selected, such as the strata surrounding a glacier and the glacier floor strata. The variation patterns (e.g., vibration patterns, fracture development patterns) and extent of occurrence (e.g., the area affected by vibration, the area of ​​fracture distribution) of the response features of the former strata (the strata surrounding a glacier) are extracted. Then, the variation patterns (e.g., the movement patterns of glacial till, the deformation patterns of the glacier floor strata) and extent of occurrence are extracted from the response features of the latter strata (the glacier floor strata). Through comparative analysis, it is determined whether the two exhibit similarities in variation patterns, such as similar vibration frequencies or consistent deformation directions; and whether there are overlapping or continuous areas in the extent of occurrence, such as whether the extent of the response from the former strata extends into the extent of the response from the latter strata.

[0049] Step S1185: Based on the similarity analysis results, determine whether the response characteristics of the previous layer cause changes in the response of the next layer. If there is similarity and there is a sequential connection in time, it is determined that there is a response transmission relationship between the two.

[0050] If the response characteristics of two adjacent strata show high similarity in terms of change pattern and range, and the timing of the response of the preceding strata plus its duration is temporally connected to the timing of the response of the following strata (i.e., the response of the preceding strata has not yet ended or has just ended when the response of the following strata begins), then it can be determined that the response characteristics of the preceding strata triggered the response changes of the following strata, and there is a response transmission relationship between the two. For example, after the vibration response of the strata surrounding a glacier lasts for a period of time, the glacial basal strata begin to show a moraine movement response, and the vibration frequencies of the two are similar, their response ranges overlap, and their timing is also connected, then it can be determined that there is a response transmission relationship between them.

[0051] Step S1186: Record adjacent layer pairs with response transit relationships, present the forward transit layer and the backward response layer, and generate layer transit pair data.

[0052] Record each pair of adjacent strata that are determined to have a response transmission relationship. Each pair includes a forward-transmitting stratum (the stratum that responds first and transmits the response characteristics) and a backward-transmitting stratum (the stratum that receives the transmitted response characteristics and responds). For example, (glacier periphery stratum, glacier subsurface stratum), (glacier subsurface stratum, glacier middle stratum), etc. Organize these stratum pairs into data format to generate stratum transmission pair data.

[0053] Step S1187: Extract the response characteristic parameters of the forward transmission layer and the response characteristic parameters of the backward response layer in the transmission pair data, analyze the correlation between the change amplitude of the forward response characteristics and the change amplitude of the backward response characteristics, and generate transmission influence characteristic data.

[0054] Select a sphere pair from the sphere transmission pair data, such as (glacier subsurface sphere, glacier mid-sphere). Extract the amplitude parameters (such as displacement amplitude, vibration amplitude, etc.) from the response characteristics of the forward transmission sphere (glacier subsurface sphere) and the amplitude parameters (such as displacement amplitude, vibration amplitude, etc.) from the response characteristics of the backward response sphere (glacier mid-sphere). By comparing the magnitudes of their amplitudes, analyze whether there is a positive correlation, negative correlation, or other relationship between them. For example, the larger the amplitude of the forward transmission sphere, the larger the amplitude of the backward response sphere, indicating a positive correlation. Record the above relationships (such as correlation coefficient, influence ratio, etc.), and perform the above analysis for each sphere transmission pair to generate transmission influence characteristic data.

[0055] Step S1188: Combining the data on the transmission of each layer with the characteristic data on the transmission impact, integrate the response transmission paths between each layer, present the layer order and response characteristic transmission details on each transmission path, and generate layer transmission path data.

[0056] Based on the data on strata transmission, we can identify which strata have direct response transmission relationships. Connecting these direct transmission relationships forms response transmission paths. For example, a transmission path could be formed from the strata surrounding a glacier to the glacier's lower strata, then from the lower strata to the middle strata, and finally to the glacier's surface strata. Simultaneously, by combining this with transmission impact characteristic data, we can mark the strata sequence and details of response characteristic transmission along each transmission path, such as the correlation of change magnitudes and transmission time delays. Integrating all transmission paths and their detailed information generates strata transmission path data.

[0057] Step S1189: Integrate the data on the transmission of the inner circle, the data on the characteristics of the transmission impact, and the data on the transmission paths of the inner circle to generate preliminary data on the transmission relationship of the inner circle response; supplement the time connection information of each transmission path in the preliminary data on the transmission relationship of the inner circle response, present the time correlation between the end of the forward inner circle response and the start of the backward inner circle response, improve the time dimension description of the transmission relationship, and generate data on the transmission relationship of the inner circle response.

[0058] By integrating data on layer-by-layer transmission pairs, transmission impact characteristics, and layer-by-layer transmission paths, preliminary layer-by-layer response transmission relationship data is formed. This preliminary data reflects the transmission pair relationships, impact characteristics, and transmission paths between each layer. Then, the start and end times (response occurrence time plus response duration) of the response for each layer are extracted from the seismic response characteristic set of each layer. For forward and backward layers along the transmission path, the relationship between the end time of the forward layer response and the start time of the backward layer response is determined; for example, whether the start time of the backward layer response is before, after, or simultaneous with the end time of the forward layer response. This temporal information is then added to the preliminary layer-by-layer response transmission relationship data to further refine the temporal description of the transmission relationship, ultimately generating the final layer-by-layer response transmission relationship data.

[0059] Step S119: Analyze the correlation basis of each link in the preliminary seismic signal glacier response chain, extract the intermediate action characteristics in the response transmission process, and generate response chain detailed data; integrate the seismic signal propagation path data of each sphere, the seismic response characteristic set of each sphere, the response delay data of each sphere, the response transmission relationship data of each sphere, and the response chain detailed data to construct the seismic signal glacier response chain information.

[0060] The preliminary seismic signal-glacier response chain constructed earlier (including seismic signal propagation path data for each sphere, seismic response feature sets for each sphere, response delay data for each sphere, and response transmission relationship data for each sphere) is analyzed to clarify the basis for the interrelationships between each link, such as how the propagation path affects the response sequence and how response features are transmitted through transmission relationships. During the response transmission process, there may be some intermediate characteristics, such as the energy attenuation characteristics of the seismic signal during propagation and the morphological transformation characteristics of the response features during transmission. These intermediate characteristics are extracted to generate detailed response chain data. Finally, the seismic signal propagation path data for each sphere, the seismic response feature sets for each sphere, the response delay data for each sphere, the response transmission relationship data for each sphere, and the detailed response chain data are all integrated to form complete seismic signal-glacier response chain information. This seismic signal-glacier response chain information comprehensively reflects the propagation of the seismic signal in the glacier region, the response of each sphere, and the transmission process of response features.

[0061] Step S120: Analyze the risk evolution nodes in the glacier response chain information of the seismic signal to generate a set of glacier debris flow risk evolution nodes, which includes response triggering nodes, energy accumulation nodes, and diffusion initiation nodes.

[0062] Step S121: Extract the seismic response feature set of each sphere from the seismic signal glacier response chain information, filter the response features in the seismic response feature set of each sphere whose response change mode exceeds the threshold of the normal fluctuation range, and generate an abnormal response feature set.

[0063] Seismic response feature sets for all glacier spheres are extracted from the seismic signal glacier response chain information. For the response variation pattern in the response feature set of each sphere, it is compared with the conventional fluctuation range threshold of the response variation pattern under normal conditions (no significant seismic signal action). The conventional fluctuation range threshold is obtained based on statistical analysis of long-term monitoring data of the glacier region and reflects the normal variation range of each sphere under natural conditions. If the response variation pattern of a certain sphere exceeds this conventional fluctuation range threshold, it indicates that the response feature is abnormal. All the above abnormal response features are collected to generate an abnormal response feature set.

[0064] Step S122: Extract the seismic signal waveform features corresponding to the abnormal response feature set, determine the key features of the seismic signal that triggered the abnormal response, associate the concentric circle identifiers corresponding to the abnormal response features with the response occurrence time points, and generate abnormal response association data.

[0065] For each anomalous response feature in the set of anomalous response features, based on its associated time period identifier, the waveform features of the seismic signal for that time period are extracted from the waveform feature set of the corresponding time period in the time-series seismic signal sequence. These waveform features are analyzed to identify which features (such as the severity of waveform fluctuations, the magnitude of signal changes, the duration of the signal, and anomalous signal propagation speed) are likely to be key factors triggering the anomalous response. For example, an anomalous response feature may be triggered by a seismic signal with a signal amplitude exceeding the normal range. After identifying the key seismic signal features that trigger the anomalous response, these key features are associated with the corresponding concentric circle identifier and the time point of the response, generating anomalous response association data.

[0066] Step S123: Based on the abnormal response correlation data, locate the glacier layer that generated the abnormal response, present the layer structure characteristics and material composition of the glacier layer, determine the initial triggering location of risk evolution, and generate the location information of the response triggering node.

