Snow rain disaster identification method based on cold and cold mountainous area and related product

Abstract

The invention relates to the technical field of snow melting disaster early warning, in particular to a snow surface rain disaster identification method based on a high and cold mountainous area and a related product, and the method comprises the steps: constructing a snow surface rain disaster single-point identification rule capable of being physically explained by using meteorological observation data and remote sensing inversion data of a small amount of weather stations in the high and cold mountainous area; the method is advantaged in that classification is carried out without depending on a black box model, precision and interpretability of single-point snow surface rain event discrimination are improved, regional popularization is carried out through dynamic downscaling simulation of a regional climate mode WRF coupling land surface process model Noah-MP, a point-surface fusion snow surface rain disaster discrimination and identification system is constructed, and the method is advantaged in that the method is simple and convenient to operate. The method is suitable for identifying snow rain disasters in complex terrain areas with high and cold areas, high altitude, large gradient, lack of data, even no data and the like, and solves the technical problems of insufficient snow rain disaster space identification capability and low precision caused by sparse observation data in the high and cold mountainous areas in the prior art.

Description

Technical Field

[0001] This invention relates to the field of snowmelt disaster early warning technology, and in particular to a method and related products for identifying snow surface rain disasters in high-altitude and cold mountainous areas. Background Technology

[0002] The phenomenon of rainfall falling on snow cover is called snow surface rain. Snow surface rain has the superimposed effect of rainfall and snowmelt, and the resulting rapid snowmelt process can easily trigger large-scale floods, avalanches, and landslides, posing a huge threat to regional water resources, ecological environment, and socio-economic development, and introducing great instability to regional socio-economic construction and sustainable ecological development. Studying the spatiotemporal changes in snowmelt disasters driven by climate change-induced rainfall has theoretical and practical significance for regional snowmelt flood prediction and water resource management. The snow surface rain disaster process is complex. Rainfall increases the water content of the snow layer, reduces albedo, increases snow layer energy, and enhances metamorphism; it leads to coarser snow grain size, changes in internal structure, intensifies snow subsidence and compaction, increases thermal conductivity, and accelerates snowmelt. At the same time, rainwater infiltrates into the snow layer or freezes on the bare surface, thus affecting vegetation growth, animal survival, and surface heat exchange. The snowmelt rate is closely related to the snow volume and is extremely sensitive to changes in temperature and precipitation pattern (rain or snow).

[0003] Snow surface rain is a complex process influenced by multiple hydrometeorological variables, including precipitation patterns, temperature, and snow cover. Threshold conditions based on precipitation and snow cover status are the primary method for determining snow surface rain occurrence. Current technologies for identifying snow surface rain events with disaster potential and for using threshold conditions for precipitation and snow cover mainly employ the critical temperature threshold method to estimate precipitation type, separate rain and snow, and identify whether a snow surface rain event has occurred.

[0004] However, due to the high spatiotemporal heterogeneity of the critical temperature for rain-snow separation and precipitation processes in high-altitude regions, as well as the sparse and incomplete ground observation networks in these areas, the identification of snow surface rain-induced snowmelt hazards becomes more difficult, thus reducing the accuracy of such identification. Therefore, improving the accuracy of snow surface rain hazard identification in high-altitude and cold mountainous areas has become an urgent technical problem to be solved. Summary of the Invention

[0005] This application aims to provide a method and related products for identifying snow rain disasters in high-altitude and cold mountainous areas. By combining meteorological and snow cover observation data analysis with regional climate model dynamic simulation, a "point-area fusion" snow rain disaster discrimination and identification system is constructed, which solves the technical problem of insufficient spatial identification capability and low accuracy of snow rain disasters due to the sparse observation data in high-altitude and cold mountainous areas in the existing technology.

[0006] Specifically, this is achieved through the following technical solutions:

[0007] Firstly, this application provides a method for identifying snow surface rain disasters in high-altitude cold mountainous areas, comprising: acquiring historical meteorological observation data and remote sensing inversion data of snow surface rain disasters occurring in high-altitude cold mountainous areas; extracting multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow surface rain disasters from the historical meteorological observation data and remote sensing inversion data; determining the discrimination thresholds and weights of the multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables; selecting several multidimensional meteorological characteristic target variables and snow cover hydrological characteristic target variables based on the weights, and combining the weights and discrimination thresholds corresponding to the multidimensional meteorological characteristic target variables and snow cover hydrological characteristic target variables through logical relationship combination to construct a single-point identification rule for snow surface rain disasters; then, constructing a rule reflecting complex terrain based on the regional climate model WRF coupled with the land surface process model Noah-MP. A dynamic downscaling simulation model of the interaction between meteorology and snow cover is proposed. ERA5 reanalysis data is used as the initial and boundary fields of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model optimizes the simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, outputting hourly gridded variables of the target mountain area within a preset time period. The hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables. Finally, the snow surface rain disaster single-point identification rules are used to identify and judge snow surface rain at each grid point according to time sequence. Based on the snow surface rain identification and judgment results, a spatial risk map of snow surface rain disasters occurring in the target mountain area within the preset time period is output, whereby the spatial risk map is used to characterize the spatial distribution of the risk of snow surface rain disasters occurring in the target mountain area within the preset time period.

