Global ocean mesoscale vortex objective tracking method, device and equipment and storage medium

Abstract

The invention belongs to the technical field of ocean big data analysis, and particularly provides a global ocean mesoscale vortex objective tracking method, device and equipment and a storage medium, and the method comprises the steps: obtaining global absolute dynamic topographic data; global mesoscale vortex data and potential vortex data are identified to form global mesoscale vortex identification data; performing particle advection on the centroids of all vortexes and potential vortexes on the ith day, and setting an integral time step length as a day to obtain longitude and latitude positions of the centroids of all vortexes and potential vortexes on the ith day from advection to the (i + a) th day; the relation Segment between the two continuous time step length vortexes is established; and establishing a linear structure Branch formed by connecting the Segments in a plurality of survival states, outputting a Branch data set, and storing the Branch data set in a file in a pkl format. The defect that subjective parameters and pseudo vortexes need to be set in an existing method is overcome, objective tracking of global mesoscale vortexes is achieved, and a data set is formed.

Description

Technical Field

[0001] The present invention relates to the technical field of ocean big data analysis, and in particular to a method, device, equipment and storage medium for objectively tracking mesoscale vortices in the global ocean. Background Art

[0002] In the field of marine science, mesoscale eddies are a natural phenomenon that is ubiquitous in the global ocean. Their unique dynamic characteristics and multifaceted impacts on the marine environment make them a hot spot and difficulty in marine research. Mesoscale eddies, with their spatial scales of tens to hundreds of kilometers, not only significantly change the water flow structure inside the ocean through rotational motion, but also profoundly affect the vertical position of the thermocline and salinity cline, thereby having a long-term impact on the global climate system. Therefore, whether in understanding basic theories such as ocean dynamics, or in practical application fields such as marine resource exploration, environmental monitoring, and marine ecological protection, the study of mesoscale eddies has shown extremely high scientific value and social significance.

[0003] At present, a variety of technical means and methods have been developed for tracking and identifying mesoscale eddies in the ocean, mainly including the pixel method, the nearest neighbor method and the similarity method. These methods have their own characteristics, but all aim to achieve automatic or semi-automatic identification and tracking of eddies by analyzing ocean observation data (such as satellite remote sensing data, buoy observation data, etc.). However, the above three methods all face certain challenges and limitations in practical applications. First, data quality is a key factor affecting the accuracy of tracking results. Noise and errors in ocean observation data may cause deviations in tracking results. In order to alleviate this problem, researchers often evaluate the performance of the algorithm by setting "pseudo-eddies", but these "pseudo-eddies" are not real eddies, so they cannot fully reflect the complexity of the real ocean environment. Secondly, these methods are more dependent on the setting of initial parameters during the tracking process, such as eddy identification threshold, tracking step size, etc. The setting of these parameters often needs to be adjusted according to the specific research area and data characteristics, which is highly subjective. Different parameter settings may lead to large differences in tracking results.

[0004] In summary, although the existing mesoscale vortex tracking methods have achieved automatic or semi-automatic identification and tracking of vortices to a certain extent, they still have problems such as strong dependence on data quality, susceptibility to noise interference, and high subjectivity in initial parameter settings. Therefore, there is an urgent need for a new objective tracking method for mesoscale vortices that does not rely on parameter settings and does not require the setting of "pseudo-vortices". Summary of the invention

[0005] In view of the problem that existing vortex tracking methods are highly subjective, the present invention provides a method, device, equipment and storage medium for objectively tracking mesoscale vortices in the global ocean.

[0006] In a first aspect, the technical solution of the present invention provides a method for objectively tracking global ocean mesoscale eddies, comprising the following steps: Obtain absolute dynamic terrain data around the world; Process the global absolute dynamic terrain data to identify the global mesoscale eddy data and potential eddy data to form the global mesoscale eddy identification data; Based on the global mesoscale eddy identification data, the centroids of all eddies and potential eddies on the i-th day are advected to the longitude and latitude positions of the i+a-th day. The integration time step is set to a day. Establish the relationship between two consecutive time step vortices Segment; Establish a linear structure Branch formed by connecting segments in multiple survival states and output the Branch data set; The output Branch dataset is stored in pkl format files, where each file consists of all eddy tracks for that year.

