Southwest low vortex whole process objective identification method based on grid reanalysis data and medium

Abstract

The invention discloses a grid reanalysis data-based whole-process objective identification method for a southwest low vortex, and a medium, and relates to the technical field of meteorological big data processing and numerical weather analysis. The method comprises the following steps: eliminating high-altitude grid points in a target identification area by using a terrain mask; searching a closed isoline according to an equivalence distance based on a preset air pressure horizon potential height field, extracting a closed low-pressure candidate body and determining a low-pressure center; calculating the attribution of a closed region, screening an effective candidate body, constructing a time sequence track by adopting adjacent time space adjacent matching, and introducing displacement constraint and duration criterion to confirm a low eddy process; and outputting starting and ending moments, duration time and a moving path, and supporting frequency, path and rainfall influence statistics. According to the method, automatic identification, stable tracking and consistency statistics of the southwest low vortex in a long-time sequence can be realized, subjective identification deviation and labor cost are reduced, and rechecking performance and service availability of an identification result under a complex terrain condition are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of meteorological big data mining and numerical weather analysis, and relates to automatic identification of mesoscale weather systems, vortex tracking and climate statistical analysis, in particular to a southwest low vortex whole-process objective identification method based on grid reanalysis data and a medium. BACKGROUND

[0002] The southwest low vortex is a mesoscale closed low vortex system with cyclonic circulation characteristics formed in the eastern Qinghai-Tibet Plateau and Sichuan Basin. It usually appears on the 700-850 hPa isobaric surface, with a horizontal scale of about 300-500 kilometers, and is one of the key weather systems that cause regional rainstorm, thunderstorm and other disastrous weather in the second half of the year in China, with importance only next to typhoon. When the southwest low vortex develops and moves eastward, it often causes extreme precipitation events in the middle and lower reaches of the Yangtze River, the Huaihe River basin, and even the North China and Northeast China. Therefore, accurate monitoring and whole-process identification of the southwest low vortex have great scientific significance and business value for improving the accuracy of rainstorm forecast and carrying out disaster prevention and mitigation.

[0003] In terms of technology, accurate identification and continuous tracking of the southwest low vortex are the basic links for carrying out climate statistical analysis, process mechanism diagnosis and forecast evaluation. However, the existing identification methods mainly rely on manual identification, which has low efficiency, huge consumption of human resources and inevitable subjective identification bias when dealing with long time series and historical climate data. Especially in the southwest region of China, the terrain is extremely complex, and the distribution of meteorological observation stations is uneven, with a large number of observation blind areas in the plateau and mountainous areas. The incompleteness of such data further aggravates the uncertainty and inconsistency of manual identification, leading to significant differences in the basic climate statistical characteristics of southwest low vortex activities among different studies.

[0004] In addition, due to the differences in the update and iteration of historical observation equipment, it is difficult to accumulate long-time and continuous uniform southwest low vortex monitoring data. The lack of such long-time series high-quality data makes it difficult to systematically grasp the sub-seasonal, interannual to interdecadal scale evolution of the southwest low vortex and its response characteristics under the background of global climate change. Although some studies have tried to use 700 hPa potential height field or wind field for objective identification, these methods often only consider a single meteorological element or simple threshold, and fail to comprehensively reflect the physical nature of the southwest low vortex as a complete weather system. These simplified objective methods still perform unsatisfactorily in dealing with local disturbances under complex terrain, continuity of system generation and extinction, etc.

[0005] For example, Chinese patent CN120470303A discloses a method for identifying and tracing a southwest vortex. It uses a multi-parameter combined with a four-layer wind field algorithm for identification, but it is insufficient in characterizing the continuity of the system's generation and dissipation under complex terrain, making it difficult to achieve robust tracking of the entire process over long time series. CN112131958B discloses a method for automatically identifying a southwest low-pressure vortex, which uses color space projection to process the geopotential height field characteristics. However, it relies too much on a single height field element and lacks a comprehensive consideration of the vortex dynamic circulation characteristics, which can easily lead to failure in identifying weak systems and make it impossible to achieve dynamic closed-loop tracking of the path.

[0006] In summary, existing technologies still have significant shortcomings in objectively identifying, continuously tracking, and statistically consistently tracking the Southwest Low Vortex under complex terrain and incomplete observation conditions. Therefore, how to establish a method that can identify and track the Southwest Low Vortex over long time series with verifiable results and stable performance throughout the entire process is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] (a) Purpose of the invention To address the aforementioned deficiencies and shortcomings of existing technologies, this invention aims to provide an objective identification method and medium for the entire process of a Southwest low-pressure vortex based on grid reanalysis data. This method involves preprocessing the geopotential height field using a topographic mask within a predefined identification area, extracting closed low-pressure candidate bodies based on closed contour lines and calculating the low-pressure center location. It then constructs candidate time-series trajectories by combining closed area attribution screening, spatial proximity correlation of adjacent time periods, and displacement constraints. Valid low-pressure processes are confirmed using duration criteria, and dynamic consistency verification is introduced when necessary to suppress false detections of non-low-pressure disturbances. The method outputs the start and end times, duration, and movement path of the low-pressure process, and can statistically evaluate the activity frequency, path type, and environmental field and precipitation impact characteristics. This achieves automated identification, stable tracking, and consistent statistics of long-term Southwest low-pressure series, reducing subjective judgment bias and labor costs, and improving the verifiability and operational availability of identification results under complex terrain conditions.

