A visualization system based on the analysis of sea ice concentration forecast results

Abstract

The present invention belongs to the technical field of ocean data analysis, and relates to a visualization system based on the analysis of sea ice concentration forecast results. It includes ocean element data acquisition and storage, data processing, visualization design, data analysis, visualization display and interaction, storage of visualization charts, system optimization and maintenance. The present invention supports a marine element visualization system that can process multiple types of data including sea surface temperature, salinity, wind field, and density, is easy to use, highly customizable, and has good visualization effects, so as to help users better understand and perceive marine elements and their change trends, and provide strong support for the research of climate change and marine ecosystems.

Description

Technical Field

[0001] The present invention belongs to the technical field of marine forecast product analysis, and relates to a visualization system based on the analysis of sea ice concentration forecast results. Background Art

[0002] Sea ice concentration is an important physical quantity in the ocean and polar environments. At the same time, its spatio-temporal variation is also one of the very key research directions in ocean science. By using modern remote sensing technology and ocean observation data, sea ice concentration data at different times and locations can be obtained, and then the spatio-temporal variation law can be grasped by visualizing the forecast results. This helps us to deeply understand the formation and distribution mechanism of sea ice, as well as its interaction with other environmental factors (such as ocean circulation, climate change, etc.).

[0003] On the other hand, the visualization of sea ice concentration data is also one of the important means to analyze the impact of climate change. Sea ice is one of the important indicators reflecting global climate change, and the change of its distribution and density directly reflects the trend of climate change. By visually analyzing sea ice concentration data, the trend and law of climate change can be better understood, so as to provide a scientific basis for formulating response measures and evaluating their effects.

[0004] In the field of marine resource development, the visualization of sea ice concentration data also has important applications. For example, in the development fields such as oil and natural gas, it is necessary to understand the sea ice concentration and distribution in order to better plan and implement the development plan. By visualizing the sea ice concentration in different regions, the resource potential and development risks of the region can be better understood, thus providing a scientific basis and reference for resource development.

[0005] In summary, it can be seen the important position of sea ice concentration data visualization in related fields. At the same time, there is an urgent need in the industry for a visualization system for sea ice concentration data that is efficient and supports personalized modification. Summary of the Invention

[0006] To overcome the deficiencies of the prior art, the present invention aims to provide an efficient, highly referential, and personalized-modification-supportive visualization system for analyzing sea ice concentration forecast results. This system directly provides a high-quality intelligent Arctic sea ice forecast product based on deep learning, as well as the climatological forecast and inertial forecast widely used in the industry. The present invention can not only analyze the forecast products of users, but also provide forecast products with considerable reference value for comparison or direct use. In addition, this system can also access the user's own forecast model and provide various training sets and test sets. The present invention can provide visualization systems in two versions, Python and Matlab, for users to connect their own forecast models. Through the system described in the present invention, users can quickly generate visual and operable visualization images of sea ice concentration data, better understand the spatio-temporal variation law of sea ice concentration, and provide a scientific basis for decision-making and practice in various fields.

[0007] The above object of the present invention is achieved by the following technical solutions: A visualization system for analyzing sea ice concentration forecast results, characterized in that it includes a data acquisition module, a region selection module, a data storage module, a visualization preview module, a forecast display module, and a data analysis module; the data acquisition module, the region selection module, the data storage module, and the visualization preview module are connected in sequence, and the forecast display module and the data analysis module are respectively connected to the visualization preview module;

[0008] The data acquisition module: is used to support the acquisition of multiple sea ice data sources, including satellite remote sensing data, ocean observation station data, simulation data, and user-defined uploads;

[0009] The region selection module: is used to support multiple region selection methods, including manual selection, automatic selection, and import selection; at the same time, this module is also used to edit and save regions for the convenience of subsequent use by users;

