Method for identifying dynamic processes of extreme weather events based on three-dimensional image connectivity

Abstract

The present application relates to a method for identifying dynamic process of extreme climate events based on three-dimensional image connectivity, comprising the steps of: selecting an extreme climate index, constructing a three-dimensional index space body; setting an extreme climate index threshold, binarizing the three-dimensional index space body; identifying and marking a three-dimensional connected body in the three-dimensional index space body; according to the layer-by-layer area and the overlapping area of adjacent two layers of the three-dimensional connected body, splitting the three-dimensional connected body into one or more new connected bodies; a new connected body independently represents an extreme climate event with three-dimensional visual evolution characteristics; constructing extreme climate event development spatiotemporal characteristic variables in the spatiotemporal dimension; using the extreme climate event development spatiotemporal characteristic variables, obtaining event development characteristics and event migration characteristics. The present application realizes simple and efficient algorithm, small cache capacity, and can objectively and reliably identify extreme climate events in a large area spanning decades; provides a new way for analyzing extreme climate based on events, and provides a certain reference basis for studying regional extreme climate mechanism, adapting to climate change, managing climate disaster risk, and implementing disaster prevention and mitigation strategy.

Description

Technical Field

[0001] This invention relates to the field of extreme climate change and climate disaster risk management, and more specifically to a method for identifying the dynamic processes of extreme climate events based on three-dimensional image connectivity. Background Technology

[0002] The latest Sixth Assessment Report released by the Intergovernmental Panel on Climate Change (IPCC) indicates that the global average temperature rose by approximately 1.09°C over the past decade (2011–2020) compared to the period of the Industrial Revolution (1850–1900). Global warming has significantly altered the characteristics of the water cycle, impacting the spatial and temporal distribution of global water resources and exacerbating extreme weather events such as torrential rains and floods, droughts, cold waves, and heat waves. The total number of natural disasters worldwide surged in the first two decades of the 21st century, particularly climate-related disasters, which saw a dramatic increase. From 1980 to 1999, there were 3,656 climate-related disasters globally, while this number rose to 6,681 from 2000 to 2019. Extreme weather events severely impact economic and social development and ecosystem security, causing serious damage to agriculture, industry, the ecological environment, and the water environment.

[0003] Developing effective climate disaster management and adaptation measures must be based on existing knowledge and requires an understanding of long-term extreme climate event patterns. Scientific identification of extreme climate events is a fundamental prerequisite for exploring their pattern characteristics and revealing their internal mechanisms.

[0004] The occurrence, development, and termination of extreme climate events exhibit distinct three-dimensional characteristics (time as the first dimension, latitude and longitude as the second and third dimensions), meaning they are continuous in time and dynamically expand or contract in space. The spatiotemporal structures of different rainstorm and flood events vary significantly. How to comprehensively consider the spatiotemporal continuity of extreme climate events and study their dynamic development processes to provide more reliable theoretical and technical support for adapting to climate change, managing climate disaster risks, and implementing disaster prevention and mitigation strategies is a question that requires further exploration.

[0005] Current state of technology:

[0006] Current domestic and international research mainly focuses on the dimensionality reduction identification and analysis of extreme climate events from the perspectives of one-dimensional time and two-dimensional plane, concentrating on the temporal evolution characteristics or spatial distribution characteristics of extreme climate events.

[0007] Deficiencies of existing technology:

[0008] Because the dimensionality reduction identification methods used in existing technologies destroy the overall structure of extreme climate events, they cannot effectively reveal the process and related mechanisms of the development and evolution of extreme climate events. Summary of the Invention

[0009] To address the aforementioned problems, this invention provides a method for identifying the dynamic processes of extreme weather events based on three-dimensional image connectivity. The aim is to achieve a simple and efficient algorithm with a small cache capacity, and to objectively and reliably identify extreme weather events spanning decades over a large area.

[0010] To solve the above problems, the technical solution provided by the present invention is as follows:

[0011] A method for identifying dynamic processes of extreme weather events based on 3D image connectivity includes the following steps:

[0012] S100. Select extreme climate indicators and construct a three-dimensional indicator space.