[0007] As a further limitation of the technical solution of the present invention, the steps of processing the global absolute dynamic terrain data to identify the global mesoscale vortex data and the potential vortex data to form the global mesoscale vortex identification data include: Perform high-pass filtering on the global absolute dynamic terrain data; Generate closed ADT contours of global vortex data based on filtered data, and identify local extreme points as vortex seed points; Traverse the ADT contour lines of global vortex data to determine whether each contour line meets the screening conditions of the vortex boundary; if it meets the conditions, identify the vortex; if it does not meet the conditions, identify the potential vortex; The identified vortex is merged with the potential vortex data, and after completing the contour traversal of the global vortex data, the results are saved and output to form the global mesoscale vortex identification data.

[0008] As a further limitation of the technical solution of the present invention, the step of identifying the potential vortex includes: When traversing the ADT contour lines of the global vortex data, the contour lines that contain only one vortex seed point but do not meet the vortex boundary screening conditions are identified as potential vortices.

[0009] As a further limitation of the technical solution of the present invention, the step of establishing a relationship Segment between two consecutive time step vortices includes: Receive the input global vortex centroid set C on the i-th day i, and conduct particle advection with a time step of a day to obtain the vortex center set C advected to the i + a day i+1 ; C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 ; Judgement in E i+1 Is there C in i+1 The vortex center in falls into the vortex within the effective boundary. If there is one, a Live Segment is constructed, otherwise a Dead Segment is constructed; C i+1 After the traversal is completed, all Live Segments are traversed again to determine whether there is C i+1 If multiple vortex centers fall within the same vortex boundary, if so, calculate the distances between the multiple vortex centers and the vortex centroid, and select the one with the smallest distance for matching; When the Live Segment traversal is completed, all Live Segments and Dead Segments are finally merged to obtain the Segment data established on the i-th day and the i+a-th day.

[0010] As a further limitation of the technical solution of the present invention, C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 The steps include: C i+1 Traverse and use the KD tree method to search for all candidate vortices matching the i + a day in a circle with the vortex center after particle advection as the center and a radius not less than the maximum moving distance of the vortex a day and generate a candidate vortex set E i+1 .

[0011] As a further limitation of the technical solution of the present invention, the KD tree method is used to screen the candidate vortices and generate the candidate vortex set E i+1 The steps also include: If C i+1 If the difference between the longitude of a vortex center and the 0° longitude is less than the set threshold, the search circle with the vortex center as the center and the maximum moving distance of the vortex a day as the radius exceeds the longitude range of 0°-360°, the area with a longitude range of (360-A)°-360° is copied and pasted to the corresponding position; similarly, if C i+1The difference between the longitude of a vortex center and the 360° longitude is less than the set threshold, so that the search circle with the vortex center as the center and a radius not less than the maximum moving distance of the vortex a day exceeds the longitude range of 0°-360°. The area with a longitude range of 0°-A° is copied and pasted to the corresponding position. Finally, a KD tree search is performed with the vortex center as the center to find the corresponding candidate vortex.

[0012] As a further limitation of the technical solution of the present invention, the steps of establishing a linear structure Branch formed by connecting segments in multiple survival states and outputting a Branch data set include: Receive all the Segment data sets of the i-th day and the i+a-th day as input; Traverse all Live Segments and find the overlapping relationship between vortex index and time in Live Segments; All Live Segments with overlapping relationships are merged into Branch, and vortices and potential vortices are marked. The entire track of potential vortices in the linear structure Branch is deleted, and finally the Branch data set is output.

[0013] In a second aspect, the technical solution of the present invention provides a global ocean mesoscale eddy objective tracking device, including a data acquisition module, a mesoscale eddy data identification module, a particle advection processing module, a relationship establishment module, a linear structure establishment module and a tracking result output module; Data acquisition module, used to obtain global absolute dynamic terrain data; A mesoscale vortex data identification module is used to process the global absolute dynamic terrain data to identify the global mesoscale vortex data and potential vortex data to form global mesoscale vortex identification data; The particle advection processing module is used to perform particle advection on the centroids of all vortices and potential vortices on the i-th day based on the global mesoscale eddy identification data, set the integration time step to a days, and obtain the centroids of all vortices and potential vortices on the i-th day to be advected to the longitude and latitude positions of the i+a-th day; Relationship building module, used to establish the relationship Segment between two consecutive time step vortices; The linear structure building module is used to build a linear structure Branch formed by connecting segments in multiple survival states and output the Branch data set; The tracking result output module is used to store the output Branch dataset in a file in pkl format; each file consists of all vortex tracks of that year.