[0008] (II) Technical Solution To achieve the objective of this invention and solve its technical problems, the present invention adopts the following technical solution: The first objective of this invention is to provide an objective identification method for the entire process of the Southwest Low-Temperature Vortex based on grid reanalysis data. This method aims to objectively identify the entire process of the Southwest Low-Temperature Vortex from its generation, development, movement, and dissipation from grid reanalysis data, and output its spatiotemporal evolution characteristics such as occurrence time, movement path, and duration. The method includes at least the following steps: S100. Identification Area Determination and Data Preprocessing: Set the target identification area covering the main activity area of ​​the southwest low vortex, acquire grid reanalysis data containing at least the preset geopotential height field of the pressure layer, generate a terrain mask based on the terrain height field, and remove grid points with an altitude exceeding the preset terrain threshold. S200. Closed low-pressure candidate extraction: For the geopotential height field at each time point, search for closed contour lines within the target identification area according to the preset contour interval, determine the closed area enclosed by the closed contour lines, calculate the low-pressure center position corresponding to each closed area, and form a set of closed low-pressure candidates; S300. Region Attribution Screening: For each closed low-pressure candidate, calculate the percentage R of the area of ​​intersection between its closed region and the target recognition region, retain closed low-pressure candidates whose intersection area percentage R reaches the preset spatial matching threshold as valid candidates, and remove the rest of the closed low-pressure candidates. S400. Temporal correlation and trajectory construction: Using effective candidates between adjacent time periods as matching objects, correlation matching is performed based on the spatial proximity of the low-pressure center location to construct candidate temporal trajectories; when there are multiple effective candidates at the same time period, corresponding candidate temporal trajectories are constructed respectively. S500. Displacement constraint process tracking: For each candidate time series trajectory, calculate the movement distance D between its adjacent time low pressure center positions. If D does not exceed the preset displacement threshold, the time series association is maintained as the same candidate body trajectory; otherwise, the time series association is interrupted and the trajectory is divided into different candidate body trajectories. S600. Duration Criterion and Low-Vortex Confirmation: For each candidate trajectory, its duration T is calculated. If the duration T reaches a preset time threshold, it is determined to be a valid southwest low-vortex process. S700. Output Results: For each determined southwest vortex process, output its start time, end time, duration, and the movement path formed by connecting the positions of the low-pressure centers in chronological order.

[0009] The second objective of this invention is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for objectively identifying the entire process of the Southwest Low-Temperature Vortex based on grid reanalysis data.

[0010] (III) Technical Effects Compared with existing technologies, the objective identification method and medium for the entire process of the Southwest Low Vortex based on grid reanalysis data of the present invention have the following beneficial and significant technical effects: (1) This invention eliminates invalid computation domains at high altitudes and introduces closed region attribution screening by using terrain mask and target recognition area constraints, making the closed low pressure identification more consistent with the southwest low vortex activity area, effectively suppressing the interference of pseudo-closed disturbances caused by complex terrain and regional boundaries, reducing false detections and missed detections, improving the spatial consistency and verifiability of the identification results, and enhancing the credibility and stability of long-term series statistics.

[0011] (2) This invention achieves continuous tracking of the entire process of low eddies from generation, development, movement and disappearance by means of closed contour line extraction, center positioning, spatial proximity association of adjacent time periods, and displacement constraint trajectory, thereby reducing the subjective differences and inconsistencies in event segmentation caused by manual identification; at the same time, it filters short-term disturbances by using duration criteria, making the identification results more in line with the temporal continuity requirements of the low eddy process, and significantly improving the efficiency of automated processing and the quality of sample library construction.

[0012] (3) In the process confirmation stage, the present invention can introduce dynamic consistency verification of relative vorticity and convergence, and support statistical evaluation of activity frequency, path type and environmental field-precipitation impact in the output stage, thereby forming a structured data product that can directly serve business and scientific research on the basis of ensuring identification purity. This product can be used for rainstorm monitoring assessment, forecast verification and mechanism diagnosis, and improve the attribution analysis capability and business application value of regional heavy precipitation events. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the 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.