[0008] Secondly, this application provides a snow rain disaster identification system based on high-altitude cold mountainous areas, including: a single-point snow rain disaster identification module, a regional reconstruction module, and a snow rain disaster risk identification module; wherein, the single-point snow rain disaster identification module acquires historical meteorological observation data and remote sensing inversion data of snow rain disasters occurring in high-altitude cold mountainous areas; extracts multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow rain disasters from the historical meteorological observation data and remote sensing inversion data; determines the discrimination thresholds and weights of the multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables; selects several multidimensional meteorological characteristic target variables and snow cover hydrological characteristic target variables based on the weights, and constructs single-point snow rain disaster identification rules by combining the weights and discrimination thresholds corresponding to the multidimensional meteorological characteristic target variables and snow cover hydrological characteristic target variables through logical relationship combination; the regional reconstruction module couples the regional climate model WRF with the land surface process model The Noah-MP model is used to construct a dynamic downscaling simulation model reflecting the interaction between meteorology and snow cover under complex terrain. ERA5 reanalysis data is used as the initial and boundary fields of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model optimizes the simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, outputting hourly gridded variables of the target mountain area within a preset time period. These hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables. Furthermore, the snow surface rain disaster risk identification module uses the single-point identification rules for snow surface rain disasters to identify and judge each grid point according to time sequence. Based on the snow surface rain identification and judgment results, a spatial risk map of snow surface rain disasters occurring in the target mountain area within the preset time period is output, whereby the spatial risk map characterizes the spatial distribution of the risk of snow surface rain disasters occurring in the target mountain area within the preset time period.

[0009] Thirdly, this application provides an electronic device comprising: at least one processor; a memory communicatively connected to the at least one processor; and a computer program stored in the memory and running on the at least one processor; wherein, when the at least one processor executes the computer program, it is used to implement the snow and rain disaster identification method based on high-altitude cold mountain areas described in the first aspect above.

[0010] Fourthly, embodiments of this application provide a computer-readable storage medium, wherein a computer program present in the computer-readable storage medium, when executed by a computer device, is used to implement the snow and rain disaster identification method based on high-altitude cold mountain areas described in the first aspect.

[0011] Fifthly, embodiments of this application provide a computer program product, which includes a computer program that, when executed by a computer device, is used to implement the snow and rain disaster identification method based on high-altitude cold mountain areas described in the first aspect.

[0012] As can be seen from the above, this application utilizes meteorological observation data and remote sensing inversion data from a limited number of meteorological stations in high-altitude and cold mountainous areas to construct physically interpretable single-point identification rules for snow rain disasters. This eliminates the need for black-box models for classification, improving the accuracy and interpretability of single-point snow rain event discrimination. Furthermore, it extends the system to the region through dynamic downscaling simulations using the regional climate model WRF coupled with the land surface process model Noah-MP, constructing a "point-area fusion" snow rain disaster discrimination and identification system. This system is applicable to snow rain disaster identification in complex terrain areas such as high-altitude, cold, steep, and data-scarce regions, solving the technical problem of insufficient spatial identification capability and low accuracy of snow rain disasters in existing technologies due to sparse observation data in high-altitude and cold mountainous areas. Simultaneously, based on the snow rain identification and judgment results, a spatial risk map of snow rain disasters occurring in the target mountainous area within a preset time period is output. This spatial risk map characterizes the spatial distribution of the risk of snow rain disasters occurring in the target mountainous area within the preset time period, demonstrating high engineering practicality. The spatial risk map can be integrated into geological disaster early warning systems or meteorological disaster early warning systems.

[0013] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart illustrating a snow surface rain disaster identification method based on high-altitude cold mountain areas, provided for an embodiment of this application;

[0016] Figure 2 A schematic diagram of a snow and rain disaster identification system based on high-altitude cold mountain areas is provided for an embodiment of this application;

[0017] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Example 1

[0020] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for identifying snow surface rain hazards in high-altitude and cold mountainous areas, provided as an embodiment of this application. Figure 1 As shown, this method for identifying snow surface rain hazards in high-altitude and cold mountainous areas includes the following:

[0021] 101. Obtain historical meteorological observation data and remote sensing inversion data of snow surface rain disasters in high-altitude and cold mountainous areas.

[0022] To address the challenge of sparse observations in high-altitude and cold mountainous areas, this application prioritizes utilizing historical meteorological data from a limited number of stations over a specific period, such as temperature and precipitation, as well as historical remote sensing inversion products, such as snow cover and snow water equivalent, to construct a sample library of snow rain events. This provides a foundation for subsequently extracting multidimensional meteorological and snow-related hydrological variables associated with snow rain disasters. Compared to traditional snow cover identification methods based solely on remote sensing, this application not only considers whether rain and snow overlap occur but also focuses on the intensity of snow rain, i.e., the combined effect of rainfall and snowmelt. This dynamic variable combination is difficult to achieve with purely remote sensing-based snow cover identification methods, thus providing a basis for improving the accuracy of snow rain disaster identification in this application.