[0014] As a further limitation of the technical solution of the present invention, the mesoscale vortex data identification module includes a filtering processing unit, a vortex data identification processing unit, a traversal identification unit and a vortex identification data generation unit; A filtering processing unit, used for high-pass filtering the global absolute dynamic terrain data; A vortex data identification and processing unit, used to generate ADT contour lines of closed global vortex data based on the filtered data, and identify local extreme points as seed points of the vortex; The traversal identification unit is used to traverse the ADT contour lines of the global vortex data to determine whether each contour line meets the screening conditions of the vortex boundary; if it meets the conditions, the vortex is identified; if it does not meet the conditions, the potential vortex is identified; The vortex identification data generation unit is used to merge the identified vortex with the potential vortex data. After completing the contour traversal of the global vortex data, it saves and outputs the results to form the global mesoscale vortex identification data.

[0015] As a further limitation of the technical solution of the present invention, when the traversal identification unit traverses the ADT contour lines of the global vortex data, the contour lines that only contain a seed point of a vortex but do not meet the vortex boundary screening conditions are identified as potential vortices.

[0016] As a further limitation of the technical solution of the present invention, the relationship building module is specifically used to receive the input global vortex centroid set C of the i-th day i , and conduct particle advection with a time step of a day to obtain the vortex center set C advected to the i + a day i+1 ; for C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 ; Judgement in E i+1 Is there C in i+1 The vortex center in falls into the vortex within the effective boundary. If there is one, a Live Segment is constructed, otherwise a Dead Segment is constructed. i+1 After the traversal is completed, all Live Segments are traversed again to determine whether there is C i+1 In the case where multiple vortex centers fall within the same vortex boundary, if so, the distances between the multiple vortex centers and the vortex centroid are calculated, and the one with the smallest distance is selected for matching; when the Live Segment traversal is completed, all LiveSegments and Dead Segments are finally merged to obtain the Segment data established on the i-th day and the i+a-th day.

[0017] As a further limitation of the technical solution of the present invention, the relationship establishment module is used to establish the relationship between Ci+1 Traverse and use the K-Dtree method to search for all candidate vortices matching the i + a day in a circle with the vortex center after particle advection as the center and a radius not less than the maximum moving distance of the vortex a day and generate a candidate vortex set E i+1 ; If C i+1 The difference between the longitude of a vortex center and the 0° longitude is less than the set threshold, so that the search circle with the vortex center as the center and the maximum moving distance of the vortex a day as the radius exceeds the longitude range of 0°-360°. At this time, the area with a longitude range of (360-A)°-360° is copied and pasted to the corresponding position; if C i+1 The difference between the longitude of a vortex center and the 360° longitude is less than the set threshold, so that the search circle with the vortex center as the center and a radius not less than the maximum moving distance of the vortex a day exceeds the longitude range of 0°-360°. The area with a longitude range of 0°-A° is copied and pasted to the corresponding position. Finally, a KD tree search is performed with the vortex center as the center to find the corresponding candidate vortex.

[0018] As a further limitation of the technical solution of the present invention, the linear structure establishment module is specifically used to receive all input Segment data sets of the i-th day and the i+a-th day; traverse all Live Segments to find the overlapping relationship between the vortex index and time in the Live Segments; merge all Live Segments with overlapping relationships into Branches, mark the vortices and potential vortices, delete a whole track of potential vortices in the linear structure Branch, and finally output the Branch data set.

[0019] In a third aspect, the technical solution of the present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can execute the objective tracking method for global ocean mesoscale eddies as described in the first aspect.

[0020] In a fourth aspect, the technical solution of the present invention also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable the computer to execute the objective tracking method of global ocean mesoscale vortices as described in the first aspect.

[0021] It can be seen from the above technical solutions that the present invention has the following advantages: the particle advection in the Lagrangian method is introduced, and there is no need to set subjective parameters and "pseudo-vortices", so that vortices at different times are linked in a more objective way, and finally the objective tracking of global mesoscale vortices is achieved. This makes up for the deficiency of the original tracking algorithm that needs to set subjective parameters and "pseudo-vortices", and achieves the objective tracking of global mesoscale vortices and forms a data set.

[0022] In addition, the invention has a reliable design principle, a simple structure and a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention.

[0025] Figure 2 Schematic diagram of the movement of a particle point under the action of the velocity field.

[0026] Figure 3 Schematic diagram of two cases of particle advection.