[0014] Figure 1 This is a flowchart of the method for objectively identifying the entire process of the Southwest Low Vortex based on grid reanalysis data provided in this embodiment of the invention; Figure 2 The image shows the topographic elevation distribution map of the Southwest region, where the horizontal axis represents longitude (°E) and the vertical axis represents latitude (°N). The color depth indicates the topographic elevation (unit: meters). Darker areas are plateau regions with an elevation of over 3000 meters, which need to be removed during the identification process using a topographic mask. Figure 3The chart shows a comparison of the frequency of occurrence of Southwest China vortices in the summer of 2012-2022 based on the objective Southwest China vortex dataset and the "Southwest China Vortex Yearbook". (a) represents the overall statistical results for summer, (b) represents the statistical results for June, (c) represents the statistical results for July, and (d) represents the statistical results for August. The bar chart represents the statistical results of the objective Southwest China vortex dataset identified by the method of this invention. The black dotted solid line represents the statistical results of manual identification in the "Southwest China Vortex Yearbook". The horizontal axis represents the year, and the vertical axis represents the frequency of occurrence of Southwest China vortices. Detailed Implementation

[0015] This invention aims to provide a method and medium for the objective identification of the entire process of the Southwest Low-Temperature Vortex based on grid reanalysis data. To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be described in more detail below with reference to the accompanying drawings. The described embodiments are some, but not all, embodiments of this invention, and are exemplary and intended to explain this invention, not to limit it. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0016] Example 1: An Objective Identification Method for the Entire Process of the Southwest Low-Temperature Vortex like Figure 1 As shown in the illustration, as a specific example, the embodiment of the present invention provides a method for objectively identifying the entire process of the Southwest Low-Pressure from its generation, development, movement to dissipation from grid reanalysis data and outputting its spatiotemporal evolution characteristics such as occurrence time, movement path, and duration. This method mainly includes the following steps: S100. Identification Area Determination and Data Preprocessing: A target identification area covering the main activity area of ​​the Southwest Low Vortex is defined, and grid reanalysis data containing at least the preset geopotential height field of the pressure layer is obtained. A terrain mask is generated based on the terrain height field, and grid points with an altitude exceeding the preset terrain threshold are removed to obtain an effective computational domain for closed structure identification and reduce the height field distortion introduced by terrain forcing.

[0017] Preferably, the target identification area covers at least the latitude and longitude range of 26~33°N and 100~109°E. The preset terrain threshold is 3000m. A terrain mask is generated according to the preset terrain threshold. The grid points in the target identification area are divided into retention areas and elimination areas. The grid points in the elimination area are removed from the geopotential height field calculation domain to reduce the interference of high terrain areas (such as the Qinghai-Tibet Plateau and other high-altitude terrain) on the identification of the southwest low vortex. This ensures that the closed structure of the subsequent closed contour lines mainly reflects the characteristics of the low-pressure vortex in the atmospheric circulation rather than the height field distortion caused by terrain forcing, thus ensuring the consistency and verifiability of the identification area. The preset pressure layer is the 700hPa isobaric surface. The horizontal spatial resolution of the grid reanalysis data is 0.25°×0.25°, and the temporal resolution is 6 hours. Steps S200 to S700 are all performed to extract, screen, associate, and confirm the closed low-pressure candidate in consecutive time units of adjacent 6-hour intervals. This ensures that the process identification results are consistent with the time sampling of the grid reanalysis data and facilitates the formation of long-term series statistical samples.

[0018] In addition, this step also includes consistent spatiotemporal gridding preprocessing of the geopotential height field: quality control and removal of missing or non-numerical grid points; registration of the geopotential height field, topographic height field and target identification area using the same coordinate system and the same grid resolution; when the grid of the original reanalysis data is inconsistent with the boundary of the target identification area, the grid points at the boundary are clipped or resampled to ensure that the subsequent closed contour line search and intersection area calculation have a consistent geometric basis, thereby reducing the false closure and false center introduced by resolution differences.

[0019] S200. Closed-loop low-pressure candidate extraction: For each time-varying geopotential height field, closed contour lines are searched within the target identification region at a preset isopleth interval to determine the closed regions enclosed by these contour lines. The location of the low-pressure center corresponding to each closed region is calculated, forming a set of closed low-pressure candidate bodies. Preferably, the preset isopleth interval is 2 gpm, and the formation of closed loops by geopotential height contour lines within the target identification region is used as the criterion for closed contour lines. For each closed region enclosed by a closed loop, its boundary and internal grid set are determined using a grid connectivity determination method. The location of the low-pressure center is determined by the grid position corresponding to the minimum geopotential height within the closed region. When multiple minimum geopotential height points exist within a closed region, the geometric center of these minimum points is calculated as the location of the low-pressure center. Furthermore, the search process for closed contour lines adopts a progressively decreasing approach, starting from high geopotential height values ​​and gradually searching towards low geopotential height values ​​until all contour lines satisfying the closure condition are identified.

[0020] Furthermore, in the process of extracting closed low-pressure candidate bodies, multiple closed regions identified at the same time are hierarchically organized according to their spatial inclusion relationship: when one closed region is completely contained within another closed region, the two are established as parent and child, and only the innermost closed region is retained as an independent closed low-pressure candidate body. This avoids different contour levels of the same low-pressure system being repeatedly identified as multiple candidates, thereby improving the independence of the candidate body set and the accuracy of subsequent matching.