[0023] 102. Extract multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow surface rain disasters from the historical meteorological observation data and remote sensing inversion data.

[0024] Based on historical snow surface rain events occurring in high-altitude and cold mountainous areas, and combining historical meteorological observation data and remote sensing inversion data, multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow surface rain disasters are extracted. In high-altitude and cold mountainous areas, due to the low level of automatic weather station construction and discontinuous data, conventional meteorological stations find it difficult to directly record this complex process of snow surface rain events. Therefore, the implementation method for obtaining historical snow surface rain events in this application embodiment can be as follows:

[0025] On the one hand, through manual investigation and disaster yearbooks, we collect disaster bulletins, yearbooks and historical archives issued by local water conservancy, emergency management, meteorology and other departments, and screen flood events described as "floods caused by melting snow after rain", "surges the next day after rain and snow", "spring melting rain disaster" and confirm whether their time, space and cause are consistent with the characteristics of "snow surface rain".

[0026] On the other hand, by using meteorological reconstruction and physical criteria, daily temperature, precipitation, snow depth (i.e., snow water equivalent), and snow cover are retrieved; then it is determined whether there is snow on the ground and the temperature is rising, and whether there is a situation where precipitation occurs simultaneously (i.e., temperature > 0℃, precipitation is rain, ground snow depth > 0cm, SCF > 25%), where SCF is snow cover.

[0027] By integrating the above two aspects and manually verifying the physical consistency of time, location, and process, multiple representative historical snow rain events can be obtained as sample events for snow rain disasters, which can be used for model building and variable selection.

[0028] As mentioned above, based on historical snow surface rain events occurring in high-altitude and cold mountainous areas, and combining historical meteorological observation data and remote sensing inversion data, the specific implementation method for extracting multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow surface rain disasters can be as follows:

[0029] Based on the principle of triggering mechanisms, multidimensional meteorological and snow cover hydrological variables related to snow rain disasters are extracted from historical snow rain events. For example, the triggering condition for snow rain requires the simultaneous fulfillment of "rainfall, snow cover, and snowmelt response." Therefore, the multidimensional meteorological and snow cover hydrological variables related to snow rain disasters need to cover multiple variables such as rainfall, temperature, snow cover, snowmelt process, and precipitation phase. Alternatively, starting from observability, meteorological and snow cover variables can be obtained by combining meteorological stations and remote sensing observations to construct a sample sequence of snow rain events. Alternatively, statistical and machine learning methods can be used to extract variables with strong discriminative power, such as information gain analysis and random forest variable importance ranking. Commonly used multidimensional meteorological and snow cover hydrological variables are shown in Table 1 below.

[0030] Variable name illustrate Daily Precipitation (PREC) Daily precipitation Daily rainfall Pure rainfall observed or simulated on the day Snowmelt Snow melt amount calculated by inversion of snow water equivalent sequence Snowmelt Ratio The proportion of snowmelt caused by rainfall to the total amount of rainfall plus snowmelt is used to determine the intensity of the disaster. Average daily temperature (T2) Determine if it is a heating process Surface air pressure (PSFC) Used to calculate wet-bulb temperature and determine precipitation phase (rain / snow / mixed). Relative humidity (RH) Used to calculate wet-bulb temperature and determine precipitation phase (rain / snow / mixed). Snow water equivalent (SWE) Characterizing the equivalent water volume of snow cover Snow cover (SCF) Characterizing regional snow cover using remote sensing products Snow depth (SD) Used to estimate snow cover changes

[0031] Table 1

[0032] The following example illustrates the snow surface rain flood event in Yili, Xinjiang in mid-May 2025: On the day of the event, the remote sensing SCF was 72%, which dropped to 31% the next day. The automatic weather station recorded a daily rainfall of 14.5 mm and an average daily temperature of 5.2℃. The SWE decreased by 11 mm, and the estimated snowmelt was 9 mm. All of these variables were extracted into the event sample and sorted in machine learning, resulting in the following example of a discrimination rule: If the rainfall is >10 mm, the snow accumulation is >10 mm, the snowmelt is >6 mm, and the snow cover is >30%, then it is determined to be a snow surface rain disaster event.

[0033] 103. Determine the discrimination thresholds and weights for the multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables.

[0034] In this embodiment, quantile statistics are used to statistically analyze the proportion of each characteristic variable in the total water volume of typical snow rain disaster events occurring in high-altitude cold mountainous areas. For example, the proportions of rainfall, snowmelt, snow water equivalent (SWE), snow cover (SCF), and snowmelt in the total water volume are analyzed. The disaster threshold value corresponding to the 75th or 80th percentile is extracted as the discrimination threshold for the corresponding characteristic variable. For example, if "daily rainfall > 10 mm", "snow water equivalent > 10 mm", "snow cover > 25%", and "snowmelt percentage > 20%", then it is determined to be a snow rain disaster event, where 10 mm is the discrimination threshold for daily rainfall.