[0067] Based on the layer identifiers in the anomaly response association data, the corresponding layer is located in the glacier layer structure dataset, thus pinpointing the location of the glacier layer that generated the anomaly response. Then, the layer structure characteristics (such as thickness, structural integrity, porosity, etc.) and material composition (such as glacial ice, rock debris, glacial till, water, etc.) of this location are extracted from the glacier layer structure dataset and presented. Since this anomaly response is a potential starting point for the evolution of glacial debris flow risk, this layer location is determined as the initial triggering location for risk evolution, and its geographic coordinates and relative position within the glacier layer structure are recorded to generate the location information of the response triggering node.

[0068] Step S124: Extract the response duration and response change amplitude from the seismic response characteristics corresponding to the location of the response trigger node, analyze the initial energy accumulation at the response trigger node, and generate the initial energy state data of the response trigger node.

[0069] From the seismic response feature set corresponding to the triggering node location, two features are extracted: response duration and response amplitude. Response duration reflects the duration of energy interaction, while response amplitude reflects the intensity of energy interaction. Combining the material composition and structural characteristics of the layer (such as the elastic modulus and density of the material), the initial energy accumulation at the triggering node under the influence of the aforementioned response duration and amplitude are analyzed, including the intensity and distribution range of energy accumulation. The analysis results are recorded to generate initial energy state data for the triggering node.

[0070] Step S125: Integrate the location information of the response trigger node with the initial energy state data to generate response trigger node information, which represents the initial triggering link of the glacial debris flow risk evolution.

[0071] By integrating the location information (geographic coordinates, sphere location, etc.) of the response trigger node with the initial energy state data (energy concentration intensity, distribution range, etc.), a complete response trigger node information is formed. This information clearly describes the location from which the evolution of glacial debris flow risk begins, as well as the initial energy state at the time of triggering, characterizing the initial triggering link of risk evolution.

[0072] Step S126: Extract layer response transmission relationship data from the seismic signal glacier response chain information, track the transmission process of response characteristics generated by the response triggering node in each subsequent layer, record the changes in response characteristics during the transmission process, and generate response transmission tracking data.

[0073] Data on the transmission relationships of responses across different spheres is obtained from the seismic signal glacier response chain information. This data includes the transmission paths, transmission pairs, and temporal connections between spheres. Based on this data, starting from the sphere containing the response trigger node, the transmission process of its generated response characteristics through subsequent spheres is tracked along the transmission path. During the tracking process, detailed records are kept of changes in the response characteristics at each transmission stage, such as whether the magnitude of the response change increases or decreases, whether the duration of the response increases or decreases, whether the morphology of the response changes, and whether the range of the response expands or shrinks. These tracking records are then compiled to generate response transmission tracking data.

[0074] Step S127: Analyze the response transmission tracking data, identify the locations of concentric circles where the response characteristics change more significantly or last longer during transmission, and generate energy accumulation area identification data.

[0075] Step S1271: Extract response feature change data of each layer from the response transmission tracking data, organize the response feature change data according to the time sequence of response transmission, and generate time-series response transmission data. The time-series response transmission data is used to reflect the dynamic changes of response features during the transmission process.

[0076] Extract response feature change data for each layer during the response feature transmission process from the response transmission tracking data, including the magnitude, duration, form, and range of change. Organize and sort this change data according to the chronological order of response feature transmission between layers to form time-series response transmission data. For example, first record the response feature change data of the response trigger node layer, then the change data of the first subsequent layer along the transmission path, followed by the change data of the second subsequent layer, and so on. In this way, the time-series response transmission data clearly reflects the dynamic changes of the response features during the transmission process.

[0077] Step S1272: Extract the response characteristic change amplitude parameter of each layer from the time-series response transmission data, associate it with the transmission time according to the layer identifier, and generate a layer amplitude time association dataset.

[0078] From the time-series response transmission data, for each concentric circle, extract the variation amplitude parameter of its response characteristics. Associate this variation amplitude parameter with the corresponding concentric circle identifier and the transmission time at which the response characteristics were received by that concentric circle, for example, "Concentric Circle A, transmission time t1, variation amplitude M1", "Concentric Circle B, transmission time t2, variation amplitude M2", etc. After organizing all the above association records, generate a concentric circle amplitude time-related dataset.

[0079] Step S1273: Obtain the baseline range threshold for the change amplitude of response characteristics of each sphere in the glacier region based on historical data statistics; compare the change amplitude parameters in the sphere amplitude time correlation dataset with the baseline range, and filter out the spheres whose change amplitude parameters exceed the upper limit of the baseline range, wherein the spheres are potential areas where the change amplitude of response characteristics increases.

[0080] By statistically analyzing historical data on multiple earthquake events and glacier responses in the glacier region, baseline threshold ranges for the variation amplitude of response characteristics in each sphere were obtained. These baseline threshold ranges include upper and lower limits, reflecting the fluctuation range of variation amplitude under normal response conditions. The variation amplitude parameter of each sphere in the temporal correlation dataset was compared with the corresponding baseline threshold range. If the variation amplitude parameter is greater than the upper limit of the baseline range, it indicates an abnormally large increase in the variation amplitude of the response characteristics of that sphere, and the sphere is marked as a potential region with an increased variation amplitude of response characteristics.

[0081] Step S1274: Extract the response feature duration parameter of each layer from the time-series response transmission data, associate it with the transmission time according to the layer identifier, and generate a layer-time long-term association dataset.

[0082] Similarly, the duration parameter of each layer response feature is extracted from the time-series response transmission data, and associated with the corresponding layer identifier and transmission time to generate a layer-time associated dataset, such as "layer A, transmission time t1, duration T1" and "layer B, transmission time t2, duration T2".

[0083] Step S1275: Obtain the baseline threshold for the duration of response characteristics of each glacier layer based on historical data statistics; compare the duration parameters in the time-long correlation dataset of the glaciers with the baseline threshold, and filter out the glaciers whose duration parameters exceed the baseline threshold.

[0084] Based on historical data, a baseline threshold for the duration of response characteristics of each glacier sphere was obtained. This baseline threshold represents the upper limit of the duration under normal response conditions. The duration parameter of each sphere in the sphere-time-long-term correlation dataset was compared with the corresponding baseline threshold, and spheres with duration parameters greater than the baseline threshold were selected.

[0085] Step S1276: Merge concentric circles whose change magnitude exceeds the upper limit of the baseline range or whose duration exceeds the baseline threshold to generate a set of candidate regions for energy accumulation.

[0086] The concentric circles whose variation amplitude exceeds the upper limit of the baseline range selected in step S1273 and whose duration exceeds the baseline threshold selected in step S1275 are merged. If a concentric circle meets both conditions, it is only counted as one candidate region. All the concentric circles obtained after merging form a set of candidate regions for energy accumulation.

[0087] Step S1277: Collect the concentric structure data corresponding to each region in the candidate region set. Based on the stability and permeability parameters of the concentric structure, select regions with high structural instability and low permeability from the candidate region set to generate energy accumulation region identification data.

[0088] Data on the layered structure of each region in the candidate region set was collected from the glacier layered structure dataset, focusing on structural stability parameters (such as fracture density, compressive strength, and shear strength) and permeability parameters (such as permeability and porosity). Regions with high structural instability are prone to deformation and failure, which is conducive to energy accumulation; regions with low permeability hinder energy loss, allowing energy to accumulate in that region. Based on preset structural instability and permeability evaluation criteria, each region in the candidate region set was evaluated, and regions with both high structural instability and low permeability were selected. The identifiers of these regions (such as layer identifiers and specific location coordinates) were recorded to generate energy accumulation region identifier data.

[0089] Step S128: Collect the layer structure data and material motion state data corresponding to the energy accumulation area identification data, analyze the energy accumulation rate and total accumulation amount in the energy accumulation area, and generate energy accumulation state data.

[0090] Based on energy accumulation region identification data, layer structure data (such as thickness, density, and material composition ratio) for the corresponding region are collected from the glacier layer structure dataset. Material motion state data (such as material velocity, vibration frequency, and displacement) for the region are also collected from the seismic response data of each layer. Using principles of energy science and glacier dynamics models, combined with the collected structural and material motion state data, the increase in energy accumulation per unit time (i.e., energy accumulation rate) and the total energy accumulation from the start of the response to the current moment are analyzed. For example, the change in kinetic energy can be calculated using the velocity and mass of the material, and combined with the change in potential energy generated by structural deformation, the energy accumulation situation is comprehensively assessed. The calculated energy accumulation rate and total accumulation are recorded to generate energy accumulation state data.

[0091] Step S129: Integrate energy accumulation area identification data and energy accumulation status data to generate energy accumulation node information. The energy accumulation node information represents the key link in the continuous energy accumulation process during the risk evolution of glacial debris flow.

[0092] By integrating energy accumulation area identification data (regional layer identifiers, location coordinates, etc.) with energy accumulation status data (energy accumulation rate, total accumulation, etc.), energy accumulation node information is formed. This information clarifies which areas are key links in the continuous accumulation of energy during the evolution of glacial debris flow risk, and the energy accumulation status of these links.