[0021] S300. Regional Attribution Filtering: For each closed low-pressure candidate, calculate the percentage R of the intersection area between its closed region and the target recognition region. Closed low-pressure candidates whose intersection area percentage R reaches the preset spatial matching threshold are retained as valid candidates, and the rest of the closed low-pressure candidates are eliminated.

[0022] As a preferred option, the intersection area ratio R is defined as the ratio of the intersection area of ​​the closed region and the target identification region to the area of ​​the closed region. Both the intersection area and the area of ​​the closed region are calculated after removing invalid computational domains corresponding to the terrain mask. The preset spatial matching threshold is 0.90. When R ≥ 90%, the main body of the closed low-pressure candidate is determined to be located in the target identification region and retained as a valid candidate. When R < 90%, the main body of the closed low-pressure candidate is determined to be located outside the target identification region and is removed. Through this regional attribution screening, it is ensured that the identified low-pressure system conforms to the geographical location definition of the Southwest Low-Pressure, thereby suppressing false detection events where the closed structure is mainly located outside the identification region.

[0023] In addition, between steps S200 and S300, there is also a closed shape validity judgment: the area, perimeter and equivalent circularity index of the closed region are calculated. When the circularity is lower than the preset shape threshold or the closed region has an obvious narrow strip structure, the corresponding closed low pressure candidate is judged as a groove-shaped disturbance rather than a vortex-type closed low pressure and is eliminated to improve the recognition purity and stability in complex background fields.

[0024] S400. Temporal correlation and trajectory construction: Using valid candidate bodies between adjacent time periods as matching objects, correlation matching is performed based on the spatial proximity of the low-pressure center location to construct candidate time-series trajectories; when multiple valid candidate bodies exist at the same time period, corresponding candidate time-series trajectories are constructed for each. The candidate time-series trajectory represents the sequence of center locations of the same low-pressure system across consecutive time periods, used to describe its generation, movement, and dissipation processes.

[0025] Preferably, when performing correlation matching based on the spatial proximity of the low-pressure center location, the following methods are included: S401. For valid candidate i at time t and valid candidate j at time t+Δt, calculate the spatial distance d between their low-pressure center locations. ij ; S402. Among all pairing combinations, if k represents all valid candidates at time t, and k' represents all valid candidates at time t+Δt, select the pair that satisfies d ij =min{d ij |i=1,2,…k,j=1,2,…k'}, that is, the spatial distance d between adjacent time intervals. ij The smallest pairing is used as the temporal association relationship, associating candidate i and candidate j as adjacent temporal states of the same candidate temporal trajectory; S403. When there are more than two valid candidates at the same time, they are sorted from smallest to largest according to the geopotential height of the low-pressure center and the top two candidates are retained for subsequent association matching. This is to take into account both the completeness of identification and the stability of trajectory construction under the coexistence of multiple vortices. Candidate time-series trajectories are independently constructed for the two retained candidates. For candidate associations that tend to merge or split, the one with the smallest distance and that satisfies the displacement constraint is used as the continuation association, so as to realize the identification of the dual-core southwest low vortex.

[0026] Furthermore, a correlation confidence assessment is introduced during the temporal correlation process: in addition to spatial distance, multi-dimensional features such as the intensity change rate of adjacent time-series candidates, consistency of movement direction, and change rate of closed area are comprehensively considered to calculate the comprehensive confidence of correlation matching; when the confidence of the minimum distance matching is lower than the preset threshold while the confidence of the suboptimal matching is significantly higher, the suboptimal matching is selected as the temporal correlation to cope with the possible trajectory intersection or rapid reorganization under complex weather conditions, and to further improve the physical rationality of trajectory construction.

[0027] S500. Process tracking of displacement constraints: For each candidate time-series trajectory, the distance D between adjacent low-pressure center locations is calculated. If D does not exceed a preset displacement threshold, the time-series association is maintained as the same candidate trajectory; otherwise, the time-series association is interrupted and the trajectory is divided into different candidate trajectories. Preferably, the preset displacement threshold is that the distance between adjacent low-pressure centers does not exceed 400 km. The calculation of the distance D includes: S501. Based on the latitude and longitude coordinates of the centers of adjacent low-pressure systems, calculate the great circle distance between two points using the spherical distance or projected plane distance formula; S502. When D≤400km, determine that the effective candidate bodies of adjacent time intervals belong to different time states of the same low-vortex system, and maintain temporal correlation; S503. If D>400km, determine that the effective candidate bodies in adjacent time periods do not belong to the same low-vortex system, interrupt the time series correlation, and divide the original candidate time series trajectory into two candidate body trajectories at that time period; S504. When the movement direction of the candidate time series trajectory deviates from the dominant direction of the 500hPa airflow by more than a preset angle threshold, it is determined that the association does not meet the large-scale guiding airflow condition and the time series association is interrupted, so as to reduce the erroneous association and trajectory jump caused by local disturbances and improve the reliability of the identification of the southwest vortex.

