A spatial prediction method for sea ice concentration based on MLR model

Abstract

The present invention discloses a spatial prediction method of sea ice density based on an MLR model, including calculating the lead-lag correlation coefficient between the Weddell Sea sea ice density and the global sea surface temperature, the mean sea level pressure and the surface air temperature in different lead times, so as to determine the atmospheric and oceanic prediction factors affecting the SIC of the Weddell Sea; constructing an MLR spatial prediction model based on the prediction factors, wherein the SIC data of the past 12 months are respectively included in the prediction factors; according to the predicted object, the MLR model performs multiple linear regression on the SIC of each grid point in the Antarctic Weddell Sea region based on the above six prediction factors, so as to realize the monthly prediction of the spatial distribution of sea ice. The present invention is to establish a spatial monthly prediction model of the Weddell Sea SIC based on the MLR statistical method, which can incorporate the climate signals of the distant atmosphere and ocean into the sea ice prediction, and can consider the influence of the persistence of the atmosphere, ocean and sea ice itself on the prediction respectively, and the MLR model shows good prediction ability.

Description

Technical Field

[0001] The present invention relates to the field of spatial prediction of sea ice density, and in particular to a method for spatial prediction of sea ice density based on an MLR model. Background Art

[0002] Polar sea ice is an important part of the Earth's climate system. Due to the limitations of observational data and the highly nonlinear dynamic changes of sea ice, the changes in sea ice are regular but extremely complex, involving physical processes at multiple scales and constrained by many meteorological factors. At present, sea ice forecasts mainly rely on sea ice numerical models, statistical prediction models, and deep learning prediction models. The numerical model of sea ice is part of the Earth's climate system model, which relies on the complex thermal and dynamic characteristics of sea ice for prediction. At present, the mainstream international sea ice models include the Los Alamous Sea Ice Model (Community IceCodE, CICE).

[0003] Statistical model prediction uses statistical methods to explore the changing patterns of data itself. For processes with complex influencing factors and physical processes, it can make up for the shortcomings of numerical model prediction to a certain extent. Statistical prediction of Arctic sea ice began in the 1980s, focusing on the relationship between sea ice and the atmosphere, ocean, etc. for a single parameter. In sea ice prediction, a variety of deep learning models have been widely used. Among them, convolutional neural networks (CNN) have powerful feature extraction capabilities and can automatically learn and extract key spatial features from sea ice images or time series data, providing rich information input for prediction models. In addition, deep neural networks (DNN), artificial neural networks (ANN) and convolutional long short-term memory networks (ConvLSTM) are also often used to establish deep learning prediction models for sea ice prediction.

[0004] The change process of sea ice is extremely complex, involving physical processes at different scales and significantly affected by a variety of meteorological factors. At present, large-scale sea ice changes have been deeply analyzed and simulated in numerical models. However, the abnormal changes in polar sea ice in recent years have put higher demands on existing numerical models. These changes increase the uncertainty of the physical parameter schemes in the model and may lead to significant deviations in the simulation results. In particular, when simulating polar sea ice events, the model shows large deviations and poor prediction results.

[0005] In addition, although deep learning models have shown significant advantages in prediction, their prediction results are largely dependent on the quality and quantity of training data. Previous sea ice predictions mostly used local meteorological elements, which ignored the impact of teleconnected sea-air signals and failed to reflect the impact of teleconnected sea-air factors on sea ice prediction. Therefore, a spatial prediction method for sea ice density based on the MLR model is needed. Summary of the invention

[0006] The purpose of the present invention is to provide a spatial prediction method for sea ice density based on the MLR model.

[0007] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0008] The present invention comprises the following steps:

[0009] The lead-lag correlation coefficients between the Weddell Sea sea ice concentration and global sea surface temperature, mean sea level pressure and surface air temperature at different lead times were calculated to determine the atmospheric and oceanic predictors that affect the Weddell Sea SIC;

[0010] Constructing an MLR spatial prediction model based on the predictors, wherein the predictors respectively include SIC data of the past 12 months;

[0011] According to the predicted object, the MLR model performs multiple linear regression on the SIC of each grid point in the Antarctic Weddell Sea region based on the above six prediction factors to achieve monthly prediction of the spatial distribution of sea ice.

[0012] Furthermore, the prediction factors include the PDO index determined by the North Pacific sea temperature with a lead time of 40 months, the eastern tropical Pacific sea temperature index ETPOSST with a lead time of 28 months, the western tropical Indian Ocean sea temperature index WTIOSST with a lead time of 3 months, the South tropical Atlantic sea temperature index STAO SST with a lead time of 57 months, the SLP index with a lead time of 6 months, and the Weddell-Amundsen Sea SAT index with a lead time of 1 month.

[0013] Furthermore, the method for constructing the MLR-based spatial prediction model comprises:

[0014] For a certain forecast quantity y, establish multiple factors x (x1, x2, ···, x p ) quantitative statistical relationship:

[0015]

[0016] Among them, b1, b2, ···, b p are independent variables x1, x2, ···, x pThe corresponding regression coefficients, b0 is the intercept of y;

[0017] Using the identified six potential influencing factors and their corresponding lead times, the following MLR equation is established:

[0018]

[0019] Among them, k represents the current forecast time, i and j represent the size of the space, where i=1,......,n, j=1,......,m.

