A small and medium scale vortex feature identification method fusing active and passive remote sensing data

Abstract

The application discloses a small and medium-sized vortex feature identification method fusing active and passive remote sensing data, and the identification method comprises the following steps: obtaining a vortex tracking data set by using sea surface height anomaly data, obtaining vortex date, vortex center point longitude and latitude, vortex radius and vortex duration; obtaining chlorophyll concentration data in the vortex range by using a stationary water color remote sensing image; obtaining surface vortex flow vectors by using a maximum correlation coefficient flow field inversion algorithm and taking the chlorophyll concentration as a tracer; obtaining vortex surface temperature by using a sea surface temperature product; establishing a time data set of vortex center point longitude and latitude coordinates, vortex radius, vortex duration, chlorophyll concentration, sea surface flow field vectors and sea surface temperature; and obtaining key feature parameters of the changes of surface vortex flow, morphological structure, inner and outer chlorophyll concentrations and surface temperature of the vortex in a life cycle. The application fuses active and passive remote sensing data, and can accurately capture the changes of marine biochemical and physical parameters and the changes in a day in the life cycle of the vortex.

Description

Technical Field

[0001] This application relates to the field of marine remote sensing, specifically to a method for identifying small- and medium-scale eddy features by fusing active and passive remote sensing data. Background Technology

[0002] Mesoscale eddies in the ocean refer to rotating fluid structures that last from days to months and span tens to hundreds of kilometers in space. These eddies store energy an order of magnitude or more higher than the surrounding average flow velocity, making them an indispensable and crucial component of dynamic oceanography research. Mesoscale eddies possess abundant kinetic energy during their motion and play a vital role in momentum transport, chemical transport, heat and mass transfer in the upper ocean. They significantly influence oceanic circulation structures, large-scale water mass distribution, marine life, and global climate change triggered by air-sea interactions. To date, scientists have been studying the interactions between oceanic mesoscale phenomena and marine biochemical processes in the shallow shelf seas and open oceans worldwide, achieving significant breakthroughs. This research has important scientific significance and practical application value.

[0003] Overall, ocean eddies are numerous, widely distributed, energy-rich, and highly mobile, making them ideal for studying the material cycle, energy transfer, and interactions between different spheres in the ocean. With the realization of full life-cycle tracking and observation of eddies, remote sensing technology in this field has achieved significant breakthroughs since the 21st century, triggering a new wave of research in eddies. Based on the basic characteristics of ocean eddies, previous researchers have proposed four remote sensing principles for their effective identification and tracking: (1) temperature anomalies; (2) material tracing; (3) closed topology; and (4) rotating flow fields. First, warm and cold eddies exist in the ocean, causing positive and negative anomalies in seawater temperature at their centers, respectively. Ideally, these anomalies can directly affect the ocean surface, forming synchronous anomalies in sea surface temperature (SST). These anomalies can be detected using infrared sensors carried by satellites. Second, eddies have a strong capacity for material entrainment and transport, carrying physical, biological, and chemical substances such as chlorophyll, organic carbon, plankton, and even sea ice along with their rotation and migration. The movement of these substances can be detected by spectrometers with underwater penetration capabilities and synthetic aperture radar (SAR) with sea surface imaging capabilities. Third, the geostrophic effect of vortex circulation results in a relatively stable closed structure of the sea surface topology. In this structure, the center of a warm vortex corresponds to the maximum value of the sea surface height (SSH), while the center of a cold vortex corresponds to the minimum value. Therefore, radar altimeters can be used to identify and track these closed structures. Fourth, the flow of seawater within a vortex can form clockwise or counterclockwise rotating flow fields, while also causing divergence and convergence of surface seawater, forming upwelling and downwelling currents. This flow alters the roughness of the sea surface and the sea surface height (SSH), which can then be captured by sensors such as SAR and radar altimeters. Based on these principles, over the past half-century, researchers have developed a series of techniques and methods for remote sensing ocean vortices using various visible light, infrared, and microwave sensors, greatly advancing the development of marine science and even Earth system science.