[0027] Figure 4 Schematic diagram of screening candidate vortices across the 0° meridian.

[0028] Figure 5 A flow chart is established for the particle advection and Segment of the present invention.

[0029] Figure 6 It is a schematic diagram of Branch connection of the present invention.

[0030] Figure 7 This is a diagram for validating the method based on the measured data of the Drifter buoy.

[0031] Figure 8 is a schematic block diagram of an apparatus according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] like Figure 1 As shown, an embodiment of the present invention provides an objective tracking method for global ocean mesoscale eddies, comprising the following steps: Step 1: Obtain global absolute dynamic terrain data; The present invention uses satellite altimeter data to objectively track global mesoscale vortices. The data is fused with Absolute Dynamic Topography (ADT) data and geostrophic flow velocity data obtained by multiple altimetry satellites. The satellite altimeter delay data used in the present invention integrates multiple satellites, mainly including Jason-3, Sentinel-3A, HY-2A, SARAL / Altika, Cryosat-2, Jason-2, Jason-1, T / P, ENVISAT, GFO and ERS1 / 2, etc. These data use the WGS-84 coordinate system, with a time resolution of 1 day and a spatial resolution of 0.25° × 0.25°. The time range is from January 1, 1993 to June 23, 2022, and the spatial range is global.

[0034] Step 2: Process the global absolute dynamic terrain data to identify the global mesoscale eddy data and potential eddy data to form global mesoscale eddy identification data; This step specifically includes: Step 21: Perform high-pass filtering on the global absolute dynamic terrain data to remove large-scale signals in the ocean. Those skilled in the art know that mesoscale vortices are a dynamic phenomenon that is widely present in the global ocean, and its spatial scale is about tens to hundreds of kilometers, while large-scale vortices are closely related to the large-scale dynamic process of ocean circulation, and its spatial scale can reach hundreds of kilometers or more.

[0035] Step 22: Generate ADT contour lines of closed global vortex data based on the filtered data, and identify local extreme points as seed points of the vortex; Step 23: Traverse the ADT contour lines of the global vortex data to determine whether each contour line meets the screening conditions of the vortex boundary; if it meets the conditions, identify the vortex; if it does not meet the conditions, identify the potential vortex; Here, the screening conditions of vortex boundaries (such as amplitude, area, etc.) are used. For the first time, the concept of potential vortex is proposed in this paper, which is defined as a "vortex" that contains only one seed point but does not meet the screening conditions of vortex boundaries, and its size, amplitude or shape test error does not meet the requirements of vortex identification. We do not regard potential vortices as real vortices, but include them in the vortex identification dataset in order to pursue the accuracy of subsequent vortex objective tracking results.

[0036] It should be noted that in the existing vortex simple linear tracking algorithm, in order to reduce the impact of data noise on the vortex tracking results, the vortex is often allowed to disappear for a few days, and the gap is filled by interpolation or directly using the state before the vortex disappears. The filled vortex is called a "pseudo-vortex". The "pseudo-vortex" is only used for recording and is not included in any analysis. In other words, the difference between a potential vortex and a pseudo-vortex is that a potential vortex is real, while a pseudo-vortex is artificially fictitious.

[0037] Step 24: Merge the identified vortex with the potential vortex data. After completing the contour traversal of the global vortex data, save and output the results to form the global mesoscale vortex identification data.

[0038] Step 3: Based on the global mesoscale eddy identification data, the centroids of all eddies and potential eddies on the i-th day are advected to the longitude and latitude positions of the i+a-th day. The integration time step is set to a day. It should be noted that in this step, within the framework of Lagrangian vortex theory, particle advection describes the movement of microscopic particles in a fluid, and explains how particles rotate, diffuse, and mix with the movement of the fluid. The trajectory of particles in the flow field can be obtained by solving the partial differential equation of the flow field:

[0039] Trajectory field of particle points Flow Map is defined as:

[0040] Reflects the particles in the flow field from Location at the moment ,go through The advection of time, arriving New location of time The movement trajectory, such as Figure 2 shown.

[0041] In the embodiment of the present invention, the centroids of all vortices and "potential vortices" on the i-th day are advected by particles, and the integration time step a is set, where a is 1 day. In this way, the centroids of all vortices and potential vortices on the i-th day can be advected to the longitude and latitude positions of the i+1-th day, such as Figure 3 The Lagrangian particle advection method can simulate the movement of particles in water more accurately, and is of great significance in studying ocean vortices, but the amount of calculation is large. Therefore, the present invention greatly reduces the amount of calculation by advecting only one particle per vortex or potential vortex.