[0093] Step S1210: Extract response range data from the seismic response feature set of each sphere from the seismic signal glacier response chain information, combine it with energy accumulation node information, identify the sphere where the response range begins to spread outward after the energy accumulation reaches a set level, determine the diffusion start time and initial diffusion direction of the sphere, and generate diffusion start node information.

[0094] The response range data for each seismic response sphere is extracted from the seismic response feature set of each sphere. This response range data describes the spatial extent of the response characteristics' influence within that sphere. Combined with the total energy accumulation in the energy accumulation node information, when the energy accumulation reaches a preset threshold (determined based on historical disaster data and glacier stability analysis, representing a critical energy value at which diffusion may occur), the response range data for each energy accumulation area is closely monitored. If the response range at a certain sphere location is found to no longer be confined to its own area but begins to expand to surrounding areas, it indicates that this sphere location is the starting point of diffusion. The time point at which the response begins to diffuse outward is extracted from the seismic response feature set of that sphere as the diffusion initiation time. The initial diffusion direction is determined based on the direction of response range expansion (e.g., along valleys, towards steeper slopes, etc.). The location of this sphere, the diffusion initiation time, and the initial diffusion direction are integrated to generate diffusion initiation node information.

[0095] Step S1211: Integrate response trigger node information, energy accumulation node information, and diffusion initiation node information to generate a set of glacial debris flow risk evolution nodes.

[0096] The previously generated response triggering node information, energy accumulation node information, and diffusion initiation node information are summarized and integrated. These node information respectively characterize the initial triggering stage, energy accumulation stage, and diffusion initiation stage of glacial debris flow risk evolution. The integrated set of glacial debris flow risk evolution nodes comprehensively reflects the key node information from triggering to energy accumulation and then to the start of diffusion of glacial debris flow risk.

[0097] Step S130: Link the set of risk evolution nodes of glacial debris flow with the topographic features of the glacial region to generate a dynamic risk evolution path of glacial debris flow.

[0098] Step S131: Collect topographic feature data of the glacier area and generate a topographic feature dataset of the glacier area. The topographic feature data includes surface slope, surface undulation, valley distribution, elevation distribution, and slope aspect distribution.

[0099] Topographic data for the glacier region was collected using methods such as UAV aerial surveying, satellite remote sensing, and ground surveying. Surface slope was calculated by determining the angle of inclination at a point on the surface, typically expressed as a percentage or degrees. Surface undulation was described by analyzing the variations in topography, such as mountains, valleys, and flat areas. Valley distribution included the number, location, direction, length, width, depth, and connectivity between valleys. Elevation distribution referred to the altitude data of points within the region. Aspect distribution indicated the direction of surface inclination, such as east, south, west, north, and northeast. This collected topographic data was then organized and digitized to generate a topographic feature dataset for the glacier region.

[0100] Step S132: The topographic feature dataset of the glacier area is structured by associating the elevation, slope, aspect, undulation morphology and valley distribution parameters of each geographical location to generate topographic feature association data indexed by geographical location.

[0101] The topographic feature dataset of the glacier region is structured by first dividing the entire glacier region into multiple small geographic location units (e.g., grid cells at certain latitude and longitude intervals). For each geographic location unit, its corresponding elevation, slope, aspect, undulation morphology (e.g., using codes to represent different undulation types), and valley distribution parameters within or around the unit (e.g., whether there are valleys passing through it, the distance to the nearest valley, etc.) are extracted from the topographic feature dataset. These parameters are then associated with the coordinate information of the geographic location unit to form a record. All records of geographic location units are indexed by geographic location coordinates to facilitate subsequent queries and matching, thereby generating topographic feature associated data.

[0102] Step S133: Extract the location information of each risk evolution node from the set of risk evolution nodes of glacier debris flow, perform spatial matching between the location information of each risk evolution node and the terrain feature association data, determine the terrain feature parameters corresponding to each risk evolution node, and generate a node terrain adaptation dataset.

[0103] From the response triggering node information, energy accumulation node information, and diffusion initiation node information in the glacial debris flow risk evolution node set, the location information (e.g., geographic coordinates) of each node is extracted. These location coordinates are then spatially matched with the geographic location index in the terrain feature association data to find the geographic location unit where each node's location coordinates are located. Next, terrain feature parameters such as elevation, slope, aspect, undulation morphology, and gully distribution parameters corresponding to this geographic location unit are extracted from the terrain feature association data. The identifier of each risk evolution node is associated with its corresponding terrain feature parameters to generate a node terrain adaptation dataset.

[0104] Step S134: Extract the terrain feature parameters corresponding to the response trigger node from the node terrain adaptation dataset, analyze the correlation between the terrain parameters and the magnitude and duration of the change in response features at the response trigger node, and generate terrain impact data of the trigger node; based on the terrain impact data of the trigger node, supplement the terrain association features in the response trigger node information and improve the evolution environment description of the response trigger node.

[0105] The terrain feature parameters corresponding to the response trigger nodes are identified from the node terrain adaptation dataset. The effects of these terrain parameters (such as slope, surface undulation complexity, and location near valleys) on the magnitude and duration of response feature changes at the trigger nodes are analyzed. For example, steeper slopes may lead to greater magnitude and shorter duration of response feature changes; nodes located near valleys may experience increased response duration due to the amplification effect of the valleys. Through statistical analysis and correlation studies, the relationship between terrain parameters and the magnitude and duration of response feature changes is determined, generating terrain influence data for the trigger nodes. This terrain influence data is then incorporated as terrain-related features into the response trigger node information, ensuring that the evolutionary environment description of the response trigger nodes includes not only layered structure and energy state but also the influence of terrain factors.

[0106] Step S135: Extract the terrain feature parameters corresponding to the energy accumulation nodes from the node terrain adaptation dataset, analyze the correlation between the terrain parameters and the energy accumulation rate and total amount at the energy accumulation nodes, and generate the terrain impact data of the accumulation nodes; based on the terrain impact data of the accumulation nodes, supplement the terrain association features in the energy accumulation node information and improve the evolutionary environment description of the energy accumulation nodes.

[0107] Similarly, terrain feature parameters corresponding to energy accumulation nodes are extracted from the node terrain adaptation dataset. The impact of these terrain parameters (such as elevation differences, surface undulation, and valley density) on the energy accumulation rate and total energy accumulation at the energy accumulation nodes is analyzed. For example, areas with large elevation differences and dramatic surface undulations may have a faster energy accumulation rate due to differences in the potential energy of material movement; areas with dense valley distribution may be conducive to energy accumulation, thereby increasing the total energy accumulation. The correlation between terrain parameters and energy accumulation rate and total energy accumulation is determined, generating terrain impact data for accumulation nodes. This data is then used as terrain-related features to supplement the energy accumulation node information, improving the description of the evolutionary environment of the energy accumulation nodes.

[0108] Step S136: Extract the terrain feature parameters corresponding to the diffusion initiation node from the node terrain adaptation dataset, analyze the correlation between the terrain parameters and the diffusion direction and diffusion rate at the diffusion initiation node, and generate diffusion node terrain influence data; based on the diffusion node terrain influence data, supplement the terrain association features in the diffusion initiation node information to improve the evolutionary environment description of the diffusion initiation node.

[0109] Step S1361: Extract the terrain feature parameters corresponding to the diffusion starting node from the node terrain adaptation dataset to generate the diffusion node terrain parameter set.

[0110] From the node terrain adaptation dataset, based on the identifier of the diffusion starting node, the terrain feature parameters corresponding to that node are extracted, including specific parameter values ​​such as surface slope, surface undulation, valley distribution, elevation distribution, and aspect distribution. These parameters are then integrated to generate the diffusion node terrain parameter set.

[0111] Step S1362: Analyze the surface slope parameters in the topographic parameter set of diffusion nodes, determine the slope change trend, and generate slope trend analysis data. The slope change trend directly affects the movement speed of the diffused material.

[0112] The surface slope parameters of the topographic parameter set of the diffusion nodes are analyzed, and the slope values ​​in different directions within a certain range around the diffusion initiation node are calculated. By comparing the magnitude and variation of these slope values, the trend of slope change is determined, such as whether it gradually increases or decreases from the node location in a certain direction, or whether it increases first and then decreases. The description of the slope change trend is recorded to generate slope trend analysis data. Since the greater the slope, the stronger the acceleration effect of gravity on the diffused material, and the faster the movement speed of the diffused material, the slope change trend directly affects the movement speed of the diffused material.

[0113] Step S1363: Combining slope trend analysis data with the principles of material motion mechanics, determine the influence of surface slope on diffusion rate and generate slope rate influence data.