[0020] Furthermore, the SIC data of the past 12 months are included in the predictors, and the prediction model of the SIC data is improved as follows:

[0021]

[0022] in, is the intercept, is the regression coefficient of each component on SIC, and the regression coefficient of each grid point is different.

[0023] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0024] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0025] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0026] The beneficial effects of the present invention are:

[0027] The present invention is a spatial prediction method for sea ice density based on the MLR model. Compared with the prior art, the present invention has the following technical effects:

[0028] This paper integrates limited direct observation data of sea ice and combines it with teleconnection climate signals from the atmosphere and ocean to predict the sea ice density in the Weddell Sea of ​​Antarctica.

[0029] The present invention reduces computing resource consumption and computing costs while significantly improving the accuracy of the sea ice statistical prediction model. It can accurately capture the changing trend of Antarctic sea ice and provide important experience and reference for sea ice prediction and climate change research. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1A flow chart of the steps of a method for spatial prediction of sea ice density based on an MLR model of the present invention;

[0031] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0032] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0033] The present invention provides a method for spatial prediction of sea ice density based on an MLR model, comprising the following steps:

[0034] like Figure 1 As shown, in this embodiment, the following steps are included:

[0035] The present invention uses the global atmospheric and oceanic factors far from the Weddell Sea as the prediction factors of sea ice density, and considers the impact of the persistence characteristics of sea ice itself on sea ice prediction. Based on the MLR linear statistical method, an Antarctic sea ice prediction model including the physical processes of global important climate factors is constructed.

[0036] The present invention is based on the atmospheric and oceanic factors of the teleconnection with the Weddell Sea SIC, and uses the MLR statistical method to establish a spatial monthly prediction model for SIC. The process is as follows: Figure 1 shown.

[0037] The steps to build a prediction model are as follows:

[0038] (1) Identify potential influencing factors

[0039] Considering the interaction of atmosphere-ocean-sea ice system, the lead-lag correlation coefficients between Weddell Sea SIC and global sea surface temperature (SST), mean sea level pressure (SLP) and sea air temperature (SAT) were calculated at lead times of 1 to 60 months to determine the atmospheric and oceanic factors affecting Weddell Sea SIC, including the PDO index determined by North Pacific SST (lead time 40 months), the Eastern Tropical Pacific SST index ETPOSST (lead time 28 months), the Western Tropical Indian Ocean SST index WTIO SST (lead time 3 months), the South Tropical Atlantic SST index STAO SST (lead time 57 months), SLP (lead time 6 months) and the Weddell-Amundsen Dipole (AWD) SAT (lead time 1 month). These predictors showed significant correlations with Weddell Sea SIC at different lead times, indicating that they can be used to predict future SIC changes.

[0040] (2) Establishing a spatial prediction model based on MLR

[0041] The MLR method is to establish multiple factors x (x1, x2, ···, x p ) quantitative statistical relationship:

[0042]

[0043] Among them, b1, b2, ···, b p are independent variables x1, x2, ···, x p The corresponding regression coefficients, b0 is the intercept of y. The significance of the multivariate linear regression equation can be determined by performing a variance test or based on the size of the multiple correlation coefficient. If the significance test is passed, the equation is considered to be statistically significant. At the same time, the t-test is used to test the statistical significance of each regression coefficient, and the corresponding independent variable x will significantly affect the dependent variable y.

[0044] In addition, the Jackknife method is used to test the stability of the model. The steps are as follows: During the fitting period, one group of data is removed from the entire time series at a time, and the regression model is rebuilt with the remaining data. The multiple correlation coefficient R is calculated. 2 , adjusted multiple correlation coefficient The f-test value, significance level (p-value) and other statistical indicators are then eliminated year by year, and the above steps are repeated to obtain multiple sets of statistical indicators. If the statistical indicators are close to the results of the original regression model, including all years in the fitting period, and the range of variation is minimal, the regression model is stable and reliable.

[0045] In order to perform spatial monthly prediction, the present invention establishes an MLR prediction model at all 11814 grid points in the Weddell Sea. According to Formula 1, the following MLR equation is established using the determined six potential influencing factors and their corresponding lead times:

[0046]

[0047] Where k represents the current forecast time, and i and j represent the size of the space, where i = 1, ..., n, and j = 1, ..., m. Sea ice concentration uses this linear relationship to generate a forecast based on six prediction factors, called Case 1. According to previous research results, the variability of sea ice is continuous and has the characteristics of interannual variation, with significantly different characteristics in the melting and freezing seasons. Therefore, sea ice concentration adds the SIC of the previous month and the SIC of the same month last year, that is, the SIC of the past 12 months, to the prediction factors to evaluate the role of sea ice persistence.