[0035] In this embodiment, a supervised learning algorithm is employed. Using the multidimensional meteorological and snow cover hydrological variables as input and snow surface rainfall intensity as output, the weights of these variables are determined. These weights characterize the contribution of each variable to the snow surface rainfall intensity during the snow surface rainfall period. Specifically, an algorithm such as a random forest is used. Variables such as rainfall, snowmelt, snow water equivalent (SWE), snow cover (SCF), and surface temperature from historical snow surface rainfall sample events are used as input, and snow surface rainfall intensity is used as output. Machine learning is employed to rank the contributions of each variable from largest to smallest (i.e., their importance to snow surface rainfall hazards), thus determining the weights of the multidimensional meteorological and snow cover hydrological variables.

[0036] 104. Select several multidimensional meteorological characteristic target variables and snow cover hydrological characteristic target variables based on weights.

[0037] As mentioned above, variables with significant contributions are selected based on the ranking results. This can be achieved by selecting the top N variables from the ranking, which will serve as the target variables for the multidimensional meteorological and snow hydrological characteristics. For example, if N = 4, then precipitation, snowmelt, snow water equivalent (SWE), and snow cover (SCF) can be selected from the ranking results as the target variables for the multidimensional meteorological and snow hydrological characteristics.

[0038] 105. Combining the weights and discrimination thresholds corresponding to the multidimensional meteorological feature target variables and the snow cover hydrological feature target variables, a single-point identification rule for snow surface rain disasters is constructed through logical relationship combination.

[0039] As described above, the weights and discrimination thresholds corresponding to the multidimensional meteorological and snow hydrological target variables are combined, and a Boolean judgment formula that is structurally clear, physically interpretable, and applicable to single-point scale identification is constructed through logical relationships. This formula serves as the single-point identification rule for snow surface rain disasters in this application embodiment. For example, if the following conditions are met: rainfall > 10 mm; snow cover > 25%; snow water equivalent > 10 mm; and the proportion of snowmelt in the total rainfall and snowmelt > 20%, i.e., snowmelt ratio > 0.2, then the period is determined to be a snow surface rain disaster event. This application's embodiments extract meteorological and hydrological characteristic variables related to snow rain disasters based on historical snow rain events and sparse meteorological observation data. Machine learning methods are then used to determine combined thresholds for several key meteorological and hydrological characteristic variables, forming a snow rain disaster discrimination rule suitable for single-point identification. Furthermore, multi-variable composite logic rules driven by snow rain intensity are used to ensure consistency with disaster impact quantification indicators. Machine learning is employed for variable screening and weight ranking to improve snow rain discrimination accuracy. This avoids reliance on black-box models, possesses good interpretability and regional adaptability, and solves the technical problem of insufficient spatial identification capability and low accuracy of snow rain disasters due to sparse observation data in high-altitude and cold mountainous areas in existing technologies.

[0040] 106. Based on the regional climate model WRF coupled with the land surface process model Noah-MP, a dynamic downscaling simulation model reflecting the interaction between meteorology and snow cover under complex terrain is constructed. ERA5 reanalysis data is used as the initial field and boundary field of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model performs optimized simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, and outputs hourly gridded variables of the target mountain area within a preset time period. The hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables.

[0041] In this embodiment of the application, in order to address the problems of sparse observation data, insufficient spatial coverage of stations, and limited timeliness of remote sensing data in high-altitude and cold mountainous areas, a regional climate model (Weather Research and Forecasting, WRF) coupled with a Noah Land Surface Model with Multiparameterization Options (Noah-MP) is adopted to construct a dynamic downscaling simulation model that can reflect the interaction between meteorology and snow cover under complex terrain. This enables regional reconstruction of key variables, can be extended to regional-scale simulation results, and achieves grid-level spatial identification. The specific implementation method includes: using coarse-resolution hourly interval ERA5 (the fifth generation ECMWF atmospheric reanalysis of the global climate, produced by the Copernicus Climate Change Service) reanalysis data as the initial and boundary fields of the dynamic downscaling simulation model; including barosphere variables (e.g., temperature, U and V components of wind, geopotential height, and relative humidity) and surface variables (e.g., surface pressure, sea level pressure, sea surface temperature, air temperature at 2 meters, specific humidity at 2 meters, U and V components at 10 meters altitude, wind speed, soil temperature, soil moisture and altitude, snow density, and snow depth). A multi-nested grid structure, such as 27km→9km→3km, is set to ensure that the regional simulation reaches the allowable convection scale (≤4km). By using the WRF model to conduct convection permissible scale simulation (≤4km), a gridded meteorological characteristic variable and snow cover characteristic variable data field considering the influence of complex terrain is generated, realizing the spatial mapping and identification from single-point rules to regional scale. Compared with traditional disaster identification methods based solely on remote sensing or statistical models, the embodiments of this application have stronger physical consistency, terrain adaptability and spatial generalization capabilities, and are particularly suitable for the spatiotemporal identification and risk zoning of snow surface rain disasters in remote, high-altitude and complex mountainous areas with insufficient data. Specifically, the above model outputs hourly (not limited to hourly, but also M-hourly, where M is a positive integer and can be preset) gridded variables for the target mountainous area within a preset time period, including meteorological characteristic variables such as near-surface air temperature (T2), near-surface air pressure (PSFC), near-surface specific humidity (Q2), 10-meter UV wind, downward shortwave radiation flux (SWDOWN), downward longwave radiation flux (GLW), and surface precipitation (PREC); and snow characteristic variables such as rainfall, snow water equivalent (SWE), snowmelt, and snow cover (SCF) obtained by enabling the Noah-MP snow multilayer structure and driving the Noah-MP land surface process model to calculate the snow process.This application's embodiments utilize a dynamic downscaling simulation model to consider the strong regulatory effects of slope aspect, elevation, and surface type on snow formation and melting processes in the target mountainous area. Even in the absence of observational data, it can generate a physically consistent and spatially continuous field of triggering factors for snow surface rain disasters. Therefore, the dynamic downscaling simulation model of this application's embodiments possesses adaptability to complex terrain and can accurately reflect the spatial differences in precipitation, snow accumulation, and snowmelt processes in high-altitude areas, thereby providing continuous and physically consistent gridded input data for regional identification of snow surface rain disasters.

