Ocean heat wave forecasting method and device, medium and product
By combining multi-source observation parameters and a random forest model, the problem of low accuracy in traditional marine heat wave forecasting has been solved, enabling efficient and accurate forecasting of marine heat waves.
Patent Information
- Application Number
- CN202511935201.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-22
- Publication Date
- 2026-02-17
AI Technical Summary
Traditional marine heat wave forecasting techniques rely on a single data source and ignore multi-source data such as wind fields, resulting in low forecast accuracy for specific marine areas, high computational complexity, and insufficient specificity.
After standardizing multi-source observation parameters (sea surface temperature, 10-meter wind speed, sea level air pressure, etc.), the data are input into a trained random forest model for marine heat wave forecasting. By comprehensively considering multiple factors, the forecast accuracy is improved.
It improves the accuracy and adaptability of marine heatwave forecasts, enhances the forecasting capability for specific marine areas, and reduces computational complexity.
Smart Images
Figure CN121542879A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of intelligent marine heatwave prediction, in particular to a marine heatwave prediction method and device, medium and product. BACKGROUND
[0002] Marine heatwaves (MHWs) are extreme high sea surface temperature events occurring in the ocean, which have important impacts on local marine ecosystems, fishery production and socio-economic activities. It is very important to predict marine heatwaves.
[0003] In the traditional method, the prediction technology of marine heatwaves is directly based on the sea surface temperature output by the data of the marine numerical prediction model.
[0004] Traditional numerical models generally have high computational complexity and require a large amount of computing resources, and their prediction ability for nearshore areas is also limited. Traditional intelligent prediction technology relies on a single data source (such as satellite remote sensing sea surface temperature), ignores the influence of multi-source data such as wind field on the evolution of marine heatwaves, and is not suitable for specific marine areas. Specific marine areas are affected by monsoons, ENSO and other multi-time scale factors, and traditional models do not optimize the feature parameters, resulting in low prediction accuracy. SUMMARY
[0005] The purpose of the present application is to provide a marine heatwave prediction method, device, medium and product to solve the problem of low prediction accuracy of marine heatwaves.
[0006] To achieve the above purpose, the present application provides the following solutions.
[0007] In a first aspect, the present application provides a marine heatwave prediction method, comprising: obtaining the measurement value of the multi-source observation parameter of the target sea area at the current time; the multi-source observation parameter comprises: first observation data and second observation data; the first observation data is sea surface temperature, and the second observation data comprises: 10-meter wind speed, sea level pressure, upward longwave radiation flux, downward longwave radiation flux, upward shortwave radiation flux, downward shortwave radiation flux, sensible heat flux, latent heat flux and 2-meter air temperature; standardizing the measurement value of the multi-source observation parameter at the current time to obtain the standardized value of the multi-source observation parameter at the current time; inputting the standardized value of the multi-source observation parameter at the current time into a marine heatwave prediction model to output the prediction value of the event parameter of the marine heatwave of the target sea area at N future times; the event parameter comprises occurrence and intensity; the occurrence is occurrence of marine heatwave or non-occurrence of marine heatwave, and the intensity is zero intensity, medium intensity, relatively strong intensity or severe intensity; the marine heatwave prediction model is obtained by training a random forest model, and N>1.
[0008] In a second aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the marine heat wave forecasting method.
[0009] In a third aspect, the present application provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the marine heat wave forecasting method.
[0010] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the marine heat wave forecasting method.