[0004] Eddy ocean color remote sensing based on the principle of material tracing began in the 1980s. Its development has roughly gone through an early exploratory stage (before the 21st century) based on CZCS satellite data, primarily for eddy identification applications, followed by a widespread application stage utilizing fused data from newer remote sensors such as SeaWiFS, MODIS, and MERIS to conduct statistical and ecological studies of eddy regional and global characteristics. GOCI (Geostationary Ocean Color Imager), with its high temporal resolution (one hour) and spatial resolution (500m), can achieve continuous hourly observations of the Bohai Sea, Yellow Sea, East China Sea, Sea of ​​Japan, and Northwest Pacific Ocean in China (Ryu JH, Han HJ, Cho S, et al. Overview of geostationary ocean color imager (GOCI) and GOCI data processing system (GDPS) [J]. Ocean Science Journal, 2012, 47(3): 223-233.). Compared to polar-orbiting satellites, GOCI's unprecedented high spatiotemporal observation results have greatly enhanced our ability to monitor highly dynamic changes in the ocean environment and provided new data support for observing the intraday variation characteristics of eddies. The Northeast Asian sea area covered by GOCI is one of the most active mesoscale eddies and a sensitive region for global atmospheric and oceanic changes. China, located on the western coast of the North Pacific, is directly affected by changes in tropical cyclones and ocean currents in the Northwest Pacific. Exploring the impact of eddy characteristics in this region on the ecological environment has significant scientific and applied value. However, this technology still faces the problem of difficulty in labeling the same eddy due to cloud cover. Therefore, it is essential to conduct feature identification of small- and medium-scale eddies throughout their entire life cycle in the Northeast Asian sea area, integrating multiple technologies. Summary of the Invention

[0005] To address the problem that existing technologies use relatively independent multi-source remote sensing data without fully integrating them, this application proposes a small-to-medium scale vortex feature recognition method. This method overcomes the shortcomings of water color remote sensing, which cannot label the same vortex due to cloud cover, and microwave remote sensing, which cannot obtain vortex feature parameters.

[0006] The objective of this application is achieved through the following technical solution:

[0007] A method for identifying small-to-medium scale vortex features by fusing active and passive remote sensing data, characterized by the following steps:

[0008] Step S1: Use sea surface height anomaly data to obtain a tracking dataset of small and medium-scale eddies, and calculate the eddy's life cycle (duration), center point latitude and longitude coordinates, and radius;

[0009] Step S2: Obtain chlorophyll concentration data within the eddy range using static water color remote sensing satellite imagery through a matching algorithm;

[0010] Step S3: Using the maximum correlation coefficient flow field inversion algorithm, surface eddy vectors are obtained with chlorophyll concentration as a tracer;

[0011] Step S4: Obtain the surface temperature of small- and medium-scale eddies using sea surface temperature product data;

[0012] Step S5: Combining the data obtained from steps S1 to S4, establish a time dataset of the latitude and longitude coordinates of the vortex center point, vortex radius, vortex duration, chlorophyll concentration, surface vortex vector, and vortex surface temperature within the vortex life cycle.

[0013] Step S6: Select key time point data with equal time intervals from Step S5 to construct key characteristic parameters of surface eddy changes, morphological structure changes, chlorophyll concentration changes, and surface temperature changes of small-scale eddies throughout their entire life cycle, so as to facilitate their use, understanding, and illustration.

[0014] This application innovatively utilizes the aforementioned small- and medium-scale vortex feature identification method to obtain the vortex's life cycle, center point latitude and longitude coordinates, and radius using SLA data; it also utilizes GOCI chlorophyll data to obtain the corresponding sea surface current field and sea surface temperature data for the vortex; and then integrates the SLA data and GOCI chlorophyll data to obtain the evolution process of the vortex throughout its entire life cycle (covering changes in sea surface eddies, morphological structure, chlorophyll distribution and concentration, and sea surface temperature), helping relevant personnel and managers understand, utilize, and analyze the impact of vortices on marine engineering, marine fisheries, and the marine ecological environment.