[0013] S200. Set the extreme climate index threshold and perform binarization processing on the three-dimensional index space volume;

[0014] S300. Identify and mark three-dimensional connected volumes in the three-dimensional index space;

[0015] S400. The three-dimensional connected component is split into one or more new connected components; each new connected component independently represents an extreme climate event with three-dimensional visualized evolution characteristics;

[0016] S500. Construct spatiotemporal characteristic variables for the development of extreme climate events in a spatiotemporal dimension; the spatiotemporal characteristic variables for the development of extreme climate events include daily event concentration points and spatiotemporal event concentration points;

[0017] Then, using the spatiotemporal characteristic variables of the extreme climate events, the event development characteristics and event migration characteristics are obtained; the event development characteristics include the daily development direction of a single event; the event migration characteristics include the migration direction between multiple events;

[0018] The event development characteristics and the event migration characteristics are the final output results of this invention.

[0019] Preferably, the three-dimensional indicator space volume is constructed based on the spatiotemporal sequence of the data and the study area, and is in the form of a three-dimensional indicator space volume matrix, expressed by the following formula:

[0020] Matrix = M * N * L

[0021] Where: Matrix is ​​the size of the three-dimensional index space matrix; M is the number of grid points in longitude within the study area; N is the number of grid points in latitude within the study area; L is the time series length of the data, in days.

[0022] Preferably, in step S200, the three-dimensional index space volume is binarized, specifically including the following steps:

[0023] S210. Compare the extreme climate index value of each grid point with the extreme climate index threshold, and then perform the following operations based on the comparison results:

[0024] If the extreme climate index value of the grid point is greater than the extreme climate index threshold, then the grid point is considered to have experienced an extreme climate event, and the grid point is considered a valid pixel.

[0025] If the extreme climate index value of the grid point is not greater than the extreme climate index threshold, then the grid point is considered to have not experienced an extreme climate event, and the grid point is considered an invalid pixel.

[0026] Preferably, in S300, based on the principle of three-dimensional image connectivity, the three-dimensional connected components are identified and marked in the three-dimensional index space volume, specifically including the following steps:

[0027] S310. Define a three-dimensional image as a set of multiple consecutive pixels; the pixels are expressed by the following formula:

[0028] Pix = {x, y, z, i}

[0029] Where: Pix is ​​the pixel; (x, y, z) are the coordinates of the pixel in three-dimensional space; x is the longitude of the pixel in three-dimensional space; y is the latitude of the pixel in three-dimensional space; z is the time of the pixel in three-dimensional space; i is the intensity value of the pixel, which is also the value of the extreme climate index;

[0030] S320. Based on the connection direction and distance between two pixels in three-dimensional space, determine whether the two pixels meet the requirement of continuity; then, based on the determination result, perform the following operations:

[0031] If the two pixels are continuous, then the two pixels are determined to be adjacent to each other;

[0032] If the two pixels do not meet the continuity requirement, then the two pixels are determined not to be adjacent to each other.

[0033] S330. Repeat S320 until each pair of pixels in the three-dimensional space has been determined to be continuous, and then execute S340.

[0034] S340. Pack all the pixels that are adjacent to each other to obtain a continuous three-dimensional image; this continuous three-dimensional image is the three-dimensional connected body;

[0035] Each day in the three-dimensional connected component is a layer along the ascending direction of the time series length of the data.

[0036] Preferably, in step S400, splitting the three-dimensional connected component into one or more new connected components specifically includes the following steps:

[0037] S410. Calculate the area of ​​the three-dimensional connected volume in each layer and the overlapping area between adjacent layers in ascending order of the time series length of the data; the overlapping area between adjacent layers is the total area of ​​the grid points that overlap between adjacent layers.

[0038] S420. Compare the layer-by-layer area and the overlapping area of ​​two adjacent layers with a pre-set area threshold, and then perform the following operations based on the comparison results:

[0039] If the area of ​​each layer is less than the area threshold, or the overlapping area of ​​two adjacent layers is less than the area threshold, then the information for that day is deleted, and the three-dimensional connected body is split off using that day as the boundary; specifically: the part of the three-dimensional connected body before that day is the first new connected body, and is marked as an extreme climate event with three-dimensional visualization evolution characteristics; the part of the three-dimensional connected body after that day is the second new connected body, and S410 is only executed on the second new connected body thereafter.

[0040] If the area of ​​each layer is not less than the area threshold, and the overlapping area of ​​two adjacent layers is not less than the area threshold, then return to and execute S410 again.