[0011] According to the embodiments of the present application, the following technical effects are disclosed: The present application discloses a marine heat wave forecasting method, device, medium and product. First, the measurement values of the multi-source observation parameters of the target sea area at the current time are obtained. The multi-source observation parameters include first observation data and second observation data. The first observation data is sea surface temperature, and the second observation data includes 10-meter wind speed, sea level pressure, upward long-wave radiation flux, downward long-wave radiation flux, upward short-wave radiation flux, downward short-wave radiation flux, sensible heat flux, latent heat flux and 2-meter air temperature. Then, the measurement values of the multi-source observation parameters at the current time are standardized to obtain the standardized values of the multi-source observation parameters at the current time. Finally, the standardized values of the multi-source observation parameters at the current time are input into a marine heat wave forecasting model to output the predicted values of the event parameters of the marine heat wave of the target sea area at N future times. The event parameters include occurrence and intensity. The occurrence is occurrence of marine heat wave or non-occurrence of marine heat wave, and the intensity is zero intensity, medium intensity, relatively strong intensity or severe intensity. The marine heat wave forecasting model is obtained by training a random forest model, and N>1. The present application comprehensively considers the sea surface temperature, 10-meter wind speed, sea level pressure, upward long-wave radiation flux, downward long-wave radiation flux, upward short-wave radiation flux, downward short-wave radiation flux, sensible heat flux, latent heat flux and 2-meter air temperature, and uses the marine heat wave forecasting model obtained by training the random forest model to determine the predicted values of the event parameters of the marine heat wave, thereby improving the marine heat wave forecasting precision. BRIEF DESCRIPTION OF DRAWINGS
[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0013] Figure 1 A flowchart of a marine heat wave forecasting method provided by an embodiment of the present application is shown in FIG. 1. Figure 2 A distribution diagram of the forecasting accuracy of different event parameters of marine heat waves with a forecasting time limit of 2 days in a certain sea area is shown in FIG. 2. Figure 3 A distribution diagram of the forecasting accuracy of different event parameters of marine heat waves with a forecasting time limit of 7 days in a certain sea area is shown in FIG. 3. Figure 4 A diagram showing the measured values of event parameters of marine heat waves in each month of 2020 in a certain sea area is shown in FIG. 4. Figure 5 A diagram showing the predicted values of event parameters of marine heat waves in each month of 2020 in a certain sea area is shown in FIG. 5. Figure 6 A structural diagram of a computer device provided by an embodiment of the present application is shown in FIG. 6. DETAILED DESCRIPTION
[0014] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.
[0015] The purpose of the present application is to provide a marine heat wave forecasting method, device, medium and product, aiming to improve the forecasting accuracy of marine heat waves.
[0016] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.
[0017] In an exemplary embodiment, as shown in FIG. 1, a marine heat wave forecasting method is provided, comprising the following steps. Figure 1
[0018] Step 1: Obtain the measured values of multi-source observation parameters of the target sea area at the current time.
[0019] The multi-source observation parameters include first observation data and second observation data; the first observation data is sea surface temperature, and the second observation data includes 10-meter wind speed, sea level pressure, upward long-wave radiation flux, downward long-wave radiation flux, upward short-wave radiation flux, downward short-wave radiation flux, sensible heat flux, latent heat flux and 2-meter air temperature.
[0020] Step 2: standardizing the measured values of the multi-source observation parameters at the current time to obtain standardized values of the multi-source observation parameters at the current time.
[0021] Specifically, step 2 comprises: using a Z-score standardization (standard deviation standardization) method to respectively standardize the measured values of each parameter in the multi-source observation parameters at the current time to obtain standardized values of each parameter in the multi-source observation parameters at the current time.
[0022] Step 3: inputting the standardized values of the multi-source observation parameters at the current time into the marine heat wave forecasting model to output predicted values of event parameters of the marine heat wave at N future times in the target sea area.
[0023] The event parameters include occurrence and intensity, the occurrence is occurrence of the marine heat wave or non-occurrence of the marine heat wave, and the intensity is zero intensity, medium intensity, relatively strong intensity or severe intensity; the marine heat wave forecasting model is obtained by training a random forest model, and N>1.
[0024] As an optional implementation, in step 3, the determination process of the marine heat wave forecasting model comprises: Step 31: constructing a data set; the data set comprises a plurality of sample data; each sample data comprises: sample values of the multi-source observation parameters at a target historical time and labeled values of the event parameters at N historical times after the target historical time.
[0025] As an optional implementation, step 31 comprises: Step 311: obtaining actual values of the first observation data at a plurality of first historical times in a historical period and actual values of the second observation data at a plurality of second historical times.
[0026] Step 312: taking the actual values of the first observation data at the plurality of first historical times in the historical period as a reference, temporally and spatially aligning the actual values of the second observation data at the plurality of second historical times to obtain aligned values of the second observation data at the plurality of first historical times.
[0027] Step 313: standardizing the actual values of the first observation data at each first historical time to obtain standardized values of the first observation data at the plurality of first historical times.
[0028] Step 314: standardizing the aligned values of the second observation data at each first historical time to obtain standardized values of the second observation data at the plurality of first historical times.
[0029] Step 315: determining the standardized value of the first observation data of the plurality of first historical moments of the historical period and the standardized value of the second observation data of the plurality of first historical moments as the sample data of the plurality of first historical moments of the historical period.
[0030] Step 316: determining the 90th percentile of the actual value of the first observation data of the plurality of first historical moments of the historical period as the marine heat wave threshold value.