[0114] Based on the slope trend analysis data, and combining relevant principles of material kinetics (such as the relationship between the component of gravity along the slope and friction and viscosity), the changes in the net force acting on diffusing materials under different slope conditions are analyzed to determine the specific impact of surface slope on the diffusion rate. For example, the diffusion rate may gradually increase in the direction of increasing slope; in areas with steeper slopes, the diffusion rate may be higher. These relationships (such as the positive correlation between slope value and diffusion rate, and the rate range corresponding to different slope intervals) are recorded to generate slope-rate impact data.

[0115] Step S1364: Analyze the valley distribution parameters in the terrain parameter set of the diffusion nodes, extract the direction parameters and connectivity parameters of the main valley, and generate valley guidance feature data.

[0116] Valley distribution parameters are extracted from the topographic parameter set of diffusion nodes, including the location, number, orientation, length, width, depth, and connectivity between valleys. By analyzing these parameters, the main valleys (usually larger and more connected) that are likely to play a major guiding role in the diffusion process are identified. Orientation parameters (such as orientation angle) and connectivity parameters (such as the degree of connection between the main valley and other valleys or downstream areas) are extracted from these main valleys. These parameters are then integrated to generate valley guidance feature data.

[0117] Step S1365: Based on the valley guidance feature data, identify the valleys that play a major guiding role in the diffusion direction, determine the direction of the valleys, and generate the main diffusion direction guidance data.

[0118] Based on the main valley orientation and connectivity parameters in the valley guidance characteristic data, the potential guiding capacity of each main valley in the diffusion direction is assessed. Generally, main valleys with clear orientations and good connectivity have a stronger guiding effect on the diffusion direction. Through comparative analysis, the valleys that play a major guiding role in the diffusion direction are identified, and their specific orientations are determined. The orientation information of these valleys is recorded to generate main diffusion direction guidance data, which indicates the direction that diffusing substances are likely to preferentially follow.

[0119] Step S1366: Analyze the surface undulation morphology parameters in the topographic parameter set of diffusion nodes, identify the raised and low-lying areas around the diffusion initiation node, and generate undulation influence feature data.

[0120] By analyzing the surface undulation parameters in the topographic parameter set of diffusion nodes, and through the analysis of elevation data, raised areas (such as small hills and ridges) and low-lying areas (such as depressions and gullies) within a certain range around the diffusion initiation node are identified. The location, size, and relative height of these raised and low-lying areas are recorded to generate undulation influence characteristic data. These undulation patterns can hinder or guide the movement direction of diffused materials; for example, raised areas may deflect the diffusion direction, while low-lying areas may attract diffused materials to converge towards them.

[0121] Step S1367: Combine the undulation impact feature data, analyze the diffusion direction deflection angle and diffusion rate adjustment coefficient caused by terrain undulation, and generate undulation diffusion impact data.

[0122] For the raised and lower-lying areas identified in the undulation influence characteristic data, their specific impact on diffusion direction and diffusion rate is analyzed. For example, when diffusing material encounters a raised area, it may deflect to either side of the raised area; the deflection angle is related to the height and slope of the raised area. When diffusing material enters a lower-lying area, its velocity may change due to open terrain or increased friction, thus requiring adjustment of the diffusion rate. The adjustment coefficient is determined based on the morphological characteristics of the lower-lying area. The analyzed diffusion direction deflection angle and diffusion rate adjustment coefficient are recorded to generate undulation diffusion influence data.

[0123] Step S1368: Analyze the slope aspect distribution parameters in the topographic parameter set of diffusion nodes, determine the angle relationship between the slope aspect and the diffusion initiation direction, and generate slope aspect direction influence data.

[0124] Slope aspect refers to the direction of the land surface inclination, and the diffusion initiation direction is the initially determined diffusion direction. The slope aspect distribution parameters in the topographic parameter set of the diffusion nodes are analyzed to determine the slope aspect at the diffusion initiation node and the slope aspect of different surrounding areas. The angle between the slope aspect and the diffusion initiation direction is calculated, and the influence of this angle on the diffusion direction is analyzed. For example, if the slope aspect is consistent with the diffusion initiation direction, it may strengthen diffusion in that direction; if the angle is large, it may cause the diffusion direction to deflect to some extent towards the slope aspect. The relationship between the slope aspect and the diffusion initiation direction, and its influence on the diffusion direction, are described and recorded to generate slope aspect influence data.

[0125] Step S1369: Integrate slope rate influence data, main diffusion direction guidance data, undulation diffusion influence data, and slope aspect direction influence data to generate diffusion node terrain influence data.

[0126] Data on the influence of slope rate (reflecting the impact of slope on diffusion rate), data on the guiding role of the main diffusion direction (reflecting the guiding effect of the main valley on the diffusion direction), data on the influence of undulation diffusion (reflecting the impact of surface undulation on diffusion direction and rate), and data on the influence of slope aspect (reflecting the influence of slope aspect on diffusion direction) are comprehensively integrated. During the integration process, the interactions and weighting relationships between the various influencing factors are analyzed; for example, the guiding effect of the main valley may be greater than the influence of slope aspect. The resulting topographic influence data for diffusion nodes comprehensively reflects the combined impact of topographic features on the diffusion direction and rate at the diffusion initiation node.

[0127] Step S137: Integrate the improved response trigger node information, energy accumulation node information, and diffusion initiation node information, and combine the response transmission relationship between each risk evolution node with topographic impact data to construct a spatial process correlation framework for the dynamic risk evolution of glacial debris flows.

[0128] This study integrates information on response triggering nodes, energy accumulation nodes, and diffusion initiation nodes, supplemented with topographic features. Based on the layered response transmission relationship data in the seismic signal-glacier response chain information, the response transmission relationships between each risk evolution node are clarified (e.g., how the response triggering node transmits to the energy accumulation node, and how the energy accumulation node develops to the diffusion initiation node). Simultaneously, combining the topographic impact data of each node (triggering node topographic impact data, accumulation node topographic impact data, and diffusion node topographic impact data), the role of topographic factors in the response transmission process between nodes is analyzed. Based on this, a spatial process correlation framework is constructed that reflects how glacial debris flow risk evolves and develops between different nodes and at different spatial locations. This spatial process correlation framework includes elements such as node location, inter-node transmission paths, and topographic influencing factors.

[0129] Step S138: Supplement the energy state data, response characteristic data, and terrain characteristic data of each risk evolution node in the spatial process association framework to generate a dynamic risk evolution path for glacial debris flow. This dynamic risk evolution path for glacial debris flow is used to present the development process and key characteristics of glacial debris flow risk at different spatial locations and different evolution stages.

[0130] Based on the spatial process correlation framework, each risk evolution node is supplemented with its energy state data (such as the initial energy state data of the response trigger node and the energy accumulation state data of the energy accumulation node), response characteristic data (such as the response change pattern, response duration, and response range of each node), and topographic feature data (obtained from the node topographic adaptation dataset). This data further enriches the framework's content, enabling it to dynamically display the development process of glacial debris flow risk from the response trigger node, through the energy accumulation node, to the diffusion initiation node, at different spatial locations (different spheres, different topographic regions) and at different evolutionary stages (triggering, accumulation, diffusion), as well as the key characteristics of each process (energy changes, response characteristics, topographic influences, etc.). This improved framework is called the dynamic risk evolution path of glacial debris flow.

[0131] Step S140: Generate dynamic risk warning content for glacier debris flows based on the dynamic risk evolution path of glacier debris flows.

[0132] Step S141: Analyze the dynamic risk evolution path of glacial debris flow, extract energy state data, response characteristic data, and topographic impact data of each risk evolution node, determine the risk development stage corresponding to each risk evolution node, and generate risk stage division data. The risk stage division data divides the risk evolution process of glacial debris flow into the initial triggering stage, the energy accumulation stage, and the diffusion development stage.

[0133] The dynamic risk evolution path of glacial debris flows was analyzed. Energy state data (initial energy state, energy accumulation rate, and total amount, etc.), response characteristic data (response change pattern, duration, and range of occurrence, etc.), and topographic impact data were extracted from response triggering nodes, energy accumulation nodes, and diffusion initiation nodes, respectively. Based on the characteristics of these data and the evolutionary relationships between nodes, the risk evolution process of glacial debris flows was divided into different risk development stages. The stage corresponding to the response triggering node is the initial triggering stage, characterized by the triggering of an abnormal response and the initial accumulation of energy. The stage corresponding to the energy accumulation node is the energy accumulation stage, in which energy continuously accumulates and reaches a critical state. The stage corresponding to the diffusion initiation node is the diffusion development stage, in which energy begins to be released and material begins to diffuse outward. The division of the above stages (including the starting nodes and main characteristics of each stage) was recorded to generate risk stage classification data.

[0134] Step S142: Extract key feature parameters for each risk development stage from the risk stage segmentation data, and generate key feature data for each risk development stage. The key feature parameters include total energy accumulation, response diffusion range, and evolution duration.