[0015] Step S1 as described in this application includes:

[0016] S11: Define the vortex centroid at time T0 as the vortex center point; set the number of consecutive times no vortex is detected, T1, to 0; T = T 0；

[0017] S12: Taking the vortex center as the origin, the larger of the following values ​​is taken westward: 1.75 times the 7-day ocean propagation distance of the baroclinic Rossby wave phase velocity and 150 km. The distance is taken as 150 km in the east, south, and north directions. Within this range, the vortex at time T = T+1 with the smallest similarity value and satisfying a similarity value less than 0.5 is selected. The similarity calculation formula is:

[0018]

[0019] In the formula, ΔS, ΔL s ,Δξ,ΔE k S0, L represent the vortex spatial distance, radius difference, relative vorticity difference, and EKE difference between two adjacent moments, respectively; s0 ,ξ0,E k0 These are their respective characteristic values, at 100km, 500km, and 10km respectively. -6 s -1 and 100cm 2 s -2 ;

[0020] S13: If a vortex of the same origin can be detected at time T, it is considered that the vortex continues to exist, and the number of consecutive vortexes not detected T1 is reset to zero before proceeding to step S12; if it cannot be detected, it is recorded as the vortex being missing once, and the number of consecutive vortexes not detected T1 is incremented by 1.

[0021] S14: Is the number of consecutive times T1 that no vortex is detected greater than the vortex disappearance confirmation setting value? If not, proceed to step S12. If yes, the life cycle of the vortex ends at the time T - the vortex disappearance confirmation setting value (i.e., regress from the current time to the vortex disappearance confirmation setting value time).

[0022] The radius of the chlorophyll extracted in step S2 of this application is equal to the radius of the vortex.

[0023] The time resolution of chlorophyll concentration and surface eddy vector in the time dataset obtained in step S5 of this application is in hours, covering the range of 8:30-15:30 Beijing time; the time resolution of vortex center point latitude and longitude coordinates, vortex radius, vortex life cycle and vortex surface temperature is in days.

[0024] In this application, the chlorophyll coverage area in the vortex accounts for more than 30% of the total vortex area.

[0025] The beneficial effects of this application are as follows:

[0026] (1) This application integrates active and passive remote sensing data and establishes a feature time dataset of the entire life cycle of vortices within the coverage area of ​​GOCI over 10 years through a series of post-processing operations such as identification and matching.

[0027] (2) This method overcomes the shortcomings of passive water color remote sensing, which cannot mark the same vortex due to the influence of cloud cover, and microwave remote sensing, which cannot obtain vortex characteristic parameters.

[0028] (3) The method of this application can accurately capture the changes in marine biochemical and physical parameters throughout the entire life cycle of small and medium-scale eddies, providing a more reliable foundation for subsequent research on marine engineering, marine fisheries, marine ecological environment, etc. Attached Figure Description

[0029] Figure 1 This is a flowchart of an embodiment of this application;

[0030] Figure 2 This is a flowchart illustrating the identification of vortices using SLA data, as described in an embodiment of this application.

[0031] Figure 3 This is a visualization of the characteristics of each moment in the vortex lifecycle according to an embodiment of this application. The vortex lifecycle is normalized, and the vortex characteristic distributions at times 0, 0.25, 0.5, 0.75, and 1 are selected. Time = 0 represents the moment the vortex is generated, and similarly, time = 1 represents the moment the vortex disappears. Figures (a)-(e) show the chlorophyll and surface eddy current distributions of the vortex at each moment, with the background field representing chlorophyll and the black vector representing surface eddy currents. Figures (f)-(j) show the SLA and morphological size distributions of the vortex at each moment, with the background field representing SLA and the black dashed line representing the fitted circle of the vortex radius. Figures (k)-(o) show the surface temperature distribution of the vortex at each moment.