[0031] Step 317: determining the marked value of the occurrence of each first historical moment based on the actual value of the first observation data of each first historical moment and the marine heat wave threshold value, respectively.
[0032] As an optional implementation, step 317 comprises: Step 3171: determining any first historical moment as a target historical moment.
[0033] Step 3172: when the actual value of the first observation data of the target historical moment is greater than or equal to the marine heat wave threshold value, determining the marked value of the occurrence of the target historical moment as a marine heat wave.
[0034] Step 3173: when the actual value of the first observation data of the target historical moment is less than the marine heat wave threshold value, determining the marked value of the occurrence of the target historical moment as no marine heat wave.
[0035] Step 318: determining the marked value of the intensity of the first historical moment with the marked value of the occurrence of no marine heat wave as zero intensity.
[0036] Step 319: determining the marked value of the intensity of each first historical moment with the marked value of the occurrence of a marine heat wave based on the actual value of the first observation data of all first historical moments and the marine heat wave threshold value.
[0037] As an optional implementation, step 319 comprises: Step 3191: determining the intensity division index of each first historical moment with the marked value of the occurrence of a marine heat wave based on the actual value of the first observation data of each first historical moment and the marine heat wave threshold value.
[0038] As an optional implementation, step 3191 comprises: calculating the intensity division index of each first historical moment with the marked value of the occurrence of a marine heat wave based on the actual value of the first observation data of each first historical moment and the marine heat wave threshold value by using the intensity division index calculation formula; the intensity division index calculation formula is: .
[0039] wherein, is an intensity division index; is an actual value of the first observation data at any first historical time when the marine heat wave occurs; is a marine heat wave threshold value; is a mean value of the actual values of the first observation data at all first historical times.
[0040] Step 3192: determining the intensity label value of the intensity of the occurrence label value at each first historical time when the marine heat wave occurs according to the intensity division index of the occurrence label value at each first historical time when the marine heat wave occurs, respectively.
[0041] As an optional implementation, step 3192 comprises: Step 31921: determining any first historical time when the marine heat wave occurs as a to-be-labeled time for the occurrence label value.
[0042] Step 31922: determining the intensity label value of the intensity of the to-be-labeled time according to the intensity division index of the to-be-labeled time by using an intensity label formula; the intensity label formula is: .
[0043] wherein, is the intensity label value.
[0044] Step 32: dividing the data set according to a preset proportion to obtain a training set and a test set.
[0045] Step 33: expanding the training set by using a K-neighbor sampling method to obtain an expanded training set.
[0046] Specifically, step 33 comprises: Step 331: determining sample data in which any historical time when the marine heat wave occurs for the occurrence label value in the training set as current sample data; determining each sample data in which all historical times when the marine heat wave occurs for the occurrence label value in the training set except the current sample data as non-current sample data; determining a sample value of any parameter in the multi-source observation parameter or the intensity label value in the current sample data as current data, and determining corresponding data of the current data in each non-current sample data as non-current data; Step 332: calculating the Euclidean distance between the current data and each non-current data corresponding to the current data, respectively; Step 333: sorting all Euclidean distances corresponding to the current data in ascending order to obtain a sorting result corresponding to the current data; Step 334: In the ranking result corresponding to the current data, select a preset number of sample data from the sample data corresponding to each non-current data with a Euclidean distance in the top K as synthetic sample data, and determine any synthetic sample data as target synthetic data; Step 335: Determine the sample value of any parameter in the multi-source observation parameter and the label value of any parameter in the event parameter in the current sample data as a current first value, respectively; and determine the sample value of any parameter in the multi-source observation parameter and the label value of any parameter in the event parameter in the target synthetic data as a current second value, respectively; Step 336: Generate a newly generated value of a corresponding parameter according to the current first value and the current second value by using a synthesis formula; the synthesis formula is: .
[0047] wherein, the newly generated value of the parameter to be generated is a newly generated value of any parameter in the multi-source observation parameter or any parameter in the event parameter; the current first value of the parameter to be generated is the current first value; the random number is a random number; the current second value of the parameter to be generated is the current second value; the absolute value is an absolute value; Step 337: Construct an expanded training set by using the newly generated values of all parameters and the training set.
[0048] Step 34: Construct a random forest model.
[0049] Step 35: Train the random forest model by using the sample values of the multi-source observation parameters of the target historical time of each sample data in the expanded training set as input and the label values of the event parameters of the N historical times after the target historical time of the corresponding sample data as output, to obtain an ocean heat wave prediction model.