[0041] S430. Repeat the above steps until the traversal is complete.

[0042] Preferably, the daily movement of the event's focal point is used to characterize the direction of the event center's movement during the event's development, expressed by the following formula:

[0043]

[0044] Where: (XX, YY) represents the event intensity-weighted centroid of the event on a specific day; X j Y represents the longitude of the j-th grid point of the event on a specific day; j Let S be the latitude of the j-th grid point of the event on a specific day; j Let J be the value of the extreme climate index at the j-th grid point on a specific day; J is the total number of grid points affected by the event on a specific day.

[0045] Preferably, the movement process of the event spatiotemporal concentration point is used to reflect the direction of movement of the event center during the development of events, and is expressed by the following formula:

[0046]

[0047] Where: (XXX, YYY, TTT) represents the centroid of the three-dimensional volume structure of the event; Xij Y represents the longitude of the j-th grid point on the i-th day of the event. ij S represents the latitude of the j-th grid point on day i of the event; ij is the value of the extreme climate index at the j-th grid point on day i of the event; D is the number of days the event lasts.

[0048] Preferably, in step S320, determining whether two pixels meet the continuity requirement based on the connection direction and distance between them in three-dimensional space specifically includes the following steps:

[0049] S321. Examine the chessboard distance between the two pixels one by one in the 26 connection directions, and then perform the following operations based on the examination results:

[0050] If the chessboard distance between these two pixels is 1 in any of the 26 connection directions, then the two pixels are determined to be continuous.

[0051] If the chessboard distance between these two pixels is not 1 in any of the 26 connection directions, then the two pixels are determined to be non-continuous.

[0052] Preferably, the extreme climate index threshold is determined and set using the relative threshold method, the absolute threshold method, or the rating standard method for climate characteristic indicators.

[0053] Preferably, in S410, the layer-by-layer area of ​​the three-dimensional connected body in each layer and the overlapping area of ​​adjacent layers are calculated based on the occurrence and development process of regional extreme climate events.

[0054] Compared with the prior art, the present invention has the following advantages:

[0055] 1. Because this invention adopts a method for identifying the dynamic development process of extreme climate events based on three-dimensional image connectivity, and is based on the principle of three-dimensional image connectivity, it achieves a simple and efficient algorithm with small cache capacity, and can objectively and reliably identify extreme climate events spanning decades over a large area.

[0056] 2. Since the method of this invention can identify extreme climate events from a three-dimensional perspective, it provides a new approach to analyzing extreme climate from an event perspective, and provides a certain reference for studying the mechanism of extreme climate in a region, adapting to climate change, managing climate disaster risks, and implementing disaster prevention and mitigation strategies. Attached Figure Description

[0057] A schematic diagram of the method for identifying extreme weather events based on three-dimensional image connectivity according to the present invention. Detailed Implementation

[0058] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0059] A method for identifying dynamic processes of extreme weather events based on 3D image connectivity includes the following steps:

[0060] S100. Select extreme climate indicators and construct a three-dimensional indicator space.

[0061] In this specific embodiment, the three-dimensional index space volume is constructed based on the spatiotemporal sequence of gridded daily precipitation data and the study area, and is in the form of a three-dimensional index space volume matrix, expressed according to equation (1):

[0062] Matrix = M * N * L (1)

[0063] Where: Matrix is ​​the size of the three-dimensional index space matrix; M is the number of grid points in longitude within the study area; N is the number of grid points in latitude within the study area; L is the time series length of the data, in days.

[0064] In this specific embodiment, M=9, N=9, L=6.

[0065] S200. Set the threshold for extreme climate indicators and perform binarization on the three-dimensional indicator space volume.

[0066] In this specific embodiment, the extreme climate index threshold is determined and set using the relative threshold method, the absolute threshold method, and the grading standard method for climate characteristic indicators.

[0067] In this specific embodiment, S200 performs binarization processing on the three-dimensional index space volume, specifically including the following steps:

[0068] S210. Compare the extreme climate index value of each grid point with the extreme climate index threshold, and then perform the following operations based on the comparison results:

[0069] If the extreme climate index value of a grid point is greater than the extreme climate index threshold, then the grid point is considered to have experienced an extreme climate event, and the grid point is considered a valid pixel.