[0032] Figure 4 This is a schematic diagram of the intraday variation of vortex chlorophyll concentration and surface vortex in an embodiment of this application; wherein, the intraday variation time range is 8:30-15:30 (Beijing time), the background field is chlorophyll, and the black vector is surface vortex. Detailed Implementation

[0033] The Northeast Asian seas covered by GOCI are among the most active mesoscale eddies and are also sensitive to global atmospheric and oceanic changes. China, located on the western coast of the North Pacific, is directly influenced by tropical cyclones and ocean currents in the Northwest Pacific. Exploring the impact of eddy characteristics in this region on the ecological environment is of significant scientific importance and practical value. Therefore, conducting feature identification of small- and medium-scale eddies throughout their entire lifecycle using multi-source remote sensing data in the Northeast Asian seas will facilitate the integration, verification, and correction of various technologies.

[0034] The method for identifying small-scale eddy features by fusing active and passive remote sensing data as described in this application includes the following steps:

[0035] Step S1: Use sea surface height anomaly data to obtain a tracking dataset of small and medium-scale eddies, and calculate the eddy's life cycle (duration), center point latitude and longitude coordinates, and radius;

[0036] Sea surface height anomaly (SLA) data is a fusion of TOPEX / Poseidon (T / P), Jason, and ERS1 / 2 altimeter data from 2011 to 2021 provided by AVISO (http: / / www.aviso.oceanobs.com / en / data.html). The fused data can better reflect the energy and structural characteristics of ocean mesoscale eddies. The eddy identification method refers to the discrimination criteria of Chelton et al. (2011) (Chelton DB, Schlax MG and Samelson RM. 2011. Global observations of nonlinear mesoscale eddies. Progress in Oceanography, 91(2):167-216), and the specific criteria are as follows:

[0037] (1) High-pass filtering is used to smooth the data in order to remove Rossby wave coherent signals;

[0038] (2) For vortices (which have two types: anti-vortices and air vortices, the same below), for anti-vortices, the sea level height (SSH) value of all pixels must be greater than a given threshold, which is the minimum value among all pixels; for air vortices, the sea level height (SSH) value of all pixels must be less than a given threshold, which is the maximum value among all pixels.

[0039] (3) The number of pixels in the connected region is greater than 8 and less than 1000;

[0040] (4) For a vortex, there is at least one local SLA extreme (maximum or minimum);

[0041] (5) The amplitude of the vortex is greater than or equal to 7.5 cm;

[0042] (6) The distance between any two points in a connected region is less than a given maximum value. It is specified that the maximum value is 1200 km at the equator and 400 km north of 25°N. Between the equator and 25°N, this value has a linear relationship with latitude.

[0043] A tracking algorithm combining the nearest distance method and the similarity method was used to track all mesoscale eddies at each time step (Qin Lijuan, Dong Qing, Fan Xing, Xue Cunjin, Hou Xueyan, Song Wanjiao. Spatiotemporal analysis of North Pacific mesoscale eddies from satellite altimeters [J]. Journal of Remote Sensing, 2015, 19(05):806-817.). The specific steps are as follows:

[0044] S11: Define the centroid of the geometry within the vortex boundary range (the outermost edge of the detected vortex range) at time T0 (the initial detection time, the initial value of detection time T) as the center point of the vortex; set the number of consecutive times no vortex is detected, T1, to 0.