[0050] Specifically, after step 35, the following steps are further included: Step 36: Evaluate the ocean heat wave prediction model by using a test set. The evaluation index is : .
[0051] .
[0052] .
[0053] wherein, the hit rate is a hit rate; the number of sample data in the test set whose occurrence label value is occurrence of marine heat wave but whose occurrence prediction value output by the marine heat wave forecasting model is non-occurrence of marine heat wave; the number of sample data in the test set whose occurrence label value is occurrence of marine heat wave but whose occurrence prediction value output by the marine heat wave forecasting model is non-occurrence of marine heat wave; the false alarm rate; the number of sample data in the test set whose occurrence label value is non-occurrence of marine heat wave but whose occurrence prediction value output by the marine heat wave forecasting model is occurrence of marine heat wave; the number of sample data in the test set whose occurrence label value is non-occurrence of marine heat wave but whose occurrence prediction value output by the marine heat wave forecasting model is occurrence of marine heat wave; the symmetric extreme dependence index.
[0054] Further, the marine heat wave forecasting method of the present application is also used to forecast marine heat waves in a certain sea area. The distribution diagram of the forecasting accuracy of different event parameters of marine heat waves in the sea area with a forecasting time of 2 days is shown in Figure 2 The distribution diagram of the forecasting accuracy of different event parameters of marine heat waves in the sea area with a forecasting time of 7 days is shown in Figure 3 The measured value diagram of the event parameters of marine heat waves in the sea area in each month of 2020 is shown in Figure 4 The predicted value diagram of the event parameters of marine heat waves in the sea area in each month of 2020 is shown in Figure 5 It can be seen that the marine heat wave forecasting method of the present application can meet the actual demand in terms of forecasting efficiency, forecasting accuracy and extreme case simulation effect.
[0055] In an exemplary embodiment, a computer device is provided, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the marine heat wave forecasting method.
[0056] In an exemplary embodiment, a computer readable storage medium is provided, which stores a computer program, the computer program being executed by a processor to implement the marine heat wave forecasting method.
[0057] In an exemplary embodiment, a computer program product is provided, comprising a computer program, the computer program being executed by a processor to implement the marine heat wave forecasting method.
[0058] In an exemplary embodiment, a computer device is provided, which can be a server or a terminal, and its internal structure diagram can be as shown in Figure 6As shown in the figure. The computer device includes a processor, a memory, an Input / Output (I / O) interface and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to implement a marine heat wave forecasting method.
[0059] Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.
[0060] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of each method can be included. Any reference to memory, database or other medium used in each embodiment provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc.
[0061] The database involved in each embodiment provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in each embodiment provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.
[0062] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.
[0063] Each technical feature of the above embodiments can be combined arbitrarily. In order to make the description simple, not all possible combinations of each technical feature in the above embodiments are described, but as long as the combination of these technical features does not exist contradictory, it should be considered as the scope of the present application.
[0064] The principles and implementations of the present application are described in detail herein with specific examples. The above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation and application range will be changed. In summary, the content of the specification should not be understood as a limitation of the present application.
Claims
1. A method for forecasting marine heat waves, characterized in that, The marine heat wave forecasting method includes: The measurement values of multi-source observation parameters for the target sea area at the current moment are obtained; the multi-source observation parameters include: first observation data and second observation data; the first observation data is sea surface temperature, and the second observation data includes: 10-meter wind speed, sea level air pressure, upward longwave radiation flux, downward longwave radiation flux, upward shortwave radiation flux, downward shortwave radiation flux, sensible heat flux, latent heat flux, and 2-meter air temperature; The measured values of the multi-source observation parameters at the current time are standardized to obtain the standardized values of the multi-source observation parameters at the current time. The standardized values of the multi-source observation parameters at the current moment are input into the marine heat wave forecasting model, which outputs the predicted values of the marine heat wave event parameters for the target sea area at N future moments. The event parameters include the occurrence status and intensity. The occurrence status is whether a marine heat wave has occurred or not, and the intensity is zero intensity, moderate intensity, strong intensity, or severe intensity. The marine heat wave forecasting model is obtained by training a random forest model, where N>1.
2. The marine heatwave forecasting method according to claim 1, characterized in that, The process of determining the marine heat wave forecasting model includes: Construct a dataset; the dataset includes multiple sample data; each sample data includes: sample values of multi-source observation parameters at the target historical moment and labeled values of event parameters at N historical moments after the target historical moment; The dataset is divided into a training set and a test set according to a preset ratio; The training set is expanded using the K-nearest neighbor sampling method to obtain the expanded training set; Construct a random forest model; The random forest model is trained by taking the sample values of the multi-source observation parameters of the target historical time of each sample data in the expanded training set as input and the label values of the event parameters of the corresponding sample data at N historical times after the target historical time as output, and thus obtaining the marine heat wave forecast model.