[0028] Furthermore, at least one preset threshold in steps S300 to S500 is determined using an adaptive update method: the background circulation intensity index is calculated based on the 500 hPa wind speed or geopotential height gradient intensity within the target identification area, and the spatial matching threshold and displacement threshold are adjusted accordingly, so that the displacement threshold is relaxed when the background circulation intensity increases and the displacement threshold is tightened when the background circulation intensity decreases, thereby taking into account the tracking robustness under different seasons and different circulation conditions.

[0029] S600. Duration Criteria and Low-Voltage Confirmation: For each candidate trajectory, its duration T is recorded. If the duration T reaches a preset time threshold, it is determined to be a valid southwest vortex process. Preferably, the preset time threshold is 12 hours, and the duration T is defined as the cumulative duration of the candidate trajectory continuously satisfying the constraints of steps S200 to S500 in adjacent time intervals on the time axis. If the duration T of the candidate trajectory is ≥ 12 hours, it is determined to be a valid southwest vortex process and its intensity evolution information is recorded. When the candidate trajectory is missing or segmented in adjacent time intervals, it is considered an interruption of continuity and the timing is restarted, thereby avoiding the accumulation of discontinuous closed disturbances into the same southwest vortex process.

[0030] In addition, step S600 includes a dynamic consistency check before determining the candidate body trajectory as a southwest low vortex process: calculating the 700hPa relative vorticity and 850hPa horizontal wind field convergence in each closed region, and requiring the relative vorticity to reach a preset vorticity threshold and the convergence to reach a preset convergence threshold; when the dynamic consistency condition is not met at any time, the corresponding time node is removed or the corresponding candidate body trajectory is terminated, so as to reduce the pseudo-closed structure formed only by local noise of geopotential height field.

[0031] Preferably, when analyzing the duration T of the candidate body trajectory, the analysis further includes the change in the intensity of the low-pressure vortex: for each candidate body trajectory, the geopotential height value at the low-pressure center at each time point is extracted, and the difference between this geopotential height value and the geopotential height value corresponding to the outermost contour line of the closed contour line is calculated. This difference is defined as the low-pressure vortex intensity index for that time point. The maximum, minimum, average, and standard deviation of the low-pressure vortex intensity index throughout the entire life history of the candidate body trajectory are statistically analyzed and used as the intensity characteristic parameters of the southwest low-pressure vortex process.

[0032] S700. Output: For each identified southwest vortex process, output its start time, end time, duration, and the movement path formed by connecting the locations of the low-pressure centers in chronological order.

[0033] Preferably, step S700 further includes statistical analysis of the activity characteristics of the southwest low-pressure system: All identified southwest vortex events were categorized and statistically analyzed according to their occurrence time, and the frequency of occurrence of southwest vortices in different months and seasons was calculated. Cluster analysis was performed on the movement paths to identify the main movement path types of southwest vortices. Statistical indicators such as the average movement speed, average duration, and average impact range of southwest vortices were calculated. A database of southwest vortex activity was established, containing complete spatiotemporal evolution information, intensity characteristics, movement characteristics, and environmental background field parameters for each vortex event. This database can be used for the analysis of the climate characteristics of southwest vortices, training of forecast models, and research on their impact on weather and climate.

[0034] This step further includes extracting the environmental field characteristics of the Southwest Low Vortex: for each identified Southwest Low Vortex process, multi-level atmospheric circulation characteristic parameters of the region surrounding the vortex center are extracted during its occurrence. These characteristic parameters include at least the 500 hPa geopotential height field, the 850 hPa temperature field, the total precipitable water content, the vertical velocity field, the relative vorticity field, and the divergence field. The dynamic and thermodynamic conditions during the occurrence and development of the Southwest Low Vortex are analyzed. The vertical vorticity advection, temperature advection, and water vapor flux divergence at the vortex center are calculated to quantitatively assess the contribution of different physical processes to the maintenance and intensification of the Southwest Low Vortex. A correlation dataset between individual Southwest Low Vortex cases and their environmental field characteristics is established to provide a data foundation for a deeper understanding of the formation mechanism of the Southwest Low Vortex.

[0035] Preferably, step S700 further includes an assessment of the impact of the southwest low-pressure system: for each identified southwest low-pressure system process, extract precipitation observation data or precipitation reanalysis data during its influence period and within its influence area; define the influence area of ​​the low-pressure system as the area within a preset radius around the low-pressure center, the preset radius being dynamically determined based on the spatial scale of closed isolines; statistically analyze the cumulative precipitation, maximum hourly precipitation intensity, and precipitation duration within the influence area; identify rainstorm events related to the southwest low-pressure system and establish the correspondence between the activity of the southwest low-pressure system and regional heavy precipitation; calculate the contribution rate of the southwest low-pressure system to regional precipitation and assess the impact of the activity of the southwest low-pressure system on water resource distribution and drought and flood disasters in the southwest region.