[0135] For each risk development stage (initial triggering stage, energy accumulation stage, and diffusion development stage) in the risk stage segmentation data, key feature parameters are extracted from the corresponding node information and evolution path data. Key feature parameters for the initial triggering stage may include the initial total energy accumulation, the initial response range, and the duration of the triggering stage evolution. Key feature parameters for the energy accumulation stage include the total energy accumulation at the end of this stage, the expansion of the response range, and the duration of the accumulation stage evolution. Key feature parameters for the diffusion development stage include the total energy at the start of diffusion, the response diffusion range (i.e., the diffusion range), the duration of the diffusion stage, and the expected duration. These key feature parameters are then organized by stage to generate key feature data for each risk development stage.

[0136] Step S143: Collect human geography data within the glacier area to generate a human geography dataset for the glacier area. The human geography data includes population distribution, infrastructure location, and the distribution range of important sites.

[0137] Human geography data for the glacier region was collected through various channels, including statistical data provided by government departments, field surveys, and map data. Population distribution data included the location, number, and resident population of settlements; infrastructure distribution data included the specific coordinates and distribution range of roads, bridges, power lines, communication base stations, water conservancy facilities, and buildings (such as houses, schools, and hospitals); and the distribution range of important locations included the boundaries of special areas such as nature reserves, tourist attractions, industrial parks, and military management zones. This data was then organized and digitized to generate a human geography dataset for the glacier region.

[0138] Step S144: Spatial information in the dynamic risk evolution path of glacial debris flow is spatially overlaid with the human geography dataset of the glacial region to determine the human geography region that may be affected by each risk development stage and generate risk impact area data.

[0139] The dynamic risk evolution path of glacial debris flows includes spatial evolution information for each risk development stage, such as the change in the diffusion range over time during the diffusion development stage. This spatial information (e.g., polygonal regions representing the impact range at different stages) is spatially overlaid with a human geography dataset of the glacier region. Using the overlay function of Geographic Information System (GIS) software, it is determined whether each human geography element (settlement, infrastructure, important locations, etc.) in the human geography dataset is located within the impact range indicated by the risk evolution path. For each risk development stage, the potentially affected human geography regions are identified, i.e., the regions containing these human geography elements. The boundaries of these regions and the types of human geography elements they contain are recorded to generate risk impact area data.

[0140] Step S145: Extract the population distribution and infrastructure location within the affected areas at each stage from the risk impact area data, present the key areas and targets that need to be focused on at each stage, and generate key target data.

[0141] From the risk-affected area data, for each stage of risk development, extract the population distribution (e.g., settlement name, location, number of residents) and infrastructure distribution (e.g., road name, bridge number, building use and location) within the affected area. Based on population size, the importance of infrastructure (e.g., hospitals, schools are critical infrastructure), and the likelihood of being affected, identify key areas (e.g., densely populated settlements, areas with concentrated critical infrastructure) and specific targets (e.g., a bridge, a village) requiring focused attention at each stage. Clearly present these key areas and targets to generate data on key targets for focus.

[0142] Step S146: For each risk development stage, perform spatial overlay analysis on the response diffusion range in the key characteristic data of the risk development stage and the target distribution range in the key target data of the risk development stage. Combine the total energy accumulation and evolution duration of the risk development stage to assess the potential impact on the key target and generate stage impact analysis data.

[0143] For example, step S1461: Extract the total energy accumulation, response diffusion range, and evolution duration of each risk development stage from the key characteristic data of each risk development stage, classify and organize them according to the risk development stage, and generate key parameter data for each stage.

[0144] From the key characteristic data of each risk development stage, three key parameters are extracted: total energy accumulation, response diffusion range, and evolution duration for the initial triggering stage, energy accumulation stage, and diffusion development stage. These parameters are then categorized according to the risk development stage, forming key parameter combinations for each stage, such as "Initial Triggering Stage: Total Energy Accumulation E1, Response Diffusion Range R1, Evolution Duration T1," "Energy Accumulation Stage: Total Energy Accumulation E2, Response Diffusion Range R2, Evolution Duration T2," and "Diffusion Development Stage: Total Energy Accumulation E3, Response Diffusion Range R3, Evolution Duration T3," etc. The categorized and organized parameter data are then used to generate stage-specific key parameter data.

[0145] Step S1462: Extract the key areas and targets corresponding to each risk development stage from the key target data, present the type and distribution range of each target, and generate detailed data of the target focus for each stage.

[0146] For each risk development stage in the data of key targets, extract the corresponding key focus areas (e.g., area A, area B) and key targets (e.g., residential area A, bridge B, school C, etc.). Clarify the type of each key target (e.g., residential area, bridge, school, hospital, etc.) and its distribution range (e.g., the specific coordinate range or area boundary of the target). Organize the above information by stage to generate detailed data on key targets for each stage, such as "Initial trigger stage key focus area: area A, key target: residential area A (type: residential area, distribution range: coordinates X1-Y1 to X2-Y2)".

[0147] Step S1463: Perform spatial overlay analysis on the response diffusion range in the key parameter data of the stage and the target distribution range in the detailed data of the target of concern in the stage to determine whether each key target of concern is within the response diffusion range of the risk development stage, and generate target coverage data.

[0148] Using GIS software, spatial overlay analysis is performed on the response diffusion range (represented by spatial polygons) of a certain risk development stage in the key parameter data of the stage and the distribution range (also represented by spatial polygons or point coordinates) of the key targets of concern in that stage in the detailed data of the stage's concerns. By determining whether the target distribution range intersects with the response diffusion range, it is determined whether each key target of concern is within the response diffusion range of that risk development stage. For example, if the distribution range polygon of settlement A overlaps with the response diffusion range polygon, it indicates that the settlement is within the response diffusion range. The coverage status of each key target of concern (whether it is covered, the coverage ratio, etc.) is recorded to generate target coverage data.

[0149] Step S1464: For key targets within the response diffusion range, extract the type parameters of the key targets, determine the vulnerability level of the key targets based on the preset vulnerability levels of various targets, and generate target vulnerability data.

[0150] For key targets within the response diffusion range in the target coverage data, extract the target type parameter (e.g., residential areas, schools, bridges, hospitals, etc.) from the phased target detail data. A set of vulnerability level evaluation criteria for each type of target is pre-defined. This vulnerability level evaluation criteria are formulated based on factors such as the target's structural strength, importance of its purpose, and risk resistance. For example, hospitals and schools, which are densely populated and important facilities, are classified as high vulnerable; sturdy bridges are classified as medium vulnerable; and temporary buildings are classified as low vulnerable. Based on this evaluation criteria, the vulnerability level of each key target within the response diffusion range is assessed, generating target vulnerability data, such as "Residential Area A: High Vulnerability Level" and "Bridge B: Medium Vulnerability Level."

[0151] Step S1465: Combining the total energy accumulation data in the key parameter data of the stage with the target vulnerability data, assess the potential damage level of the risk energy to the key targets of concern, and generate energy damage correlation data.

[0152] This method combines the total energy accumulation at each stage of risk development from the key parameter data with the vulnerability levels of key targets from the target vulnerability data. Generally, the larger the total energy accumulation, the stronger the potential destructive power to the target; the higher the target vulnerability level, the more easily it is damaged under the same energy. Based on a pre-defined damage level evaluation matrix corresponding to the combination of energy and vulnerability level (e.g., high energy accumulation and high vulnerability level correspond to extremely high damage level, medium energy and medium vulnerability correspond to medium damage level, etc.), the potential damage level that each key target may suffer is assessed. The assessment results are recorded to generate energy-damage correlation data, such as "Residential Site A, Total Energy Accumulation E, High Vulnerability Level, Extremely High Damage Level".

[0153] Step S1466: Based on the evolution duration in the key parameter data of the stage, assess the cumulative effect level of the duration of risk impact on the key focus targets, and generate duration cumulative impact data.

[0154] The evolution duration in the key parameter data of each stage reflects the potential length of time that risk development stage may last. The longer the duration of the risk impact, the more severe the cumulative effects (such as repeated impacts, prolonged immersion, and continuous vibration) on key targets of concern may be. Based on the duration of evolution and the type of target, a pre-defined cumulative effect level evaluation standard is established. For example, if the duration exceeds a certain threshold and the target is a vulnerable type (such as earthen houses), the cumulative effect level is high. Based on this standard, the cumulative effect level of each key target of concern is assessed, generating duration-based cumulative impact data, such as "Residential site A, evolution duration T, high cumulative effect level".

[0155] Step S1467: Integrate target coverage data, target vulnerability data, energy destruction correlation data, and cumulative impact data over time to calculate the comprehensive impact level of each key target at this risk development stage and generate detailed target impact data.

[0156] This process integrates data on target coverage (whether the target is within the diffusion range), target vulnerability (vulnerability level), energy damage correlation (potential damage level), and cumulative impact over time (cumulative effect level). A weight is assigned to each data item, and the comprehensive impact level of each key target at this stage of risk development is calculated using weighted summation or other comprehensive evaluation methods. The comprehensive impact level can be categorized into extremely high, high, medium, and low levels, comprehensively reflecting the severity of the risk impact on the target. The comprehensive impact level and various sub-indicators for each target are recorded to generate detailed target impact data.