[0070] If the extreme climate index value of a grid point is not greater than the extreme climate index threshold, it is considered that no extreme climate event has occurred at this grid point, and this grid point is also considered an invalid pixel.

[0071] In this specific embodiment, the extreme precipitation threshold for each grid point is taken as the 90th percentile of the daily precipitation over many years for that grid point. When the daily precipitation of a grid point exceeds this threshold, it is assigned a value of 1, indicating that an extreme precipitation event has occurred, and the grid point is considered a valid pixel; otherwise, it is assigned a value of 0, indicating that no extreme precipitation event has occurred, and the grid point is considered an invalid pixel. This completes the binarization processing of the three-dimensional precipitation index spatial volume in S200.

[0072] S300. Identify and label the three-dimensional connected volume E in the three-dimensional index space.

[0073] In this specific embodiment, S300, based on the principle of three-dimensional image connectivity, identifies and marks the three-dimensional connected volume E in the three-dimensional index space volume, specifically including the following steps:

[0074] S310. Define a three-dimensional image as a set of multiple consecutive pixels; pixels are expressed according to equation (2):

[0075] Pix = {x, y, z, i} (2)

[0076] Where: Pix is ​​a pixel; (x, y, z) are the coordinates of the pixel in three-dimensional space; x is the longitude of the pixel in three-dimensional space; y is the latitude of the pixel in three-dimensional space; z is the time of the pixel in three-dimensional space; i is the pixel intensity value, which is also the value of the extreme climate index.

[0077] S320. Determine whether two pixels in three-dimensional space are continuous based on the connection direction and distance between them.

[0078] In this specific embodiment, step S320, which determines whether two pixels meet the continuity requirement based on the connection direction and distance between them in three-dimensional space, specifically includes the following steps:

[0079] S321. Examine the chessboard distance between these two pixels one by one in the 26 connecting directions, and then perform the following operations based on the examination results:

[0080] If the chessboard distance between two pixels is 1 in any of the 26 connection directions, then the two pixels are considered to be continuous.

[0081] If the chessboard distance between these two pixels is not 1 in any of the 26 connection directions, then these two pixels are determined not to be continuous.

[0082] Then, based on the judgment result, the following operations are performed:

[0083] If two pixels are continuous, then they are considered to be neighbors.

[0084] If these two pixels do not meet the continuity requirement, then these two pixels are determined not to be adjacent.

[0085] S330. Repeat S320 until every pair of pixels in the three-dimensional space has been determined to be continuous, then execute S340.

[0086] S340. Pack all neighboring pixels together to obtain a continuous 3D image; this continuous 3D image is the 3D connected volume E.

[0087] It should be noted that the three-dimensional images exhibit different colors, shapes, and heights due to the varying intensity values ​​of these pixels.

[0088] It should be further explained that a complete extreme weather event can be generalized as a three-dimensional connected entity, which is similar to a three-dimensional connected image. Current image connectivity recognition technology is relatively mature, with a series of computationally simple and efficient algorithms. Therefore, this invention introduces it into the extreme weather event recognition process and implements it using the bwconncomp function in MATLAB software.

[0089] It needs to be further explained that the principle of connectivity in three-dimensional images is as follows: Figure 1 As shown, a 3D image can be viewed as being composed of a large number of continuous pixels. Whether any two pixels in 3D space are continuous depends on their connection direction and distance. Regarding the connection direction, for any pixel, its cubic neighborhood is taken in 3D space. Starting from each corner point, the center point of each face, and the center point of each side of the cube, a set of 26 connection directions can be formed, representing the possible connection directions between the target pixel and all its neighbors. Regarding the distance, a chessboard distance is used here. If the chessboard distance between two pixels is 1 in any of the 26 connection directions, then these two pixels are considered continuous and are neighbors. Figure 1 The target pixel (dark blue dot) is given 26 connection directions and 26 neighboring pixels (light blue dots) in 3D space. All neighboring pixels can form a continuous 3D image, and the 3D image presents different colors, shapes, and heights due to the different intensity values ​​of these pixels.

[0090] Based on the above principle of three-dimensional image connectivity, the bwconncomp function in MATLAB software can be used to quickly identify, label, and extract three-dimensional connected components E.