[0042] The Noah-MP model, which drives land surface processes as described above, is used to calculate snow accumulation processes, yielding snow accumulation characteristic variables such as rainfall (rainll), snow water equivalent (SWE), snowmelt (snowmelt), and snow cover (SCF). A specific implementation method is as follows:

[0043] (1) Rainfall calculation:

[0044] The Noah-MP land surface process model uses the following formula to perform rain and snow segmentation calculations, obtaining the conditional probability F of snowfall. snowfall :

[0045]

[0046] in,

[0047] 'a' is a scale factor for adjusting the slope of precipitation phase transition; in this embodiment, it is taken as 6.99 × 10⁻⁶. -5 ;

[0048] b is the temperature sensitivity coefficient, which is 2.0 (unit: 1 / K) in this embodiment.

[0049] c is the temperature offset correction term, which is 3.97 (in K) in this embodiment of the application;

[0050] T wetb This refers to the wet-bulb temperature.

[0051] wet-bulb temperature T wetb Solve iteratively using the following formula:

[0052]

[0053] in,

[0054] e sat (T wetb The saturated vapor pressure is calculated using an empirical formula.

[0055] T2 is the near-surface temperature, expressed in Kelvin (K).

[0056] e air The actual water vapor pressure is calculated from the specific humidity and air pressure.

[0057] γ is a psychological constant, and in this embodiment, it is approximately 66 (in Pa / K).

[0058] The equation is solved iteratively until e is reached. sat (T wetb ) equals e air This process employs an iterative method to determine the wet-bulb temperature under energy balance conditions, serving as the core variable for distinguishing between rainfall and snowfall.

[0059] The above saturated water vapor pressure e sat (T wetb )have:

[0060]

[0061] The above-mentioned actual water vapor pressure e air have:

[0062]

[0063] in,

[0064] Q2 is the near-surface specific humidity;

[0065] PSFC stands for near-surface air pressure, measured in Pa.

[0066] Therefore, the rainfall and snowfall per unit time calculated by the Noah-MP land surface process model are as follows:

[0067] rainfall = PREC × (1-F) snowfall );

[0068] snowfall = PREC × F snowfall ;

[0069] PREC represents the surface precipitation (including rain and snow) output by the regional climate model (WRF) per unit time, expressed in mm / hr or mm / step, indicating the total precipitation within the time step.

[0070] (2) Snow water equivalent calculation:

[0071] The Noah-MP land surface process model targets the rate of change of total snow water equivalent (i.e., the equivalent liquid water depth obtainable after all snow has melted) over time. The calculation is as follows:

[0072]

[0073] in,

[0074] ΔWsnow This represents the change in snow water equivalent (SWE).

[0075] Δt is the time step;

[0076] Q snow (Unit: mm / s) represents the rate at which snowfall falls from the forest canopy to the snow surface;

[0077] Q sno,frost (Unit: mm / s) represents the total amount of water vapor sublimation that forms frost on the snow surface;

[0078] Q sno,melt (Unit: mm / s) represents the total amount of snow melt driven by the energy balance of the snow layer surface;

[0079] Q sno,subl (Unit: mm / s) represents the amount of water vapor lost from the direct sublimation of ice in the snow layer.

[0080] (3) Snowmelt calculation:

[0081] Total snowmelt (Q) sno,melt The calculation formula for ) is as follows:

[0082]

[0083] in,

[0084] C LH,fus The latent heat of fusion of ice (unit: J / kg);

[0085] SW grd,net LW grd,net H pr,grd H L,grd H s,grd and H snow2soil (W / m 2 These are the net shortwave and net longwave radiation fluxes over the snow surface, the downflow heat flux, latent heat, sensible heat flux, and the snow-soil interface heat flux, respectively.

[0086] The calculation process requires no external fitting or manual parameter setting, and has good physical constraints and adaptability.