[0157] Step S1468: Summarize the detailed impact data of all key targets within each risk development stage, count the number and distribution of affected targets, assess the overall severity level of the risk impact, and generate overall impact data for the stage.

[0158] Summarize the detailed impact data of all key targets within each risk development stage. Calculate the number of targets with an overall impact level of extremely high, high, medium, and low within that stage, and their distribution range (e.g., concentrated in which regions). Based on factors such as the number of affected targets, the proportion of high-level impact targets, and the size of their distribution range, assess the overall severity level of the risk impact at that stage (e.g., severe, relatively severe, moderate, minor). Record the statistical results and assessment levels to generate overall impact data for that stage.

[0159] Step S1469: Combine the topographic feature data of the glacier area to determine the amplification or weakening coefficient of the topography on the risk impact and generate topographic impact adjustment data; based on the topographic impact adjustment data, adjust the impact range and severity level in the overall impact data of the stage to generate stage impact analysis data.

[0160] Topographic feature data (such as slope, gully density, and elevation difference) within the risk impact area are extracted from the topographic feature dataset of the glacier region. This analysis examines how these topographic features affect the movement and destructive force of glacial debris flows. For example, steep terrain may amplify the flow velocity and impact force of debris flows, thus increasing the risk impact; while gentle, open terrain may weaken the energy of debris flows, reducing the risk impact. Based on the different topographic features, corresponding risk impact amplification or weakening coefficients are determined, generating topographic impact adjustment data. This adjusted data is then applied to the overall impact data for each stage, adjusting the original impact range (which may expand or shrink after considering the guiding effect of terrain) and severity level (which may increase the severity level under the amplification effect of terrain), ultimately generating stage impact analysis data.

[0161] Step S147: Based on the stage impact analysis data, formulate corresponding response measures for each stage of risk development and generate stage response measure data. The response measures cover the path planning and timing selection related to personnel transfer, and the formulation and implementation process of infrastructure protection measures.

[0162] Based on the key concerns, comprehensive impact level, and overall severity level of each risk development stage in the phased impact analysis data, response measures are formulated for each stage. For the initial triggering stage, possible responses include strengthening monitoring and early warning, and notifying personnel in relevant areas to prepare. For the energy accumulation stage, it may be necessary to begin organizing the evacuation of personnel from highly vulnerable target areas and to conduct initial reinforcement of critical infrastructure. For the diffusion and development stage, a comprehensive evacuation of personnel is required, emergency response plans must be activated, and emergency protection measures must be taken for infrastructure. Route planning for personnel evacuation should consider terrain conditions and road capacity, selecting safe and efficient routes; timing should be based on risk trend forecast data to ensure that evacuation is completed before the risk impacts the target area. Measures related to infrastructure protection should be formulated according to the type of facility and the type of risk faced (such as impact, flooding, vibration, etc.), with specific methods determined, and the implementation process clearly defining the responsible parties, steps, and time requirements. These response measures are then organized by stage to generate phased response measure data.

[0163] Step S148: Extract the rate data of risk evolution from the dynamic risk evolution path of glacial debris flow, combine it with the key characteristic data of each risk development stage, predict the duration of each risk development stage and the development trend of subsequent stages, and generate risk trend prediction data.

[0164] This study analyzes the rate data of risk evolution from the dynamic risk evolution path of glacial debris flows, such as the growth rate of response characteristic changes, energy accumulation rate, and diffusion range expansion rate. Combining parameters like evolution duration, total energy accumulation, and response diffusion range from key characteristic data of each risk development stage, methods such as trend extrapolation and dynamic models are used to predict how long the current risk development stage will continue, and how energy will change, how the diffusion range will expand, and how the affected area will change in subsequent stages. For example, based on the current energy accumulation rate and the target energy value of the energy accumulation stage, the study predicts the duration of the energy accumulation stage; based on the current diffusion rate, it predicts the change curve of the diffusion range over time in the diffusion development stage. These prediction results are recorded to generate risk trend prediction data.

[0165] Step S149: Integrate risk stage classification data, key characteristic data of each risk development stage, risk impact area data, key focus target data, stage impact analysis data, stage response measure data, and risk trend prediction data to construct a framework for dynamic risk early warning content; establish the correspondence between stage response measure data and each impact area in stage impact analysis data to generate correlation mapping data; integrate the above data and correlation mapping data to generate dynamic risk early warning content for glacier debris flows.

[0166] First, data on risk stage classification, key characteristic data for each risk development stage, data on risk-affected areas, data on key targets, data on stage-specific impact analysis, data on stage-specific response measures, and risk trend prediction are integrated to construct a basic framework for dynamic risk early warning content. This basic framework includes all the elements of early warning. Then, the correspondence between various response measures in the stage-specific response measures data and the various affected areas (or key targets) in the stage-specific impact analysis data is established, clarifying which measures are applicable to which areas or targets, and generating correlation mapping data. Finally, the correlation mapping data is also integrated into the framework of dynamic risk early warning content, so that the early warning content not only includes risk information but also clarifies specific response measures for different regions and targets, forming a complete dynamic risk early warning content for glacial debris flows.

[0167] Step S150: The on-site response data of the dynamic risk warning content is back-linked to the glacier response chain construction process of the seismic signal to complete the optimization and update of the glacier response chain information of the seismic signal.

[0168] After the release of dynamic risk warnings, on-site response data is collected through channels such as field observation, emergency command feedback, and reports from affected areas. This on-site response data includes the actual situation at each risk development stage after the warning release (e.g., actual impact range, impact degree, and effectiveness of response measures), and deviations from predicted data. The aforementioned on-site response data is then inversely correlated with each stage of the glacier response chain construction process for earthquake signals. For example, the actual energy accumulation is compared with the analysis process of energy accumulation nodes in the response chain, and the actual diffusion path is compared with the propagation path analysis and topographic impact analysis in the response chain. Potential errors or deficiencies in the construction process are identified (e.g., inaccurate extraction of response features for certain spheres, discrepancies between propagation path simulation and reality, and incomplete consideration of topographic influence factors). Based on the on-site response data, relevant parameters, models, and data in the glacier response chain information for earthquake signals (e.g., earthquake signal sphere propagation path data, sets of earthquake response features for each sphere, and sphere response transmission relationship data) are adjusted, optimized, and updated to improve the accuracy and reliability of subsequent warnings.

[0169] In one exemplary embodiment, a dynamic risk early warning system for glacial debris flows based on seismic signals is provided. This system can be a terminal, server, etc., and its internal structure diagram can be as follows: Figure 2 As shown, this seismic signal-based dynamic risk early warning system for glacial debris flows includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, near-field communication, or other technologies. When the computer program is executed by the processor, it implements a seismic signal-based dynamic risk early warning method for glacial debris flows. The display unit is used to generate a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, or a button, trackball, or touchpad set on the shell of the dynamic risk early warning system for glaciers and debris flows based on earthquake signals. It can also be an external keyboard, touchpad, or mouse, etc.

[0170] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A dynamic risk early warning method for glacial debris flows based on seismic signals, characterized in that, The method includes: A glacier response chain for seismic signals is constructed, and seismic signal glacier response chain information is generated. The seismic signal is a continuous seismic wave signal in and around the glacier region. The seismic signal glacier response chain information includes the seismic signal propagation path, the response sequence of each glacier layer, and the transmission relationship of response characteristics. The risk evolution nodes in the glacier response chain information of the seismic signal are analyzed to generate a set of glacier debris flow risk evolution nodes, which includes response triggering nodes, energy accumulation nodes, and diffusion initiation nodes. By linking the set of risk evolution nodes of glacial debris flows with the topographic features of glacial areas, a dynamic risk evolution path for glacial debris flows is generated; Based on the dynamic risk evolution path of glacial debris flow, generate dynamic risk warning content for glacial debris flow; By inversely linking the on-site response data of dynamic risk warning content to the glacier response chain construction process of seismic signals, the optimization and updating of glacier response chain information of seismic signals is completed.