[0091] Each day in the three-dimensional connected volume E is a layer along the ascending direction of the time series length of the data.

[0092] S400. Split the three-dimensional connected component E into one or more new connected components; each new connected component independently represents an extreme climate event with three-dimensional visualization of its evolution.

[0093] In this specific embodiment, step S400, which splits the three-dimensional connected component E into a first new connected component E1 and a second new connected component E2, specifically includes the following steps:

[0094] S410. Calculate the layer-by-layer area g of the three-dimensional connected body E in each layer and the overlapping area h between adjacent layers along the ascending direction of the time series length of the data; the overlapping area h between adjacent layers is the total area of ​​the pixels that overlap in space between adjacent layers.

[0095] In this specific embodiment, in S410, the layer-by-layer area g of the three-dimensional connected body E and the overlapping area h of adjacent layers are calculated based on the occurrence and development process of regional extreme climate events.

[0096] S420. Compare the area g of each layer and the overlapping area h of two adjacent layers with the manually preset area threshold f, and then perform the following operations based on the comparison results:

[0097] If the area g of each layer is less than the area threshold f, or the overlapping area h of two adjacent layers is less than the area threshold f, then the information for that day is deleted, and the three-dimensional connected component E is split off using that day as the boundary. Specifically: the part of the three-dimensional connected component E before that day is the first new connected component E1, and is marked as an extreme climate event with three-dimensional visualization evolution characteristics; the part of the three-dimensional connected component E after that day is the second new connected component E2, and S410 is only executed on the second new connected component E2 thereafter.

[0098] If the area g of each layer is not less than the area threshold f, and the overlapping area h of two adjacent layers is not less than the area threshold f, then return to and execute S410 again.

[0099] S430. Repeat the above steps until the traversal is complete.

[0100] It should be noted that the steps S400 to S430 above can also be expressed as follows:

[0101] Taking into account factors such as the occurrence and development of regional extreme climate events, the layer-by-layer area g and the overlapping area h of two adjacent layers of the three-dimensional connected body E (duration is m, unit is days) are calculated.

[0102] If g or h is less than the area threshold f on day i, then the three-dimensional connected component E is split into a new connected component E1 (duration i-1, in days) and a new connected component E2 (duration mi, in days) on day i.

[0103] It should be further noted that the threshold for the occurrence and overlap of extreme climate events is related to the area of ​​the study region, and is usually set as a percentage of the study region area. For example, when identifying drought events, the area threshold can be set to 1% to 2% of the study region area.

[0104] In this specific embodiment, the layer-by-layer area of ​​the three-dimensional connected body E and the overlapping area of ​​two adjacent layers are calculated. The results show that the event area on the 4th day is less than the area threshold. Therefore, the information of this layer is deleted, resulting in the split new connected body E1 (lasting 3 days) and the new connected body E2 (lasting 2 days).

[0105] In this specific embodiment, the newly connected components E1 and E2 after splitting are respectively a regional continuous extreme precipitation event with three-dimensional visualization evolution characteristics.

[0106] S500. Construct spatiotemporal characteristic variables for the development of extreme climate events in the spatiotemporal dimension; the spatiotemporal characteristic variables for the development of extreme climate events include the daily concentration points of events and the spatiotemporal concentration points of events.

[0107] In this specific embodiment, the daily movement of the event's focal point is used to characterize the direction of the event center's movement during the event's development, expressed by equation (3):

[0108]

[0109] Where: (XX, YY) represents the event intensity-weighted centroid of the event on a specific day; X j Y represents the longitude of the j-th grid point of the event on a specific day; j Let S be the latitude of the j-th grid point of the event on a specific day; j Let J be the value of the extreme climate index at the j-th grid point on a specific day; J is the total number of grid points affected by the event on a specific day.

[0110] It should be noted that the intensity of the events on that day was concentrated at this point.

[0111] In this specific embodiment, the movement process of the event spatiotemporal concentration point is used to reflect the movement direction of the event center during the development of events, and is expressed by equation (4):

[0112]

[0113] Where: (XXX, YYY, TTT) represents the centroid of the three-dimensional volume structure of the event; X ij Y represents the longitude of the j-th grid point on the i-th day of the event. ij S represents the latitude of the j-th grid point on day i of the event; ij is the value of the extreme climate index at the j-th grid point on day i of the event; D is the number of days the event lasts.