[0042] Step 4: Establish the relationship Segment between the vortices of two consecutive time steps; The Segment of the present invention includes two relationships: survival and death, which are Live Segment and Dead Segment respectively. The process of establishing the relationship between two consecutive time step vortices Segment is as follows: Figure 5 As shown; Step 1: Input the global vortex centroid set C on day i i , perform particle advection with a time step of 1 day, and obtain the vortex center set C advected to the i + 1th day i+1 .

[0043] Step 2: C i+1 Traverse and use the KD tree method to search for all possible matching candidate vortex sets E on the i + 1 day within a circle with a radius of 0.5° (not less than the maximum moving distance of the vortex in one day) and the vortex center after particle advection. i+1 When using the KD tree method to screen candidate vortices, it is important to note that if C i+1 The longitude of a vortex center in is very close to the 0° longitude, so that the search circle with the vortex center as the center and a radius of 0.5° exceeds the longitude range of 0°-360°. Then we copy and paste the area with a longitude range of 355°-360° to the corresponding position. Similarly, if C i+1 The longitude of a vortex center in is very close to the 360° longitude, so that the search circle with the vortex center as the center and a radius of 0.5° exceeds the longitude range of 0°-360°. Then we copy and paste the area with a longitude range of 0°-5° to the corresponding position, and finally perform a KD tree search with the vortex center as the center to find its corresponding candidate vortex, such as Figure 4 shown.

[0044] Step 3: Determine the i+1 Is there C in i+1The vortex center in falls into the vortex within the effective boundary. If there is one, a Live Segment is constructed, otherwise a Dead Segment is constructed; Step 4: C i+1 After the traversal is completed, all Live Segments are traversed again to determine whether there is C i+1 If multiple vortex centers in the same vortex fall within the boundary of the same vortex, the distances between these vortex centers and the centroid of this vortex are calculated, and only the one with the smallest distance is matched, such as Figure 3 shown.

[0045] Step 5: When the Live Segment traversal is completed, all Live Segments and Dead Segments are finally merged to obtain the Segment data established on the i-th day and the i + 1-th day.

[0046] Step 5: Establish a linear structure Branch formed by connecting segments of multiple survival states and output the Branch data set; 5. Establish a linear structure Branch formed by connecting multiple live segments: It should be noted that a branch is a track formed by connecting multiple live segments that have overlapping relationships in time and vortex index in time order. The method for establishing a branch is: first input all the segment data sets of the i-th day and the i+1-th day, then traverse all the live segments, find their overlapping relationships in vortex index and time, and then merge all the live segments with this overlapping relationship into a branch, such as Figure 6 As shown, mark the vortex and "potential vortex". If it is a vortex, mark it as "0", if it is a "potential vortex", mark it as "-1", delete the entire track that is all "potential vortex", and finally output the Branch data set.

[0047] Step 6: Store the output Branch dataset in pkl format files in units of years; each file consists of all vortex tracks in that year. Each point in each track contains six attribute information, as follows: Date of vortex location: the date on which the point on the record track is the vortex center of mass; Vortex index: the serial number of the point on the recorded trajectory in the vortex centroid array of the day; Longitude of the vortex centroid: Longitude of the point on the recorded trajectory; Latitude of the vortex centroid: the latitude of the point on the recorded trajectory; Vortex type (distinguishing between cold and warm vortices): distinguish whether the point on the trajectory belongs to the centroid of a cyclonic vortex or the centroid of an anticyclonic vortex. The anticyclone is 1 and the cyclone is -1. The type of all points on a trajectory must be the same. Vortex type (distinguishing between vortex and potential vortex): Distinguish whether the point on the trajectory belongs to the vortex center of mass or the potential vortex center of mass. The vortex is 0 and the potential vortex is -1. A trajectory may be entirely composed of vortex center of mass, or may have both vortex center of mass and potential vortex center of mass. However, the present invention does not consider the case where a trajectory is entirely composed of potential vortex center of mass, and all such cases are deleted.