3. The marine heatwave forecasting method according to claim 1, characterized in that, Constructing the dataset includes: Obtain the actual values of the first observation data at multiple first historical moments and the actual values of the second observation data at multiple second historical moments within a historical period; Based on the actual values of the first observation data at multiple first historical moments in a historical period, the actual values of the second observation data at multiple second historical moments are spatiotemporally aligned to obtain the aligned values of the second observation data at multiple first historical moments. The actual values of the first observation data at each first historical moment are standardized to obtain standardized values of the first observation data at multiple first historical moments. The alignment values of the second observation data at each first historical moment are standardized to obtain standardized values of the second observation data at multiple first historical moments. The standardized values of the first observation data and the standardized values of the second observation data of multiple first historical moments in a historical period are determined as the sample data of multiple first historical moments in the historical period. The 90th percentile of the actual values of the first observation data at multiple first historical moments in a historical period is determined as the ocean heat wave threshold; Based on the actual values of the first observation data at each first historical moment and the ocean heat wave threshold, the occurrence value of each first historical moment is determined. The intensity of the event at the first historical moment when no ocean heat wave occurred is defined as zero intensity. Based on the actual values of the first observation data at all first historical moments and the ocean heat wave threshold, the marker value for determining the occurrence is the marker value of the intensity of each first historical moment at which the ocean heat wave occurs; Each of the first historical moments in the first QN of the historical time periods is determined as the target historical moment; Q>N, where Q is the total number of first historical moments in the historical time period; Multiple sample data are obtained by constructing corresponding sample data based on the sample values of multi-source observation parameters at each historical moment of the target and the labeled values of event parameters at N historical moments after the historical moment of the target. The dataset is constructed based on multiple sample data.
4. The marine heatwave forecasting method according to claim 3, characterized in that, Based on the actual values of the first observation data at each first historical moment and the ocean heat wave threshold, a marker value for the occurrence of each first historical moment is determined, including: Define any first historical moment as the target historical moment; When the actual value of the first observation data at the target historical moment is greater than or equal to the ocean heat wave threshold, the marker value of the occurrence of the target historical moment is determined to be an ocean heat wave. When the actual value of the first observation data at the target historical moment is less than the ocean heat wave threshold, the marker value of the occurrence of the target historical moment is determined as no ocean heat wave has occurred.
5. The marine heatwave forecasting method according to claim 3, characterized in that, Based on the actual values of the first observation data at all first historical moments and the ocean heat wave threshold, the marker values for determining the occurrence of an ocean heat wave are the marker values of the intensity at each first historical moment, including: Based on the actual values of the first observation data at each first historical moment and the ocean heat wave threshold, the marker value for determining the occurrence is the intensity classification index for each first historical moment in which the ocean heat wave occurred; The intensity of each first historical moment of a marine heat wave is determined by classifying the intensity of the occurrence based on the marker value of the occurrence.
6. The marine heatwave forecasting method according to claim 5, characterized in that, Based on the actual values of the first observation data at each first historical moment and the aforementioned marine heat wave threshold, the marker value for determining the occurrence of the marine heat wave is the intensity classification index for each first historical moment in which the marine heat wave occurred, including: Using the intensity classification index calculation formula, based on the actual values of the first observation data at each first historical moment and the aforementioned marine heat wave threshold, the occurrence status is marked as the intensity classification index for each first historical moment in which the marine heat wave occurred; the intensity classification index calculation formula is as follows: ； in, Intensity classification index; The marker value for the occurrence is the actual value of the first observation data at any first historical moment of the occurrence of a marine heat wave; The threshold for ocean heat waves; It is the mean of the actual values of the first observation data for all first historical moments.
7. The marine heatwave forecasting method according to claim 6, characterized in that, The intensity of the marine heat wave at each first historical moment is determined based on the intensity index of the occurrence status, including: The occurrence of the event is marked as any first historical moment of the ocean heat wave as the moment to be marked; Using the intensity marking formula, the intensity marking value at the time to be marked is determined based on the intensity classification index. The intensity marking formula is: ； in, This is the value marked for intensity.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the marine heatwave forecasting method according to any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the marine heatwave forecasting method according to any one of claims 1-7.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the marine heatwave forecasting method according to any one of claims 1-7.