[0114] It should be noted that the intensity of the entire event was concentrated at this point.

[0115] Then, by utilizing the spatiotemporal characteristic variables of extreme climate events, we obtain the event development characteristics and event migration characteristics; the event development characteristics include the daily development direction of a single event; the event migration characteristics include the migration direction between multiple events.

[0116] This specific embodiment also provides the following example of calculating the daily development direction of a single event to demonstrate the technical effects of the present invention:

[0117] Calculate the daily concentration points for two consecutive days; determine the angle θ between the line connecting the two concentration points and the positive X-axis in a counter-clockwise direction; the range of angle θ can be divided into eight directions, as shown in Table 1:

[0118] Table 1. Criteria for classifying the direction of event development (unit: °)

[0119] (0,22.5) East (202.5,247.5) southwest (22.5,67.5) northeast (247.5,292.5) South (67.5,112.5) north (292.5,337.5) southeast (112.5,157.5) northwest (337.5,360.5) East (157.5,202.5) West -

[0120] The event development characteristics and event migration characteristics are the final output results of this invention.

[0121] In this specific embodiment, spatiotemporal characteristic variables of extreme precipitation event development are constructed in the spatiotemporal dimension to obtain the event development and movement characteristics, as shown in Table 2:

[0122] Table 2. Development and migration characteristics of extreme precipitation events (unit: °)

[0123]

[0124]

[0125] In the above detailed description, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features of the single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, wherein each claim stands alone as a preferred embodiment of the invention.

[0126] The disclosed embodiments have been described above to enable any person skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments without departing from the spirit and scope of this disclosure. Therefore, this disclosure is not limited to the embodiments given herein, but is consistent with the broadest scope of the principles and novel features disclosed in this application.

[0127] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

[0128] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying the dynamic processes of extreme weather events based on three-dimensional image connectivity, characterized in that: Includes the following steps: S100. Select extreme climate indicators and construct a three-dimensional indicator space. S200. Set the extreme climate index threshold and perform binarization processing on the three-dimensional index space volume; S300. Identify and mark three-dimensional connected volumes in the three-dimensional index space; S400. The three-dimensional connected component is split into one or more new connected components; each new connected component independently represents an extreme climate event with three-dimensional visualized evolution characteristics; S500. Construct spatiotemporal characteristic variables for the development of extreme climate events in a spatiotemporal dimension; the spatiotemporal characteristic variables for the development of extreme climate events include daily event concentration points and spatiotemporal event concentration points; Then, using the spatiotemporal characteristic variables of the extreme climate events, the event development characteristics and event migration characteristics are obtained; the event development characteristics include the daily development direction of a single event; the event migration characteristics include the migration direction between multiple events; The event development characteristics and the event migration characteristics are the final output results of this invention. The three-dimensional indicator space volume is constructed based on the spatiotemporal sequence of the data and the scope of the study area. It is in the form of a three-dimensional indicator space volume matrix, expressed by the following formula: in: The size of the three-dimensional index space volume matrix; This represents the number of grid points representing longitudes within the study area; This represents the number of grid points representing the latitude within the study area; The time series length of the data, in days; S200 performs binarization processing on the three-dimensional index space volume, specifically including the following steps: S210. Compare the extreme climate index value of each grid point with the extreme climate index threshold, and then perform the following operations based on the comparison results: If the extreme climate index value of the grid point is greater than the extreme climate index threshold, then the grid point is considered to have experienced an extreme climate event, and the grid point is considered a valid pixel. If the extreme climate index value of the grid point is not greater than the extreme climate index threshold, then the grid point is considered to have not experienced an extreme climate event, and the grid point is considered an invalid pixel. Based on the principle of three-dimensional image connectivity, S300 identifies and marks the three-dimensional connected volume in the three-dimensional index space volume, specifically including the following steps: S310. Define a three-dimensional image as a set of multiple consecutive pixels; the pixels are expressed by the following formula: in: For the pixel point; The coordinates of the pixel in three-dimensional space; The longitude of the pixel in three-dimensional space; Let be the latitude of the pixel in three-dimensional space; The time of the pixel in three-dimensional space; This is the pixel intensity value, and also the value of the extreme climate index; S320. Based on the connection direction and distance between two pixels in three-dimensional space, determine whether the two pixels conform to continuity; specifically: Examine the chessboard distance between the two pixels one by one in the 26 connection directions, and then perform the following operations based on the determination results: If the chessboard distance between these two pixels is 1 in any of the 26 connection directions, then these two pixels are determined to be continuous and are neighbors. If the chessboard distance between these two pixels is not 1 in any of the 26 connection directions, then the two pixels are determined to be non-continuous. S330. Repeat S320 until each pair of pixels in the three-dimensional space has been determined to be continuous, and then execute S340. S340. Pack all the pixels that are adjacent to each other to obtain a continuous three-dimensional image; this continuous three-dimensional image is the three-dimensional connected body; Each day in the three-dimensional connected component is a layer along the ascending direction of the time series length of the data.