[0048] The Drifter buoy in the ocean will move accordingly with the vortex movement, and the Drifter buoy is usually equipped with a positioning system to record its real-time position, so the Drifter buoy can record the survival and movement of the vortex. If the Drifter buoy is captured by the vortex, it will rotate with the vortex. In the northern hemisphere, if the buoy is captured by an anticyclonic vortex, its trajectory will be a clockwise circle, and if it is captured by a cyclonic vortex, it will be a counterclockwise circle. In the southern hemisphere, it is exactly the opposite. The present invention uses the measured data of the Drifter buoy to match and verify the results of mesoscale vortex tracking trajectories in the Pacific Ocean, Indian Ocean, and Atlantic Ocean, respectively. Figure 7 As shown in Figure 1, the blue line is the cyclonic vortex track, the red line is the anticyclonic vortex track, and the green line is the Drifter buoy track, and the matching effect is good. Among them, the mesoscale vortex tracking track results in the Pacific are shown in Figure 1. Figure 7 As shown in (a) in the figure, the mesoscale eddy tracking results in the Indian Ocean are as follows Figure 7 As shown in (b) in the figure, the tracking results of the mesoscale eddy in the Atlantic are as follows Figure 7 As shown in (c) in .

[0049] like Figure 8 As shown, an embodiment of the present invention provides a global ocean mesoscale eddy objective tracking device, including a data acquisition module, a mesoscale eddy data identification module, a particle advection processing module, a relationship establishment module, a linear structure establishment module and a tracking result output module; Data acquisition module, used to obtain global absolute dynamic terrain data; A mesoscale vortex data identification module is used to process the global absolute dynamic terrain data to identify the global mesoscale vortex data and potential vortex data to form global mesoscale vortex identification data; The particle advection processing module is used to perform particle advection on the centroids of all vortices and potential vortices on the i-th day based on the global mesoscale eddy identification data, set the integration time step to a days, and obtain the centroids of all vortices and potential vortices on the i-th day to be advected to the longitude and latitude positions of the i+a-th day; Relationship building module, used to establish the relationship Segment between two consecutive time step vortices; The linear structure building module is used to build a linear structure Branch formed by connecting segments in multiple survival states and output the Branch data set; The tracking result output module is used to store the output Branch dataset in pkl format files in units of years; each file consists of all vortex tracks in that year.

[0050] In some embodiments, the mesoscale vortex data identification module includes a filtering processing unit, a vortex data identification processing unit, a traversal identification unit, and a vortex identification data generation unit; A filtering processing unit, used for high-pass filtering the global absolute dynamic terrain data; A vortex data identification and processing unit, used to generate ADT contour lines of closed global vortex data based on the filtered data, and identify local extreme points as seed points of the vortex; The traversal identification unit is used to traverse the ADT contour lines of the global vortex data to determine whether each contour line meets the screening conditions of the vortex boundary; if it meets the conditions, the vortex is identified; if it does not meet the conditions, the potential vortex is identified; The vortex identification data generation unit is used to merge the identified vortex with the potential vortex data. After completing the contour traversal of the global vortex data, it saves and outputs the results to form the global mesoscale vortex identification data.

[0051] In some embodiments, when the traversal identification unit traverses the ADT contour lines of the global vortex data, the contour lines that only contain a seed point of a vortex but do not meet the vortex boundary screening conditions are identified as potential vortices.

[0052] In some embodiments, the relationship building module is specifically configured to receive the input global vortex centroid set C of the i-th day. i , and conduct particle advection with a time step of a day to obtain the vortex center set C advected to the i + a day i+1 ; for C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 ; Judgement in E i+1 Is there C in i+1 The vortex center in falls into the vortex within the effective boundary. If there is one, a Live Segment is constructed, otherwise a Dead Segment is constructed. i+1 After the traversal is completed, all Live Segments are traversed again to determine whether there is C i+1If multiple vortex centers fall within the same vortex boundary, if so, the distances between the multiple vortex centers and the vortex centroid are calculated, and the one with the smallest distance is selected for matching; when the Live Segment traversal is completed, all Live Segments and Dead Segments are finally merged to obtain the Segment data established on the i-th day and the i+a-th day. It should be noted that for C i+1 Traverse and use the KD tree method to search for all candidate vortices matching the i + a day in a circle with the vortex center after particle advection as the center and a radius not less than the maximum moving distance of the vortex a day and generate a candidate vortex set E i+1 ; If C i+1 The difference between the longitude of a vortex center and the 0° longitude is less than the set threshold, so that the search circle with the vortex center as the center and the maximum moving distance of the vortex a day as the radius exceeds the longitude range of 0°-360°. At this time, the area with a longitude range of (360-A)°-360° is copied and pasted to the corresponding position; if C i+1 The difference between the longitude of a vortex center and the 360° longitude is less than the set threshold, so that the search circle with the vortex center as the center and the maximum moving distance of the vortex a day as the radius exceeds the longitude range of 0°-360°. The area with a longitude range of 0°-A° is copied and pasted to the corresponding position, and finally the KD tree search is performed with the vortex center as the center to find the corresponding candidate vortex. Here A is 5.