[0087] (4) Snow cover calculation:

[0088] The Noah-MP land surface process model calculates snow cover (SCF) as follows:

[0089]

[0090] in,

[0091] h sno (Unit: meters) represents snow depth;

[0092] SC factor For adjustable semi-empirical snow cover factor parameters;

[0093] f melt The melting factor of SCF on the ground.

[0094] In this embodiment, the snow cover (SCF) is calculated based on the semi-empirical function structure in the Noah-MP land surface process model, using snow depth, scale parameters, and density adjustment factors. Relevant parameters such as snow depth, snow water equivalent, snow density, and grid resolution are all derived from model simulation results or constants already set in the model, requiring no external data or manual intervention, thus ensuring the consistency and automation of snow cover calculation.

[0095] Among them, SC factor The parameters are calculated as follows:

[0096] SC factor =β 1,fsno sinh(β 2,fsno DX)+β 2,fsno ;

[0097] in,

[0098] DX is the grid size configured in the regional climate model (WRF), i.e., the grid resolution (in km);

[0099] β 1,fsno and β 2,fsno These are empirical parameters built into the Noah-MP land surface process model, representing the response coefficient of the snow cover factor to spatial scale and the basic offset, respectively. In the embodiments of this application, these parameters are taken as 0.0062 and 0.0555, respectively.

[0100] Among them, the above f melt The parameters are calculated as follows:

[0101]

[0102] Where, ρ new The density of fresh snow is set to 100 kg / m³ in this embodiment of the application. 3 ;

[0103] m is an adjustable semi-empirical melting index, which is related to the spatial scale (the larger the scale, the larger the value of m is usually); We have:

[0104] m = β 1,m +tanh(β 2,m ·DX);

[0105] Where, β 1,m and β 2,mThese are empirical constants, with values ​​of 0.9713 and 0.7436, respectively.

[0106] ρ snow This refers to the overall snow density.

[0107] Where, ρ snow The calculation formula is:

[0108]

[0109] Where, m sno This indicates the mass of snow cover, specifically snow water equivalent (SWE), expressed in kg / m³. 2 .

[0110] 107. Using the aforementioned single-point identification rule for snow surface rain disasters, snow surface rain identification and judgment are performed on each grid point according to the time sequence.

[0111] Based on the output of the aforementioned dynamic downscaling simulation model, this embodiment applies the constructed single-point identification rules for snow rain disasters to the regional simulation grid (i.e., the hourly gridded meteorological and snow cover characteristic variable data field of the target mountain area within a preset time period output by the aforementioned dynamic downscaling simulation model). It then identifies each simulation grid of the target mountain area within the preset time period, thereby achieving spatial identification and risk zone extraction of snow rain disasters. Specifically, as described above, the meteorological and snow cover characteristic target variables used in the aforementioned single-point identification rules for snow rain disasters, such as rainfall, snowmelt, snow water equivalent, and snow cover, are matched with the hourly gridded meteorological and snow cover characteristic variables of the target mountain area within the preset time period output by the aforementioned dynamic downscaling simulation model. The specific implementation of this matching can be as follows: for example, if there are differences in definition or inconsistencies in units, normalization, physical conversion, or other methods can be used for unified processing.

[0112] Then, a grid-by-grid logical discrimination calculation is performed: for each simulated grid point, key variables corresponding to the time series are extracted; and the above-mentioned single-point identification rules for snow rain disasters are used to identify and judge snow rain. For example, if the rainfall on a certain day is >10mm, and the snow water equivalent on that day is >10mm, and the snow cover is >25%, and the snowmelt amount / (rainfall + snowmelt) is >20%, then it is determined that the location corresponding to the simulated grid point on that day constitutes a snow rain disaster event, and the spatial Boolean matrix outputs 1; otherwise, it outputs 0. Combining the identification results of multiple consecutive days, a process time sequence of snow rain disaster events is formed. At the same time, in a preferred embodiment, indicators such as the frequency, intensity, duration, and cumulative rainfall of disasters are further statistically analyzed to cross-validate with historical disaster records, which can effectively improve the recognition reliability.

[0113] In this embodiment, the previously constructed single-point identification rules for snow surface rain disasters are applied one by one to each grid cell within the simulated area. Logical discrimination is performed according to the time sequence to form a spatial Boolean matrix, thereby realizing the discrimination and spatial identification of the regional-scale disaster risk of snow surface rain disasters in the target mountainous area within a preset time period.

[0114] 108. Based on the snow surface rain identification and judgment results, output a spatial risk map of snow surface rain disasters in the target mountainous area within a preset time period.

[0115] The spatial risk map is used to characterize the spatial distribution of the risk of snow and rain disasters in the target mountainous area within a preset time period.

[0116] The specific implementation method is as follows: The above-mentioned judgment results (spatial Boolean matrix) are projected onto a geographic coordinate system to generate a spatial risk map of snow surface rain disasters occurring in the target mountainous area within a preset time period. This includes, for example, an event occurrence area map, an intensity level map, and a three-dimensional distribution map. This map can be used for disaster early warning and emergency decision support, assisting in disaster prevention and mitigation deployment. The spatial risk map can be integrated into geological disaster early warning systems or meteorological disaster early warning systems, demonstrating high engineering practicality. This application's embodiment can achieve dynamic identification of the location, impact range, and risk level of snow surface rain disasters in areas lacking data, possessing advantages such as high resolution, physical consistency, and regional adaptability.