[0036] It should be noted that the objective identification method for the entire process of the Southwest Low-Temperature Vortex proposed in this embodiment organically combines multiple layers of algorithms, including closed contour line search, regional attribution screening, spatial proximity association, displacement constraint tracking, and duration criteria, to construct a complete identification mechanism from the geopotential height field to the spatiotemporal evolution characteristics of the low-pressure vortex. In step S300, the method introduces the quantitative indicator of the proportion of intersecting areas in the regional attribution screening, solving the problem of objectively determining the geographical attribution of the low-pressure vortex. In steps S400-S500, the trajectory construction comprehensively utilizes spatial distance optimization, displacement threshold constraints, and consistency checks of guiding airflow, effectively avoiding trajectory jumps and erroneous associations. In step S600, the low-pressure vortex confirmation integrates duration criteria and dynamic consistency verification, ensuring the reliability of the identification results from both temporal scale and physical mechanism dimensions. The synergistic effect of the above key technical steps enables this method to accurately capture the entire process of the formation, development, movement and dissipation of the Southwest Low Vortex, and effectively suppress false detections and missed detections under complex background fields. This significantly improves the accuracy, stability and automation of the objective identification of the Southwest Low Vortex, and provides reliable technical support for the study of the climate characteristics of the Southwest Low Vortex and its weather and climate impact assessment.

[0037] Example 2: Application Case Based on the above Example 1, Example 2 further provides an application case of objective identification of the Southwest Low Vortex based on the first-generation global atmospheric and land surface reanalysis product CRA-40 independently developed by the China Meteorological Administration, in order to verify the actual identification effect, statistical analysis capability and business application value of the method of the present invention.

[0038] Data Acquisition and Computation Domain Setup: The CRA-40 reanalysis dataset was selected as the input data, with a time resolution of 6 hours and a horizontal resolution of approximately 25 km. Following the requirements of step S100 in Example 1, the target identification region was defined as 26–33°N and 100–109°E, and 700 hPa isobaric surface geopotential height field data was extracted. A terrain mask was generated based on the terrain height field data, and grid points with an altitude exceeding 3000 m were removed. Figure 2As shown, the excluded areas are mainly distributed in the eastern edge of the Qinghai-Tibet Plateau in the northwestern part of the identified area. Regarding parameter configuration, the preset isointerval is 2 gpm, the preset spatial matching threshold is 90%, the preset displacement threshold is 400 km, and the preset time threshold is 12 hours.

[0039] Closed low-pressure candidate body extraction (corresponding to step S200): Perform closed contour line search on the 700 hPa geopotential height field for every 6 hours, with the contour line analysis interval set to 2 gpm. For each closed region, extract the point with the minimum geopotential height within the closed region as the low-pressure center location, and record the boundary, area, and outer scale of the closed region to form a set of closed low-pressure candidate bodies.

[0040] Regional attribution screening (corresponding to step S300): For each closed low-pressure candidate, calculate the percentage of the area R of the intersection between its closed region and the target identification region. If R ≥ 90%, it is retained as a valid candidate; otherwise, it is discarded to ensure that the identified object is mainly located within the defined area of ​​the southwest low vortex.

[0041] Displacement and steering airflow consistency constraint (corresponding to step S500): Calculate the displacement D of the sub-low pressure center over the next 6 hours for each candidate trajectory. If D ≤ 400 km, maintain the association; if D > 400 km, interrupt the association and split the trajectory. Simultaneously, apply consistency constraints to the trajectory movement direction with reference to the 500 hPa airflow direction.

[0042] Duration criterion confirmation (corresponding to step S600): Calculate the continuous duration T of each trajectory. When T≥12 hours, it is determined to be an effective southwest low vortex process, and output its start and end times, duration, hourly sub-center position and movement path, etc.

[0043] Results Output, Comparative Evaluation, and Product Display (corresponding to step S700): The identified Southwest Low-Pressure Processes are statistically analyzed monthly and quarterly, outputting indices such as frequency of occurrence, average duration, and intensity, and forming a queryable monitoring product. To verify the reliability of the identification results of the method of this invention, the identified frequencies of occurrence in the summer of 2012-2022 and from June to August are compared with the manually identified results in the "Southwest Low-Pressure Yearbook" (see...). Figure 3 ).like Figure 3As shown, in the overall summer statistics, the frequency of southwest vortices identified by this method shows a high degree of consistency with the yearbook results, with a correlation coefficient of 0.92. In the monthly comparisons for June, July, and August, the interannual trends of the two methods are basically consistent, with correlation coefficients of 0.89, 0.91, and 0.88, respectively. Statistical analysis shows that the average deviation rate of this method compared to manual identification is -3.2%, with a standard deviation of 2.1 times / year, proving that the method of this invention can accurately and objectively identify southwest vortex processes. Furthermore, integrating the identification results into the operational platform to form a productized output of monthly frequency, duration, and intensity allows for the retrieval of southwest vortex activity paths and intensity indicators by any time period, demonstrating that the method of this invention has operational feasibility. It can support the construction and real-time monitoring application of long-term historical datasets of southwest vortices, filling the gap in objective monitoring of southwest vortices in the sub-season in my country and providing key technological support for weather forecasting, climate monitoring, and disaster early warning by meteorological operational departments.