[0045] S12: Taking the center point as the origin, take a distance westward that is 1.75 times the ocean propagation distance within 7 days (d) of the baroclinic Rossby wave phase velocity (or 150 km if less than 150 km), and take 150 km for the other three directions. Within this range, select the vortex at the next moment (T = T + 1) with the minimum similarity value and a similarity value less than 0.5. Similarity calculation formula:

[0046]

[0047] In the formula, ΔS, ΔL s ,Δξ,ΔE k S0, L represent the vortex spatial distance, radius difference, relative vorticity difference, and EKE difference between two adjacent moments, respectively; s0 ,ξ0,E k0 These are their respective characteristic values, at 100km, 500km, and 10km respectively. -6 s -1 and 100cm 2 s -2 ;

[0048] S13: If a vortex of the same origin can be detected at time T, it is considered that the vortex continues to exist, and the number of consecutive vortexes not detected T1 is reset to zero before proceeding to step S12; if it cannot be detected, it is recorded as the vortex being missing once, and the number of consecutive vortexes not detected T1 is incremented by 1.

[0049] S14: Is the number of consecutive times T1 that no vortex is detected greater than the vortex disappearance confirmation setting value? If not, proceed to step S12. If yes, the life cycle of the vortex ends at the time T - vortex disappearance confirmation setting value, and the vortex tracking is completed.

[0050] The "moment" mentioned in this application refers to a set detection period, which is usually one day or more, with one day as an exception. This data can be determined comprehensively based on needs, probability, and data volume. The vortex disappearance confirmation setting is usually set to 4, meaning that a vortex is considered to have disappeared only after it has been continuously missing for more than 5 moments.

[0051] Step S2: Using a matching algorithm, obtain chlorophyll concentration data within the eddy range from still water color remote sensing satellite images, where the chlorophyll coverage must reach 30% or more of the eddy area.

[0052] For the eddy tracking dataset obtained in step S1, chlorophyll concentration data within the eddy range are acquired using stationary water color remote sensing satellite imagery through a matching algorithm. Specifically, the eddy tracking dataset is categorized by eddy type and year; for the GOCI chlorophyll data, missing days and time periods are filled with zero-value data, and the dataset is also categorized by year to facilitate the sea surface current field operation of the matching program. The information after matching mainly includes the number of eddies that meet the matching criteria and the average chlorophyll concentration within the eddies.

[0053] Step S3: Using the maximum correlation coefficient flow field inversion algorithm, surface eddy vectors are obtained with chlorophyll concentration as a tracer;

[0054] The maximum correlation coefficient method, based on template matching, obtains the sea surface current field vector for a given period from two consecutive GOCI images taken on the same day. The first image used to estimate the current location is called the "template window," and the second image is called the "search window." The MCC algorithm uses correlation to track changes in tracer structure based on template matching. A suitable matching window is determined by calculating the maximum correlation coefficient between the template window and the search window. If the correlation coefficient between the template window and the search window is greater than a similarity threshold between them, the matching window is considered the correct location reached by the template window after moving for one hour. Then, the above steps are repeated to obtain a relatively complete sea surface current field.

[0055] Step S4: Obtain the surface temperature of small- and medium-scale eddies using sea surface temperature product data;

[0056] Sea surface temperature data comes from the OSTIA (Operational SST & Sea Ice Analysis) product (http: / / ghrsstppmetoffice.com / pages / latest_analysis / ostia.html), with daily data at a resolution of 0.05° × 0.05°. This product uses data from sensors such as AVHRR, AMSR, and AATSR, as well as measured data. To eliminate errors caused by daytime sea surface warming, the OSTIA product's basic temperature filter excludes observations with daytime wind speeds less than 6 m / s, and also references AATSR data and satellite-tracked buoy data to adjust for errors. Using this sea surface temperature product data, relatively accurate surface temperatures of small- to medium-scale eddies can be obtained.

[0057] Step S5: Combining the data obtained from steps S1 to S4, establish a time dataset of the latitude and longitude coordinates of the vortex center point, vortex radius, vortex duration, chlorophyll concentration, surface vortex vector, and vortex surface temperature within the vortex duration.

[0058] The established time dataset includes basic parameters identifying vortices, such as the latitude and longitude coordinates of the center point, the date, and the trajectory identification number; it also includes specific biochemical and physical parameters, such as chlorophyll concentration and radius. The time resolution for chlorophyll concentration and surface vortex vector is hourly, covering the period from 8:30 to 15:30 (Beijing time), while the time resolution for the other parameters is daily.