[0048] Adding past SIC data improves the forecast model as follows:

[0049]

[0050] in, is the intercept, is the regression coefficient of each component on SIC, and the regression coefficient of each grid point is different.

[0051] In summary, the above three cases are set up to compare the prediction ability of sea ice density: the model contains only six basic atmospheric and oceanic factors, Case 1 (Eq. 2), the above six factors plus the SIC of January, Case 2 (Eq. 3), and the above six factors plus the SIC of January and 12 months ago, that is, Case 3 (Eq. 4). In this way, by comparing the prediction results of the models in different cases, the impact of different factors on the SIC of the Weddell Sea can be clearly evaluated, as shown in Table 1.

[0052] Table 1 Three combinations of predictors.

[0053]

[0054] Note: ki indicates the time when the predictor leads the SIC change in the Weddell Sea.

[0055] The MLR model uses the above prediction factors to perform multiple linear regression on the SIC of each grid point in the Antarctic Weddell Sea region, that is, one prediction equation for each grid point. There are a total of 11,814 prediction equations in the Weddell Sea region, and each equation is different. It can be rolled in real time according to the predicted object, and thus the monthly prediction of the spatial distribution of sea ice can be achieved.

[0056] The present invention is based on the MLR statistical method to establish a spatial monthly prediction model for the Weddell Sea SIC, which can incorporate distant atmospheric and oceanic climate signals into sea ice prediction, and can consider the impact of the persistence of the atmosphere, ocean and sea ice itself on the prediction. The MLR model shows good prediction ability, especially in the north, west and south of the Weddell Sea. The prediction effect is better when adding past SIC data, and the predicted values ​​are more concentrated and there are fewer extreme values.

[0057] Compared with the previous sea ice statistical model, the prediction effect of the prediction model established by the present invention has been improved, and it has been strengthened in the entire Weddell Sea, especially after considering the past SIC data. The MLR prediction model can predict the trend of sea ice increase in the 21st century and the maximum value in 2014, and can also predict the sharp decrease in sea ice in 2015-2017 and the minimum value of sea ice in 2023. The anomaly correlation coefficient has increased significantly and the root mean square error has decreased significantly. Moreover, the MLR statistical prediction model can improve the sea ice prediction level based on limited fitting data, which can effectively reduce the calculation cost and improve the calculation efficiency.

[0058] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0059] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0060] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0061] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a spatial prediction device for sea ice density based on the MLR model at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned spatial prediction methods for sea ice density based on the MLR model.

[0062] The above application Figure 1The spatial prediction method of sea ice density based on the MLR model disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0063] The electronic device may also perform Figure 1 A spatial prediction method for sea ice density based on the MLR model was developed and implemented Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0064] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple applications, they execute any of the aforementioned MLR model-based sea ice density spatial prediction methods.

[0065] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0066] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0067] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0068] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0069] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0070] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0071] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0072] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0073] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A spatial prediction method for sea ice density based on the MLR model, characterized in that: The following steps are involved: The lead-lag correlation coefficients between the Weddell Sea sea ice concentration and global sea surface temperature, mean sea level pressure and surface air temperature at different lead times were calculated to determine the atmospheric and oceanic predictors that affect the Weddell Sea SIC; Building an MLR spatial prediction model based on the prediction factors, and adding 12 months of SIC data to the prediction factors respectively; According to the predicted object, the MLR model performs multiple linear regression on the SIC of each grid point in the Antarctic Weddell Sea region based on the predicted factors to achieve monthly prediction of the spatial distribution of sea ice; The prediction factors include the PDO index determined by the North Pacific SST with a lead time of 40 months, the Eastern Tropical Pacific SST index ETPOSST with a lead time of 28 months, the Western Tropical Indian Ocean SST index WTIO SST with a lead time of 3 months, the South Tropical Atlantic SST index STAO SST with a lead time of 57 months, the SLP index with a lead time of 6 months, and the Weddell-Amundsen Sea ADWvSAT index with a lead time of 1 month.

2. The method for spatial prediction of sea ice density based on the MLR model according to claim 1, characterized in that: The method for constructing the MLR-based spatial prediction model comprises: For a certain forecast quantity y, establish multiple factors x (x1, x2, ···, x p ) quantitative statistical relationship: Among them, b1, b2, ···, b p are independent variables x1, x2, ···, x p The corresponding regression coefficients, b0 is the intercept of y; Using the identified six potential influencing factors and their corresponding lead times, the following MLR equation is established: Where k represents the current forecast time, i and j represent the size of the space, where i = 1, ..., n, j = 1, ..., m, is the intercept, is the regression coefficient of each component on SIC.

3. The method for spatial prediction of sea ice density based on the MLR model according to claim 1, characterized in that: The SIC data of the past 12 months are included in the predictors respectively, and the prediction model of the SIC data is improved as follows: in, is the intercept, is the regression coefficient of each component on SIC, and the regression coefficient of each grid point is different.

4. An electronic device comprising: processor; and a memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 3.

5. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute any one of the methods of claims 1 to 4.

Images