[0044] The objectives of this invention have been fully and effectively achieved through the above embodiments. Those skilled in the art will understand that this invention includes, but is not limited to, the contents described in the accompanying drawings and the specific embodiments described above. Although the invention has been described with reference to what is currently considered the most practical and preferred embodiments, it should be understood that the invention is not limited to the disclosed embodiments, and any modifications that do not depart from the functional and structural principles of the invention will be included within the scope of the claims.

Claims

1. A method for objective identification of the whole process of southwest vortex based on grid reanalysis data, characterized in that, At least comprising the following steps: S100. Set a target identification area covering the main activity area of the southwest vortex, obtain grid reanalysis data containing at least the potential height field of a preset pressure layer, generate a terrain mask based on the terrain height field, and remove grid points with an altitude exceeding a preset terrain threshold; S200. For each time instance of the potential height field, search for closed contour lines in the target identification area at a preset contour interval, determine the closed regions surrounded by the closed contour lines, calculate the low-pressure center positions corresponding to each closed region, and form a closed low-pressure candidate set; S300. For each closed low-pressure candidate, calculate the intersection area ratio R of the closed region and the target identification area, and retain the candidate that reaches a preset spatial matching threshold as an effective candidate; S400. Take the effective candidates between adjacent time instances as matching objects, and perform correlation matching based on the spatial proximity of the low-pressure center positions to construct candidate time sequence trajectories; when there are multiple effective candidates at the same time instance, corresponding candidate time sequence trajectories are constructed respectively; S500. For each candidate time sequence trajectory, calculate the moving distance D between the low-pressure center positions of adjacent time instances, if D does not exceed a preset displacement threshold, maintain the time sequence correlation as the same candidate trajectory, otherwise, interrupt the time sequence correlation and divide the trajectory into different candidate trajectories; S600. For each candidate trajectory, calculate its duration T, if T reaches a preset time threshold, determine it as an effective southwest vortex process; S700. For each determined southwest vortex process, output its start time, end time, duration, and moving path formed by connecting the low-pressure center positions in time sequence.

2. The method of claim 1, wherein, In step S100, the target identification area at least covers the latitude and longitude range of 26-33 °N and 100-109 °E; the preset terrain threshold is 3000 m, the terrain mask is generated according to the preset terrain threshold, the grid points in the target identification area are divided into a retained area and a removed area, and the grid points in the removed area are removed from the potential height field calculation domain; the preset pressure layer is the 700 hPa isobaric surface, the horizontal spatial resolution of the grid reanalysis data is 0.25 °x 0.25 °, the time resolution is 6 hours, and steps S200-S700 are executed for the extraction, screening, correlation and confirmation of closed low-pressure candidates with adjacent 6-hour time instances as continuous time sequence units.

3. The method according to claim 1 or 2, characterized in that, In step S100, the potential height field is also subjected to consistent spatiotemporal gridding preprocessing: quality control and removal of missing or non-numeric grid points; the potential height field, terrain height field and target identification area are registered using the same coordinate system and grid resolution; When the grid of the original reanalysis data is inconsistent with the boundary of the target identification area, the grid points at the boundary are cropped or resampled.

4. The method of claim 1, wherein, In step S200, the preset equivalent distance is 2 gpm, and the potential height contour is used to form a closed loop in the target identification area as a closed contour criterion; for each closed area surrounded by a closed loop, a grid connected domain determination method is used to determine the boundary and internal grid set of the closed area, and the grid position corresponding to the minimum value of the potential height in the closed area is used to determine the position of the low pressure center. When there are multiple minimum value points with equal potential height in a closed area, the geometric center of these minimum value points is calculated as the position of the low pressure center; and the search process of the closed contour is performed in a step-by-step decreasing manner, starting from a high potential height value and gradually searching towards a low potential height value, until all contours meeting the closed condition are identified.

5. The method of claim 1, wherein, In step S300, the intersection area ratio R is defined as the ratio of the intersection area of the closed area and the target identification area to the area of the closed area, and both the intersection area and the area of the closed area are calculated after excluding the invalid calculation domain corresponding to the terrain mask; The preset spatial matching threshold is 0.90, when R≥90%, it is determined that the main body of the closed low pressure candidate is located in the target identification area and is retained as an effective candidate, and when R<90%, it is determined that the main body of the closed low pressure candidate is located outside the target identification area and is excluded.