[0117] Example 2

[0118] The above combination Figure 1 This application provides a detailed description of a snow and rain disaster identification method based on high-altitude cold mountain areas, as illustrated in its embodiments. The following is in conjunction with... Figure 2 This document describes in detail a snow and rain disaster identification system based on a high-altitude cold mountain area, which is used to implement the snow and rain disaster identification method provided in the embodiments of this application. Figure 2 A schematic diagram of a snow and rain disaster identification system based on high-altitude cold mountain areas is provided for embodiments of this application; please refer to... Figure 2 The system includes the following:

[0119] The snow surface rain disaster single-point identification module 201 acquires historical meteorological observation data and remote sensing inversion data of snow surface rain disasters occurring in the target high-altitude mountainous area; extracts multidimensional meteorological feature variables and snow cover hydrological feature variables related to snow surface rain disasters from the historical meteorological observation data and remote sensing inversion data; determines the discrimination threshold and weight of the multidimensional meteorological feature variables and snow cover hydrological feature variables; selects several multidimensional meteorological feature target variables and snow cover hydrological feature target variables based on the weights, and constructs single-point identification rules for snow surface rain disasters by combining the weights and discrimination thresholds corresponding to the multidimensional meteorological feature target variables and snow cover hydrological feature target variables through logical relationship combination.

[0120] The regional reconstruction module 202 constructs a dynamic downscaling simulation model reflecting the interaction between meteorology and snow cover under complex terrain, based on the regional climate model WRF coupled with the land surface process model Noah-MP. ERA5 reanalysis data is used as the initial and boundary fields of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model optimizes the simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, outputting hourly gridded variables for the target high-altitude mountainous area within a preset time period. The hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables.

[0121] The snow surface rain disaster risk identification module 203 uses the single-point identification rule of the snow surface rain disaster to identify and judge the snow surface rain for each grid point in a time sequence; based on the snow surface rain identification and judgment results, it outputs a spatial risk map of the snow surface rain disaster occurring in the target high-altitude cold mountain area within a preset time period, wherein the spatial risk map is used to characterize the spatial distribution of the risk of snow surface rain disaster occurring in the target high-altitude cold mountain area within a preset time period.

[0122] In a preferred embodiment, the snow surface rain disaster single-point identification module 201 is specifically used to: use quantile statistics to statistically analyze the proportion of each characteristic variable in the total water volume of a snow surface rain disaster event occurring in the target high-altitude mountainous area; and extract the disaster threshold value corresponding to the preset percentile as the discrimination threshold for each corresponding characteristic variable.

[0123] In a preferred embodiment, the single-point identification module 201 for snow surface rain disaster is specifically used to: employ a supervised learning algorithm, taking the multidimensional meteorological feature variables and snow cover hydrological feature variables as inputs and snow surface rain intensity as output, to determine the weights of the multidimensional meteorological feature variables and snow cover hydrological feature variables, wherein the weights are used to characterize the contribution of the multidimensional meteorological feature variables and snow cover hydrological feature variables to the snow surface rain intensity during the snow surface rain time.

[0124] The specific implementation method and technical effects of the snow surface rain disaster identification system based on high-altitude cold mountain areas can be referred to the aforementioned snow surface rain disaster identification method based on high-altitude cold mountain areas, and will not be repeated here.

[0125] Example 3

[0126] This invention also provides an electronic device. Figure 3 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 3As shown, the electronic device includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage section 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304. The following components are connected to the I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A driver 310 is also connected to the I / O interface 305 as needed. Removable media 311, such as disks, optical discs, magneto-optical discs, semiconductor memories, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0127] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0128] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention is not limited to the specific combination of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this invention.

Claims

1. A method for identifying snow surface rain hazards in high-altitude and cold mountainous areas, characterized in that, include: Acquire historical meteorological observation data and remote sensing inversion data of snow rain disasters in high-altitude and cold mountainous areas; extract multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow rain disasters from the historical meteorological observation data and remote sensing inversion data; Determine the discrimination thresholds and weights of the multidimensional meteorological feature variables and snow cover hydrological feature variables; select several multidimensional meteorological feature target variables and snow cover hydrological feature target variables based on the weights and discrimination thresholds corresponding to the multidimensional meteorological feature target variables and snow cover hydrological feature target variables, and construct a single-point identification rule for snow surface rain disaster by combining logical relationships. Based on the regional climate model WRF coupled with the land surface process model Noah-MP, a dynamic downscaling simulation model reflecting the interaction between meteorology and snow cover under complex terrain is constructed. ERA5 reanalysis data is used as the initial and boundary fields of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model performs optimized simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, outputting hourly gridded variables of the target mountain area within a preset time period. The hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables. The snow surface rain disaster single-point identification rule is used to identify and judge snow surface rain for each grid point according to time sequence; based on the snow surface rain identification and judgment results, a spatial risk map of snow surface rain disasters occurring in the target mountain area within a preset time period is output, wherein the spatial risk map is used to characterize the spatial distribution of the risk of snow surface rain disasters occurring in the target mountain area within the preset time period.