[0059] Step S6: Based on step S5, obtain key characteristic parameters such as surface eddy changes, morphological structure changes, chlorophyll concentration changes, and surface temperature changes of small- and medium-scale eddies throughout their entire life cycle. This step selects some representative time point data from step S5, such as eddy data at five key time points in the eddy life cycle: 0, 0.25, 0.5, 0.75, and 1. Alternatively, it can divide the eddy life cycle into several time points (three or more, including at least the three key time points of 0, 0.5, and 1) to facilitate a direct understanding of the eddy change process and to visualize the eddy.

[0060] Based on the established vortex time dataset, vortices can be selected according to their identification number and observation sequence number, thereby obtaining key characteristic parameters such as surface eddy changes, morphological structure changes, chlorophyll concentration changes, and surface temperature changes of small and medium-scale vortices throughout their entire life cycle.

[0061] The specific implementation steps of this application and the identification results of special cases are explained below with reference to the accompanying drawings.

[0062] Figure 1 This is a flowchart of the small- and medium-scale eddy feature recognition method described in this application. The flowchart includes three main parts of the process. The first part is to obtain a tracking dataset of small- and medium-scale eddies using sea surface height anomaly data. The second part is to obtain chlorophyll concentration data within the eddy range using static water color remote sensing satellite imagery through a matching algorithm, and then obtain the surface eddy vector using the maximum correlation coefficient algorithm. The third part is to combine the results of the first two parts with sea surface temperature data to establish a feature parameter dataset for the duration of the eddy, and then obtain the feature parameter changes of small- and medium-scale eddies throughout their entire life cycle.

[0063] Figure 2 This is a flowchart for identifying vortices using sea surface height anomaly data. By preprocessing the sea surface height anomaly data with high-pass filtering to remove Rossby wave coherent signals, and then based on the vortex discrimination criteria of Chelton et al. (2011), the vortex type can be identified.

[0064] Figure 3This is a visualization of the characteristics of each moment in the life cycle of a vortex according to an embodiment of this application. The life cycle of the vortex is normalized, and the vortex characteristic distribution at times 0, 0.25, 0.5, 0.75, and 1 is selected. Time = 0 represents the moment the vortex is generated, and similarly, time = 1 represents the moment the vortex disappears. Figures (a)-(e) show the chlorophyll and surface eddy current distribution of the vortex at each moment. The background field is chlorophyll, and the black vector represents the surface eddy current. It can be seen that from the generation to the disappearance of the vortex, the chlorophyll concentration inside the vortex tends to increase, while the velocity of the surface eddy current tends to decrease. Figures (f)-(j) show the SLA and morphological size distribution of the vortex at each moment. The background field is SLA, and the black dashed line is the fitted circle of the vortex radius. It can be seen that from the generation to the disappearance of the vortex, the position changes, and the overall vortex moves westward, while the morphological size does not change significantly. Figures (k)-(o) show the surface temperature distribution of the vortex at each moment. It can be seen that from the generation to the disappearance of the vortex, the surface temperature tends to increase.

[0065] Figure 4 This is a schematic diagram illustrating the intraday variation of chlorophyll concentration and surface eddies in a vortex according to an embodiment of this application. The intraday variation time range is 8:30-15:30 (Beijing time). The background field represents chlorophyll, and the black vectors represent surface eddies. It can be seen that the chlorophyll concentration inside the vortex exhibits a diday variation trend of first increasing and then decreasing, and the number of eddies at the vortex center also shows a trend of first increasing and then decreasing. Combined with... Figure 3 and Figure 4 It can be seen that this application can accurately capture the changes in marine biochemical and physical parameters throughout the entire life cycle of small and medium-scale eddies, and can reflect the intra-diurnal variation characteristics of small and medium-scale eddies.