2. The method for identifying the dynamic process of extreme climate events based on three-dimensional image connectivity according to claim 1, characterized in that: In S400, the three-dimensional connected component is split into one or more new connected components, specifically including the following steps: S410. Calculate the layer-by-layer area of ​​the three-dimensional connected volume and the overlapping area between adjacent layers in each layer along the ascending direction of the time series length of the data; the overlapping area between adjacent layers is the total area of ​​the pixels in adjacent layers that overlap in space. S420. Compare the layer-by-layer area and the overlapping area of ​​two adjacent layers with a pre-set area threshold, and then perform the following operations based on the comparison results: If the area of ​​each layer is less than the area threshold, or the overlapping area of ​​two adjacent layers is less than the area threshold, then the information for that day is deleted, and the three-dimensional connected body is split off using that day as the boundary; specifically: the part of the three-dimensional connected body before that day is the first new connected body, and is marked as an extreme climate event with three-dimensional visualization evolution characteristics; the part of the three-dimensional connected body after that day is the second new connected body, and S410 is only executed on the second new connected body thereafter. If the area of ​​each layer is not less than the area threshold, and the overlapping area of ​​two adjacent layers is not less than the area threshold, then return to and execute S410 again. S430. Repeat the above steps until the traversal is complete.

3. The method for identifying the dynamic process of extreme weather events based on three-dimensional image connectivity according to claim 2, characterized in that: The daily movement of the event's focal point is used to characterize the direction of the event center's movement during the event's development, and is expressed by the following formula: in: The event intensity-weighted centroid represents the event intensity on a specific day; For the event on a specific day j Longitude of each grid point; For the event on a specific day j The latitude of each grid point; For the event on a specific day j The values ​​of the extreme climate index at each grid point; This represents the total number of grid points affected by the event on a specific day.

4. The method for identifying the dynamic process of extreme weather events based on three-dimensional image connectivity according to claim 3, characterized in that: The movement process of the spatiotemporal focal point of the event is used to reflect the direction of movement of the event center during the development of events, and is expressed by the following formula: in: The centroid of the three-dimensional volume structure representing the event; For the event number i Heavenly j Longitude of each grid point; For the event number i Heavenly j The latitude of each grid point; For the event number i Heavenly j The values ​​of the extreme climate index at each grid point; The number of days the event lasted.

5. The method for identifying the dynamic process of extreme climate events based on three-dimensional image connectivity according to claim 4, characterized in that: S320 determines whether two pixels meet the continuity requirement based on the connection direction and distance between them in three-dimensional space, specifically including the following steps: S321. Examine the chessboard distance between the two pixels one by one in the 26 connection directions, and then perform the following operations based on the examination results: If the chessboard distance between these two pixels is 1 in any of the 26 connection directions, then the two pixels are determined to be continuous. If the chessboard distance between these two pixels is not 1 in any of the 26 connection directions, then the two pixels are determined to be non-continuous.

6. The method for identifying the dynamic process of extreme climate events based on three-dimensional image connectivity according to claim 5, characterized in that: The extreme climate index thresholds are determined and set using the relative threshold method, the absolute threshold method, and the grading standard method for climate characteristic indicators.

7. The method for identifying the dynamic process of extreme climate events based on three-dimensional image connectivity according to claim 6, characterized in that: In S410, the layer-by-layer area of ​​the three-dimensional connected body and the overlapping area of ​​adjacent layers are calculated based on the occurrence and development process of regional extreme climate events.

Images