2. The method for identifying snow surface rain hazards in high-altitude cold mountainous areas as described in claim 1, characterized in that, Determining the discrimination thresholds for the multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables includes: The proportion of each characteristic variable in the total water volume during snow surface rain disasters in high-altitude and cold mountainous areas was statistically analyzed using quantile statistics. The disaster threshold value corresponding to the preset percentile was extracted as the discrimination threshold for each characteristic variable.

3. The method for identifying snow surface rain hazards in high-altitude cold mountainous areas as described in claim 2, characterized in that, Determining the weights of the multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables includes: A supervised learning algorithm is used, with the multidimensional meteorological feature variables and snow cover hydrological feature variables as inputs and snow surface rain intensity as output, to determine the weights of the multidimensional meteorological feature variables and snow cover hydrological feature variables. The weights are used to characterize the contribution of the multidimensional meteorological feature variables and snow cover hydrological feature variables to the snow surface rain intensity during the snow surface rain time.

4. The method for identifying snow surface rain hazards in high-altitude cold mountainous areas as described in claim 3, characterized in that, The hourly gridded variables include rainfall, snow water equivalent, snow cover, and snowmelt.

5. A snow surface rain disaster identification system based on high-altitude cold mountain areas, characterized in that, include: The snow surface rain disaster single-point identification module acquires historical meteorological observation data and remote sensing inversion data of snow surface rain disasters occurring in high-altitude and cold mountainous areas; and extracts multidimensional meteorological characteristic variables and snow cover hydrological characteristic variables related to snow surface rain disasters from the historical meteorological observation data and remote sensing inversion data. Determine the discrimination thresholds and weights of the multidimensional meteorological feature variables and snow cover hydrological feature variables; select several multidimensional meteorological feature target variables and snow cover hydrological feature target variables based on the weights and discrimination thresholds corresponding to the multidimensional meteorological feature target variables and snow cover hydrological feature target variables, and construct a single-point identification rule for snow surface rain disaster by combining logical relationships. The regional reconstruction module constructs a dynamic downscaling simulation model reflecting the interaction between meteorology and snow cover under complex terrain, based on the regional climate model WRF coupled with the land surface process model Noah-MP. ERA5 reanalysis data is used as the initial and boundary fields of the dynamic downscaling simulation model, and a multi-nested grid structure is set. The dynamic downscaling simulation model optimizes the simulation of snow surface rain disaster processes based on the ERA5 reanalysis data, outputting hourly gridded variables of the target mountain area within a preset time period. These hourly gridded variables correspond and match with the multidimensional meteorological characteristic target variables and the snow cover hydrological characteristic target variables. The snow rain disaster risk identification module uses the single-point identification rules for snow rain disasters to identify and judge snow rain at each grid point according to time sequence; based on the snow rain disaster identification and judgment results, it outputs a spatial risk map of snow rain disasters occurring in the target mountain area within a preset time period, wherein the spatial risk map is used to characterize the spatial distribution of the risk of snow rain disasters occurring in the target mountain area within the preset time period.

6. A snow and rain disaster identification system based on high-altitude cold mountain areas as described in claim 5, characterized in that, The single-point identification module for snow surface rain disasters is specifically used for: The proportion of each characteristic variable in the total water volume during snow surface rain disasters in high-altitude and cold mountainous areas was statistically analyzed using quantile statistics. The disaster threshold value corresponding to the preset percentile was extracted as the discrimination threshold for each characteristic variable.

7. A snow and rain disaster identification system based on high-altitude cold mountain areas as described in claim 6, characterized in that, The single-point identification module for snow surface rain disasters is specifically used for: A supervised learning algorithm is used, with the multidimensional meteorological feature variables and snow cover hydrological feature variables as inputs and snow surface rain intensity as output, to determine the weights of the multidimensional meteorological feature variables and snow cover hydrological feature variables. The weights are used to characterize the contribution of the multidimensional meteorological feature variables and snow cover hydrological feature variables to the snow surface rain intensity during the snow surface rain time.

8. An electronic device, characterized in that, The electronic device includes: at least one processor; a memory communicatively connected to the at least one processor; and a computer program stored in the memory and running on the at least one processor; wherein, when the at least one processor executes the computer program, it is used to implement a snow and rain disaster identification method based on high-altitude cold mountain areas as described in any one of claims 1 to 4.

9. A computer-readable storage medium, characterized in that, The computer program present in the computer-readable storage medium, when executed by a computer device, is used to implement the snow and rain disaster identification method based on high-altitude cold mountain areas as described in any one of claims 1 to 4.

10. A computer program product, the computer program product comprising a computer program, characterized in that, When executed by a computer device, the computer program is used to implement a snow and rain disaster identification method based on high-altitude cold mountainous areas as described in any one of claims 1 to 4.

Images