[0053] In some embodiments, the linear structure establishment module is specifically used to receive input of all Segment data sets of the i-th day and the i+a-th day; traverse all Live Segments to find the overlapping relationship between the vortex index and time in the Live Segments; merge all Live Segments with overlapping relationships into Branches, mark the vortices and potential vortices, delete an entire track of potential vortices in the linear structure Branch, and finally output the Branch data set.

[0054] The embodiment of the present invention also provides an electronic device, which includes: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The communication bus can be used for information transmission between the electronic device and the sensor. The processor can call the logic instructions in the memory to execute the following method: obtain global absolute dynamic terrain data; process the global absolute dynamic terrain data to identify global mesoscale vortex data and potential vortex data to form global mesoscale vortex identification data; based on the global mesoscale vortex identification data, the center of mass of all vortices and potential vortices on the i-th day is advected by particles, and the integral time step is set to a day, and the center of mass of all vortices and potential vortices on the i-th day is advected to the longitude and latitude position of the i+a day; establish the relationship Segment between two consecutive time step vortices; establish a linear structure Branch formed by connecting multiple segments of survival states and output the Branch data set; store the output Branch data set in a file in pkl format in units of years; wherein each file is composed of all vortex trajectories of that year.

[0055] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0056] An embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute the method provided by the above method embodiment, for example, including: obtaining global absolute dynamic terrain data; processing the global absolute dynamic terrain data to identify global mesoscale vortex data and potential vortex data to form global mesoscale vortex identification data; based on the global mesoscale vortex identification data, advecting the center of mass of all vortices and potential vortices on the i-th day, setting the integration time step to a days, and obtaining the center of mass of all vortices and potential vortices on the i-th day advected to the longitude and latitude position of the i+a-th day; establishing a relationship Segment between two consecutive time step vortices; establishing a linear structure Branch formed by connecting Segments in multiple survival states and outputting a Branch data set; storing the output Branch data set in a file in pkl format in units of years; wherein each file is composed of all vortex trajectories of that year.

[0057] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0058] An embodiment of an objective tracking device for mesoscale eddies in the global ocean is provided in an embodiment of the present invention. The device and the objective tracking method for mesoscale eddies in the global ocean described in the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the objective tracking device for mesoscale eddies in the global ocean, reference can be made to the embodiment of the objective tracking method for mesoscale eddies in the global ocean described in the above-mentioned embodiments.

[0059] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention.

Claims

1. A global ocean mesoscale eddy objective tracking method, characterized in that: The steps include: Obtain absolute dynamic terrain data around the world; Process the global absolute dynamic terrain data to identify the global mesoscale eddy data and potential eddy data to form the global mesoscale eddy identification data; Based on the global mesoscale eddy identification data, the centroids of all eddies and potential eddies on the i-th day are advected to the longitude and latitude positions of the i+a-th day. The integration time step is set to a day. Establish the relationship between two consecutive time step vortices Segment; Establish a linear structure Branch formed by connecting segments in multiple survival states and output the Branch data set; The output Branch dataset is stored in pkl format files, where each file consists of all eddy tracks for that year.

2. The objective tracking method of global ocean mesoscale eddies according to claim 1, characterized in that: The steps of processing the global absolute dynamic terrain data to identify the global mesoscale eddy data and potential eddy data to form the global mesoscale eddy identification data include: Perform high-pass filtering on the global absolute dynamic terrain data; Generate closed ADT contours of global vortex data based on filtered data, and identify local extreme points as vortex seed points; Traverse the ADT contour lines of global vortex data to determine whether each contour line meets the screening conditions of the vortex boundary; if it meets the conditions, identify the vortex; if it does not meet the conditions, identify the potential vortex; The identified vortex is merged with the potential vortex data, and after completing the contour traversal of the global vortex data, the results are saved and output to form the global mesoscale vortex identification data.