6. The method of claim 1, wherein, In step S400, the spatial proximity relationship of the low pressure center position is used for correlation matching, including: S401. For the valid candidate i at time t and the valid candidate j at time t+Δt, calculate the spatial distance d between the two low-pressure center positions ij ; S402. Among all pairing combinations, if k represents all valid candidates at time t, and k' represents all valid candidates at time t+Δt, select the pair that satisfies d ij =min{d ij |i=1,2,…k,j=1,2,…k'}, that is, the spatial distance d between adjacent time intervals. ij The smallest pairing is used as the temporal association relationship, associating candidate i and candidate j as adjacent temporal states of the same candidate temporal trajectory; S403. When there are more than 2 effective candidates at the same time, the low pressure center potential height is sorted from small to large, and the first 2 candidates in the order are retained for subsequent correlation matching. The candidate time sequence trajectory is independently constructed for the retained 2 candidates, and the candidate correlation that tends to merge or split is continued by distance minimization and displacement constraint satisfaction, so as to realize the identification of the double-core southwest low vortex.

7. The method of claim 1, wherein, In step S500, the preset displacement threshold is that the moving distance of the low pressure center between adjacent time is not more than 400 km, and the calculation of the moving distance D includes: S501. According to the longitude and latitude coordinates of the low pressure center positions of adjacent times, the great circle distance between two points is calculated by using the spherical distance or projection plane distance formula; S502. When D≤400km, it is determined that the effective candidates of adjacent times belong to different time states of the same low vortex system, and the time sequence correlation is maintained; S503. If D>400km, it is determined that the effective candidates of adjacent times do not belong to the same low vortex system, and the time sequence correlation is interrupted, and the original candidate time sequence trajectory is divided into two candidate body trajectories at this time; S504. When the moving direction of the candidate time sequence trajectory deviates from the dominant direction of the 500 hPa air flow by more than a preset angle threshold, it is determined that the correlation does not meet the large-scale guiding air flow condition and the time sequence correlation is interrupted.

8. The method of claim 1, wherein, In step S600, the preset time threshold is 12 hours, and the continuous time length T is defined as the cumulative time length of the candidate body trajectory satisfying the constraints of steps S200-S500 in adjacent time on the time axis. If the continuous time length T of the candidate body trajectory is greater than or equal to 12 hours, it is determined that the candidate body trajectory is a valid southwest vortex process, and the intensity evolution information of the southwest vortex process is recorded. When the candidate body trajectory is missing or segmented in adjacent time, it is considered that the continuity is interrupted, and the time is restarted.

9. The method according to claim 1 or 8, characterized in that, In step S600, before determining that the candidate body trajectory is a southwest vortex process, a dynamic consistency review is further included. The relative vorticity at 700 hPa and the convergence of the horizontal wind field at 850 hPa are calculated in each closed region, and it is required that the relative vorticity reaches a preset vorticity threshold and the convergence reaches a preset convergence threshold. When any time does not satisfy the dynamic consistency condition, the corresponding time node is removed or the corresponding candidate body trajectory is terminated.

10. The method of claim 1, wherein, Step S700 further includes statistical analysis of the characteristics of the southwest vortex activity. For all identified southwest vortex processes, the occurrence time is classified and counted, and the occurrence frequency of the southwest vortex in different months and different seasons is calculated. Cluster analysis is performed on the moving path to identify the main moving path type of the southwest vortex. The average moving speed, average duration, and average influence range of the southwest vortex are calculated.

11. The method of claim 1, wherein, Step S700 further includes the extraction of the environmental field characteristics of the southwest vortex. For each determined southwest vortex process, the multi-level atmospheric circulation characteristic parameters of the region around the low vortex center during the occurrence of the southwest vortex are extracted, and the characteristic parameters at least include the 500 hPa geopotential height field, the 850 hPa temperature field, the whole layer precipitable water, the vertical velocity field, the relative vorticity field, and the divergence field. The dynamic and thermal conditions during the occurrence and development of the southwest vortex are analyzed. The vertical vorticity advection, temperature advection, and water vapor flux divergence at the low vortex center are calculated to quantitatively evaluate the contribution of different physical processes to the maintenance and strengthening of the southwest vortex.

12. The method of claim 1, 10 or 11, wherein, Step S700 further includes the evaluation of the influence of the southwest vortex. For each determined southwest vortex process, the precipitation observation data or the precipitation reanalysis data in the influence period and the influence region are extracted. The influence region of the low vortex is defined as the region within a preset radius around the low pressure center, and the preset radius is dynamically determined according to the spatial scale of the closed contour. The cumulative precipitation, maximum hourly precipitation intensity, and precipitation duration in the influence region are counted. Rainstorm events related to the southwest vortex are identified, and the corresponding relationship between the southwest vortex activity and the regional heavy precipitation is established. The contribution rate of the southwest vortex to the regional precipitation is calculated to evaluate the influence of the southwest vortex activity on the water resource distribution and the drought and flood disasters in the southwest region.

13. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the southwest vortex whole-process objective identification method based on grid reanalysis data according to any one of claims 1-12.

Images