2. The method for dynamic risk early warning of glacial debris flow based on seismic signals according to claim 1, characterized in that, The construction of the glacier response chain for seismic signals, generating seismic signal glacier response chain information, includes: Seismic signals from the glacier region and its surroundings are continuously collected and processed sequentially according to the time of collection to generate a time-series seismic signal sequence containing seismic signals from multiple consecutive time periods. Each time period in the time-series seismic signal sequence is associated with a corresponding collection time identifier, which is used to establish a temporal correspondence with the environmental response data of the glacier region. Waveform analysis processing is performed on the seismic signals of each time period in the time-series seismic signal sequence to extract waveform-related features such as waveform fluctuation morphology, signal change amplitude, signal duration, and signal propagation speed of the seismic signals of each time period, generating a set of waveform features for each time period. The set of waveform features for each time period is associated with the corresponding time period identifier to distinguish the signal features of different time periods. Data on the layered structure of glaciers is collected to generate a glacier layered structure dataset. The data on the layered structure of glaciers covers the distribution range, structural characteristics, and material composition of the glacier surface layer, the glacier middle layer, the glacier bottom layer, the glacier surrounding strata, and the glacier surrounding water body layer. Based on the signal propagation velocity characteristics in the time-series seismic signal sequence, and combined with the material composition and structural characteristics of each layer in the glacier stratum structure dataset, the propagation path of seismic signals in each glacier stratum and surrounding areas is analyzed, the order in which seismic signals enter each glacier stratum from the periphery of the glacier region is determined, and seismic signal stratum propagation path data is generated. Under the continuous monitoring of time-series seismic signal sequences, the state changes of each sphere in the glacier stratum structure data are captured, and the state change information of each sphere under the action of seismic signals at different time periods is captured. The state change information covers changes in structural integrity, changes in material motion state, and changes in spatial position displacement. Seismic response data of each sphere is generated, and the seismic response data of each sphere is associated with the corresponding seismic signal time period identifier. Response features are extracted from the seismic response data of each sphere to generate a set of seismic response features for each sphere. The response features cover the time point of the response, the duration of the response, the form of the response change, and the range of the response. The set of seismic response features for each sphere is associated with the corresponding sphere identifier and time segment identifier. By comparing the response occurrence time in the seismic response feature set of each sphere with the signal arrival time in the seismic signal propagation path data of each sphere, the response delay period of each sphere to the seismic signal is divided, and sphere response delay data is generated. The sphere response delay data is used to characterize the time interval between the arrival of the seismic signal and the generation of the response in the sphere. Based on the layer response delay data and the seismic signal layer propagation path data, the response order of each glacier layer to the seismic signal is determined, the transmission relationship between the response characteristics of each layer is integrated, and layer response transmission relationship data is generated. By integrating seismic signal propagation path data, seismic response feature sets of each sphere, response delay data of each sphere, and response transmission relationship data of each sphere, a preliminary seismic signal glacier response chain is constructed. The preliminary seismic signal glacier response chain presents the correlation logic between the propagation of seismic signals in the glacier region and the responses of each sphere. The correlation between each link in the preliminary seismic signal-glacier response chain is analyzed, the intermediate action characteristics in the response transmission process are extracted, and detailed response chain data are generated. The seismic signal propagation path data of each sphere, the seismic response characteristic set of each sphere, the response delay data of each sphere, the response transmission relationship data of each sphere, and the detailed response chain data are integrated to construct the seismic signal-glacier response chain information.

3. The method according to claim 1, wherein, The risk evolution nodes in the glacier response chain information of the analyzed seismic signal are used to generate a set of glacier debris flow risk evolution nodes, including: From the earthquake signal glacier response chain information, extract the earthquake response feature set of each sphere, filter the response features in each sphere earthquake response feature set whose response change mode exceeds the threshold of the normal fluctuation range, and generate an abnormal response feature set; Extract the seismic signal waveform features corresponding to the abnormal response feature set, determine the key seismic signal features that trigger the abnormal response, associate the concentric circle identifiers corresponding to the abnormal response features with the response occurrence time points, and generate abnormal response association data; Based on the correlation data of abnormal responses, the location of the glacier layer that generated the abnormal response is located, the glacier layer structure characteristics and material composition at that location are presented, the initial triggering location of risk evolution is determined, and the location information of the response triggering node is generated. Extract the response duration and response change amplitude from the seismic response characteristics corresponding to the location of the response trigger node, analyze the initial energy accumulation at the response trigger node, and generate the initial energy state data of the response trigger node. By integrating the location information of the response triggering node with the initial energy state data, response triggering node information is generated, which characterizes the initial triggering link in the evolution of glacial debris flow risk. Extract layer response transmission relationship data from seismic signal glacier response chain information, track the transmission process of response characteristics generated by response triggering nodes in each subsequent layer, record the changes of response characteristics in the transmission process, and generate response transmission tracking data. Analyze response transmission tracking data to identify the locations of concentric zones where response characteristics change more significantly or last longer during transmission, and generate energy accumulation area identification data; Collect the layer structure data and material motion state data corresponding to the energy accumulation area identification data, analyze the energy accumulation rate and total accumulation in the energy accumulation area, and generate energy accumulation state data. By integrating energy accumulation area identification data and energy accumulation status data, energy accumulation node information is generated. This energy accumulation node information represents the key link in the continuous energy accumulation process during the risk evolution of glacial debris flows. The response range data of each layer of seismic response feature set is extracted from the seismic signal glacier response chain information. Combined with the energy accumulation node information, the layer location where the response range begins to spread outward after the energy accumulation reaches a set level is identified. The diffusion start time and initial diffusion direction of the layer location are determined, and diffusion start node information is generated. By integrating response trigger node information, energy accumulation node information, and diffusion initiation node information, a set of nodes for the risk evolution of glacial debris flows is generated.

4. The method of claim 1, wherein the method further comprises: The set of linked glacial debris flow risk evolution nodes and the topographic features of the glacial region generate a dynamic risk evolution path for glacial debris flows, including: Collect topographic feature data of glacier areas to generate a topographic feature dataset of glacier areas. The topographic feature data includes surface slope, surface undulation, valley distribution, elevation distribution, and slope aspect distribution. The topographic feature dataset of the glacier region is structured, and the elevation, slope, aspect, undulation morphology and valley distribution parameters of each geographical location are associated to generate topographic feature association data indexed by geographical location. The location information of each risk evolution node is extracted from the set of risk evolution nodes of glacial debris flow. The location information of each risk evolution node is spatially matched with the terrain feature association data to determine the terrain feature parameters corresponding to each risk evolution node and generate a node terrain adaptation dataset. Extract terrain feature parameters corresponding to response trigger nodes from the node terrain adaptation dataset, analyze the correlation between terrain parameters and the magnitude and duration of response feature changes at response trigger nodes, and generate terrain impact data of trigger nodes. Based on the terrain impact data of the trigger nodes, supplement the terrain association features in the response trigger node information to improve the evolution environment description of the response trigger nodes. Extract terrain feature parameters corresponding to energy accumulation nodes from the node terrain adaptation dataset, analyze the correlation between terrain parameters and energy accumulation rate and total amount at energy accumulation nodes, and generate terrain impact data of accumulation nodes; Based on the terrain impact data of accumulated nodes, the terrain association features in the energy accumulation node information are supplemented to improve the description of the evolutionary environment of energy accumulation nodes. The terrain feature parameters corresponding to the diffusion initiation node are extracted from the node terrain adaptation dataset. The correlation between the terrain parameters and the diffusion direction and diffusion rate at the diffusion initiation node is analyzed to generate diffusion node terrain impact data. Based on the topographic impact data of diffusion nodes, the topographic association features in the information of diffusion initiation nodes are supplemented to improve the evolutionary environment description of diffusion initiation nodes. By integrating and improving the response trigger node information, energy accumulation node information, and diffusion initiation node information, and combining the response transmission relationship between each risk evolution node with topographic impact data, a spatial process correlation framework for the dynamic risk evolution of glacial debris flows is constructed. The energy state data, response characteristic data, and terrain characteristic data of each risk evolution node are supplemented in the spatial process association framework to generate a dynamic risk evolution path for glacial debris flow. This dynamic risk evolution path for glacial debris flow is used to present the development process and key characteristics of glacial debris flow risk at different spatial locations and different evolution stages.

5. The method of claim 1, wherein the method further comprises: The generation of dynamic risk early warning content for glacier debris flows based on the dynamic risk evolution path of glacier debris flows includes: The dynamic risk evolution path of glacial debris flow is analyzed, and the energy state data, response characteristic data, and topographic impact data of each risk evolution node are extracted. The risk development stage corresponding to each risk evolution node is determined, and risk stage division data is generated. The risk stage division data divides the risk evolution process of glacial debris flow into the initial triggering stage, the energy accumulation stage, and the diffusion and development stage. Key characteristic parameters for each risk development stage are extracted from the risk stage segmentation data to generate key characteristic data for each risk development stage. The key characteristic parameters include total energy accumulation, response diffusion range, and evolution duration. Human geography data within the glacier region is collected to generate a human geography dataset for the glacier region. The human geography data includes population distribution, infrastructure location, and the distribution range of important sites. Spatial information from the dynamic risk evolution path of glacial debris flows is spatially overlaid with human geography datasets of glacial regions to identify the human geography regions that may be affected at each risk development stage, thereby generating risk impact area data. Extract population distribution and infrastructure location within the affected areas at each stage from the risk impact area data, present the key areas and targets that need to be focused on at each stage, and generate data on key targets of focus; For each risk development stage, the response diffusion range in the key characteristic data of the risk development stage is spatially overlaid with the target distribution range in the key target data of the risk development stage. Combined with the total energy accumulation and evolution duration of the risk development stage, the potential impact on the key target is assessed, and stage impact analysis data is generated. Based on the stage impact analysis data, corresponding response measures are formulated for each stage of risk development, and stage response measure data is generated. The response measures cover the path planning and timing selection related to personnel transfer, and the formulation and implementation process of infrastructure protection measures. The rate data of risk evolution is extracted from the dynamic risk evolution path of glacial debris flow. Combined with the key characteristic data of each risk development stage, the duration of each risk development stage and the development trend of subsequent stages are predicted to generate risk trend prediction data. By integrating data on risk stage classification, key characteristic data of each risk development stage, data on risk-affected areas, data on key targets of concern, data on stage impact analysis, data on stage response measures, and data on risk trend prediction, a framework for dynamic risk early warning content is constructed. Establish the correspondence between the phased response measures data and the phased impact analysis data for each affected area, and generate correlation mapping data; integrate the risk phase division data, key characteristic data of each risk development phase, risk impact area data, key focus target data, phased impact analysis data, phased response measures data, risk trend prediction data, and correlation mapping data to generate dynamic risk early warning content for glacial debris flows.