3. The objective tracking method of global ocean mesoscale eddies according to claim 2, characterized in that: The steps for identifying potential vortices include: When traversing the ADT contour lines of the global vortex data, the contour lines that contain only one vortex seed point but do not meet the vortex boundary screening conditions are identified as potential vortices.

4. The objective tracking method of global ocean mesoscale eddies according to claim 3, characterized in that: The steps to establish the relationship Segment between two consecutive time step vortices include: Receive the input global vortex centroid set C on the i-th day i , and conduct particle advection with a time step of a day to obtain the vortex center set C advected to the i + a day i+1 ; C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 ; Judgement in E i+1 Is there C in i+1 The vortex center in falls into the vortex within the effective boundary. If there is one, a Live Segment is constructed, otherwise a Dead Segment is constructed; C i+1 After the traversal is completed, all Live Segments are traversed again to determine whether there is C i+1 If multiple vortex centers fall within the same vortex boundary, if so, calculate the distances between the multiple vortex centers and the vortex centroid, and select the one with the smallest distance for matching; When the Live Segment traversal is completed, all Live Segments and Dead Segments are finally merged to obtain the Segment data established on the i-th day and the i+a-th day.

5. The objective tracking method of global ocean mesoscale eddies according to claim 4, characterized in that: C i+1 Traverse and use the KD tree method to screen candidate vortices and generate a candidate vortex set E i+1 The steps include: C i+1 Traverse and use the KD tree method to search for all candidate vortices matching the i + a day in a circle with the vortex center after particle advection as the center and a radius not less than the maximum moving distance of the vortex a day and generate a candidate vortex set E i+1 .

6. The objective tracking method of global ocean mesoscale eddies according to claim 5, characterized in that: Use K-Dtree method to screen candidate vortices and generate candidate vortex set E i+1 The steps also include: If C i+1 If the difference between the longitude of a vortex center and the 0° longitude is less than the set threshold, the search circle with the vortex center as the center and the maximum moving distance of the vortex a day as the radius exceeds the longitude range of 0°-360°, the area with a longitude range of (360-A)°-360° is copied and pasted to the corresponding position; similarly, if C i+1 The difference between the longitude of a vortex center and the 360° longitude is less than the set threshold, so that the search circle with the vortex center as the center and a radius not less than the maximum moving distance of the vortex a day exceeds the longitude range of 0°-360°. The area with a longitude range of 0°-A° is copied and pasted to the corresponding position. Finally, a KD tree search is performed with the vortex center as the center to find the corresponding candidate vortex.

7. The objective tracking method of global ocean mesoscale eddies according to claim 5, characterized in that: The steps of establishing a linear structure Branch formed by connecting segments in multiple survival states and outputting the Branch data set include: Receive all the Segment data sets of the i-th day and the i+a-th day as input; Traverse all Live Segments and find the overlapping relationship between vortex index and time in Live Segments; All Live Segments with overlapping relationships are merged into Branch, and vortices and potential vortices are marked. The entire track of potential vortices in the linear structure Branch is deleted, and finally the Branch data set is output.

8. An objective tracking device for global ocean mesoscale eddies, characterized in that: It includes a data acquisition module, a mesoscale vortex data identification module, a particle advection processing module, a relationship establishment module, a linear structure establishment module and a tracking result output module; Data acquisition module, used to obtain global absolute dynamic terrain data; A mesoscale vortex data identification module is used to process the global absolute dynamic terrain data to identify the global mesoscale vortex data and potential vortex data to form global mesoscale vortex identification data; The particle advection processing module is used to perform particle advection on the centroids of all vortices and potential vortices on the i-th day based on the global mesoscale eddy identification data, set the integration time step to a days, and obtain the centroids of all vortices and potential vortices on the i-th day to be advected to the longitude and latitude positions of the i+a-th day; Relationship building module, used to establish the relationship Segment between two consecutive time step vortices; The linear structure building module is used to build a linear structure Branch formed by connecting segments in multiple survival states and output the Branch data set; The tracking result output module is used to store the output Branch dataset in a file in pkl format; each file consists of all vortex tracks of that year.

9. An electronic device, characterized in that: The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; the memory stores computer program instructions executable by the at least one processor, and the computer program instructions are executed by the at least one processor to enable the at least one processor to execute the global ocean mesoscale eddy objective tracking method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the global ocean mesoscale eddy objective tracking method as described in any one of claims 1 to 7.

Images