Claims

1. A method for identifying small-to-medium scale vortex features by fusing active and passive remote sensing data, characterized in that... Includes the following steps: Step S1: Use sea surface height anomaly data to obtain a tracking dataset of small and medium-scale eddies, and calculate the eddy's life cycle, center point latitude and longitude coordinates, and radius; S11: Define the centroid of the vortex at time T0 as the center point of the vortex; set the number of times no vortex was detected, T1, to 0; T = T0; S12: Taking the vortex center as the origin, the larger of the following values ​​is taken westward: 1.75 times the 7-day ocean propagation distance of the baroclinic Rossby wave phase velocity and 150 km. The distance is taken as 150 km in the east, south, and north directions. Within this range, the vortex at time T = T+1 with the smallest similarity value and satisfying a similarity value less than 0.5 is selected. The similarity calculation formula is: In the formula, ΔS, ΔL s ,Δξ,ΔE k S0, L represent the vortex spatial distance, radius difference, relative vorticity difference, and EKE difference between two adjacent moments, respectively; s0 ,ξ0,E k0 These are their respective characteristic values, at 100km, 500km, and 10km respectively. -6 s -1 and 100cm 2 s -2 ; S13: If a vortex of the same origin can be detected at time T, it is considered that the vortex continues to exist, and the number of consecutive vortexes not detected T1 is reset to zero before proceeding to step S12; if it cannot be detected, it is recorded as the vortex being missing once, and the number of consecutive vortexes not detected T1 is incremented by 1. S14: Is the number of consecutive times T1 that no vortex is detected greater than the vortex disappearance confirmation setting value? If not, proceed to step S12. If yes, the life cycle of the vortex ends at the time T - vortex disappearance confirmation setting value. Step S2: Obtain chlorophyll concentration data within the eddy range using static water color remote sensing satellite imagery through a matching algorithm; Step S3: Using the maximum correlation coefficient flow field inversion algorithm, surface eddy vectors are obtained with chlorophyll concentration as a tracer; Step S4: Obtain the surface temperature of small- and medium-scale eddies using sea surface temperature product data; Step S5: Combining the data obtained from steps S1 to S4, establish a time dataset of the latitude and longitude coordinates of the vortex center point, vortex radius, vortex duration, chlorophyll concentration, surface vortex vector, and vortex surface temperature within the vortex life cycle. Step S6: Select key time point data with equal time intervals from Step S5 to construct key characteristic parameters of surface eddy changes, morphological structure changes, chlorophyll concentration changes, and surface temperature changes of small-scale eddies throughout their entire life cycle.

2. The method for identifying small- and medium-scale vortex features by fusing active and passive remote sensing data according to claim 1, characterized in that, The vortex identification process must meet the following requirements: (1) High-pass filtering is used to smooth the data in order to remove Rossby wave coherent signals; (2) The SSH values ​​of all pixels are within the SSH threshold range; (3) The number of pixels in the connected region is greater than 8 and less than 1000; (4) There is at least one local SLA extreme value; (5) The vortex amplitude is greater than or equal to 7.5 cm; (6) The distance between any two points in a connected region is less than a given maximum value.

3. The method for identifying small- and medium-scale vortex features by fusing active and passive remote sensing data according to claim 1, characterized in that, In step S2, the radius of the chlorophyll extraction is equal to the radius of the vortex.

4. The method for identifying small-to-medium scale vortex features by fusing active and passive remote sensing data according to claim 1, characterized in that, The time resolution of chlorophyll concentration and surface eddy vector in the time dataset obtained in step S5 is in hours, covering the range of 8:30-15:30 Beijing time; the time resolution of vortex center point latitude and longitude coordinates, vortex radius, vortex life cycle and vortex surface temperature is in days.

5. The method for identifying small-to-medium scale vortex features by fusing active and passive remote sensing data according to claim 1, characterized in that, The chlorophyll coverage area in the vortex accounts for more than 30% of the total vortex area.

Images