6. The method of claim 2, wherein the method further comprises: The process involves comparing the response occurrence time points in the seismic response characteristic sets of each sphere with the signal arrival time in the seismic signal propagation path data of each sphere, dividing the response delay period of each sphere to the seismic signal, and generating sphere response delay data, including: The response occurrence time points corresponding to each sphere are extracted from the seismic response characteristic set of each sphere, and the response occurrence time points are classified and organized according to the sphere identifier to generate a sphere response time association dataset. The time points of earthquake signal arrival at each layer are extracted from the earthquake signal propagation path data, and the signal arrival time points are classified and organized according to the layer identifier to generate a layer signal arrival time association dataset. Match the concentric response time association dataset with the concentric signal arrival time association dataset according to the concentric identifier, establish the correspondence between the response occurrence time point and the signal arrival time point of each concentric layer, and generate the concentric time corresponding dataset; For the time-corresponding data of each concentric circle, the response occurrence time point and the signal arrival time point are extracted, and the time interval between the two time points is calculated. This time interval is used as the response delay duration of the concentric circle to the seismic signal. Record the response latency of each layer, associate it with the corresponding layer identifier, and generate preliminary layer response latency data; Structural feature data of each sphere is extracted from the glacier sphere structure dataset; the correlation between the material composition and structural integrity parameters of different spheres and the response delay time is analyzed to generate structural impact analysis data; Based on the structural impact analysis data, for the concentric circles in the preliminary concentric response delay data where the correlation between structural features and delay duration does not conform to the preset rules, the accuracy of the response occurrence time and signal arrival time is re-verified, and the calculation results of the response delay duration are corrected; the corrected response delay duration, concentric circle identifiers and structural feature data of all concentric circles are integrated to generate concentric response delay data. The response delay data of the spheres is structured and organized, and the response delay information of each sphere is arranged according to the distribution order of the spheres in the glacier region to generate the final sphere response delay data.

7. The method according to claim 2, wherein, The method, based on layer response delay data and seismic signal propagation path data, determines the response order of each glacier layer to seismic signals, integrates the transmission relationships between the response characteristics of each layer, and generates layer response transmission relationship data, including: The response delay duration of each layer is extracted from the layer response delay data. Combined with the time point when the seismic signal arrives at each layer in the seismic signal propagation path data, the absolute time when each layer actually generates a response is calculated to generate layer response absolute time data. The spheres are sorted according to the absolute time sequence of the absolute time data of the sphere response to determine the response order of each glacier sphere to the seismic signal, and the sphere response order data is generated. The seismic response feature set of each sphere is extracted from the seismic signal glacier response chain information, and the response features of each sphere are organized in the order of the sphere response sequence data to generate an ordered sphere response feature set. Analyze the response features of two adjacent concentric circles in an ordered concentric circle response feature set, extract the change patterns and generation ranges in the response features of the previous concentric circle, and identify the similarity between the response features and the response features of the next concentric circle in terms of change patterns and generation ranges. Based on the similarity analysis results, determine whether the response characteristics of the previous layer trigger the response changes of the next layer. If there is similarity and there is a sequential connection in time, it is determined that there is a response transmission relationship between the two. Record adjacent concentric circle pairs that have response transit relationships, present the forward transit concentric circle and the backward response concentric circle, and generate concentric circle transit pair data; Extract the response feature parameters of the forward transmission layer and the response feature parameters of the backward response layer in the transmission layer data, analyze the correlation between the change amplitude of the forward response feature and the change amplitude of the backward response feature, and generate transmission influence feature data. By combining the data on the transmission of layers and the characteristic data of the transmission impact, the response transmission paths between each layer are integrated to present the layer order and response characteristic transmission details on each transmission path, and generate layer transmission path data. By integrating data on the transmission of layers, data on the characteristics of transmission impact, and data on the transmission paths of layers, preliminary data on the transmission relationships of layers are generated. The time connection information of each transmission path in the preliminary data on the transmission relationships of layers is supplemented to present the time correlation between the end of the forward layer response and the start of the backward layer response, thereby improving the time dimension description of the transmission relationship and generating data on the transmission relationships of layers.

8. The method according to claim 3, wherein, The analysis of response transmission tracking data identifies the locations of concentric layers where the magnitude of change in response characteristics increases or the duration of change lengthens during transmission, generating energy accumulation region identification data, including: The response feature change data of each layer is extracted from the response transmission tracking data, and the response feature change data is organized according to the time sequence of response transmission to generate time-series response transmission data. The time-series response transmission data is used to reflect the dynamic changes of response features during the transmission process. Extract the response characteristic change amplitude parameter of each layer from the time-series response transmission data, associate it with the transmission time according to the layer identifier, and generate a layer amplitude time-related dataset; Obtain the baseline range threshold for the variation amplitude of response characteristics of each sphere in the glacier region, obtained based on historical data statistics; The change amplitude parameters in the time-related dataset of the concentric circles are compared with the benchmark range, and the concentric circles whose change amplitude parameters exceed the upper limit of the benchmark range are selected. These concentric circles are potential regions where the change amplitude of the response features increases. Extract the response feature duration parameter of each layer from the time-series response transmission data, associate it with the transmission time according to the layer identifier, and generate a layer-time long-term association dataset; Obtain a baseline threshold for the duration of response characteristics of each glacier layer based on historical data statistics; compare the duration parameters in the time-long-term correlation dataset of the glaciers with the baseline threshold, and filter out the glaciers whose duration parameters exceed the baseline threshold; Merge concentric circles whose change magnitude exceeds the upper limit of the baseline range or whose duration exceeds the baseline threshold to generate a set of candidate regions for energy accumulation; Collect the concentric structure data corresponding to each region in the candidate region set. Based on the stability and permeability parameters of the concentric structure, select regions with high structural instability and low permeability from the candidate region set to generate energy accumulation region identification data.

9. The method according to claim 4, wherein, The process involves extracting terrain feature parameters corresponding to the diffusion initiation nodes from the node terrain adaptation dataset, analyzing the correlation between the terrain parameters and the diffusion direction and diffusion rate at the diffusion initiation nodes, and generating diffusion node terrain impact data, including: Extract the terrain feature parameters corresponding to the diffusion initiation node from the node terrain adaptation dataset to generate the diffusion node terrain parameter set; Analyze the surface slope parameters in the topographic parameter set of diffusion nodes, determine the slope change trend, and generate slope trend analysis data. The slope change trend directly affects the movement speed of the diffused material. By combining slope trend analysis data, the impact of surface slope on diffusion rate is determined, and slope rate impact data is generated. Analyze the valley distribution parameters in the topographic parameter set of the diffusion nodes, extract the orientation parameters and connectivity parameters of the main valley, and generate valley guidance feature data; Based on the valley guidance feature data, the valleys that play a major guiding role in the diffusion direction are identified, the direction of the valleys is determined, and the main diffusion direction guidance data is generated. Analyze the surface undulation morphology parameters in the topographic parameter set of diffusion nodes, identify the raised and low-lying areas around the diffusion initiation node, and generate undulation influence characteristic data. By combining the undulation impact characteristic data, the diffusion direction deflection angle and diffusion rate adjustment coefficient caused by terrain undulation are calculated to generate undulation diffusion impact data; Analyze the slope aspect distribution parameters in the topographic parameter set of diffusion nodes, determine the angle relationship between the slope aspect and the diffusion initiation direction, and generate slope aspect direction influence data; By integrating slope rate influence data, main diffusion direction guidance data, undulation diffusion influence data, and slope aspect direction influence data, topographic influence data of diffusion nodes is generated.

10. A dynamic risk warning system for glacial debris flow based on seismic signals, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the dynamic risk warning method for glacial debris flows based on seismic signals according to any one of claims 1 to 9 by executing the machine-executable instructions.

Images