[0010] The data storage module: is used to store the finally generated images in the user-defined storage path and can choose whether to generate them in batches; this module supports the storage of multiple image formats, including PNG, JPEG, and SVG; at the same time, this module is also used to compress and encrypt the images to ensure the security and privacy of the images;

[0011] The visualization preview module: is used to generate images according to the data set selected by the user according to the system default recommended parameters and display them to the user; this module supports the following several types of images: Spatial distribution map: This type of image can show the distribution of sea ice in different regions; Time series graph: This type of image can show the change trend of sea ice in different time periods, reflecting the sea ice coverage rate and sea ice drift;

[0012] The prediction display module: It is used to support a variety of image editing and processing tools, including cropping, rotation, scaling, and filters; at the same time, this module also supports a variety of interaction methods, including gestures and keyboard shortcuts.

[0013] The data analysis module: It is used to make a prediction after selecting a dataset, a sea area, and a user-selected prediction method by the user, obtain the prediction result and conduct an analysis, and provide a prediction product or a prediction product evaluation to the user.

[0014] In the region selection module, the data information of the selected region must be included in the dataset.

[0015] The implementation plan and technical route of the system are as follows:

[0016] The first step:

[0017] After the user selects to connect their own prediction mode to the system, in the data acquisition module, the user can select the required dataset through simple operations; in addition to the sea ice concentration datasets of the European Centre, the National Snow and Ice Data Center, and the University of Bremen in Germany provided by default in the system, this module also supports online datasets and local sea ice concentration datasets uploaded by the user.

[0018] The second step:

[0019] In the region selection module, the user can select the sea area range for which they want to conduct a prediction analysis; specifically, the user needs to input the boundary values of the sea area, including the north-south boundary values and the east-west boundary values; among them, the north-south boundary values can be customized with the specific degrees of south latitude or north latitude, and the east-west boundary values can be customized with the specific degrees of east longitude or west longitude.

[0020] The user can select different sea area ranges according to actual needs to more accurately study the distribution and changes of sea ice; at the same time, this module also supports the user to select multiple sea area ranges for comparison and analysis to more comprehensively understand the spatio-temporal distribution law of sea ice.

[0021] After selecting the sea area range, this module can also check and verify the sea area range to ensure that the selected sea area range meets the actual situation and research needs; at the same time, this module can also provide a visual display function of the sea area range to facilitate the user to more intuitively understand and select the sea area range.

[0022] The third step:

[0023] When saving an image, the user can select the image format and quality to be saved and customize the local storage path for saving; at the same time, this module also supports the user to save images in batches to more efficiently conduct data analysis and research.

[0024] When sharing images, users can select the sharing platforms and methods and customize the shared content and format. At the same time, this module also supports users to share images in batches for more efficient data analysis and research.

[0025] Step 4:

[0026] In the visualization preview module, the system generates visualization images of the selected area and the forecast days according to the default parameters and stores them in the specified path. At the same time, the generated images are displayed in the interface, and the images support users to perform interactive operations such as zooming in, zooming out, and dragging, so that users can more intuitively understand and analyze the spatio-temporal distribution law of sea ice concentration.

[0027] When generating images, the system automatically generates high-quality and visual images according to the sea area range and other parameters selected by the user. At the same time, the system can also adjust the parameters and algorithms according to the user's needs and feedback to improve the accuracy and interpretability of the images.

[0028] When displaying images, the system provides a variety of interactive operations and tools to facilitate users to perform data analysis and research more conveniently. Users can view the sea ice concentration in different areas through zooming in, zooming out, and dragging operations. Users can also perform data analysis and comparison through tools such as annotation and measurement. At the same time, the system also supports users to customize the styles and parameters of the images to more personalized display and analyze data.

[0029] Step 5:

[0030] If users are not satisfied with the images generated by the system by default or have more detailed modification intentions, the data visualization preview module also supports further customization. Users can modify various parameters of the images to meet different research needs and personalized requirements.

[0031] The parameters that can be modified include:

[0032] The style of the Colorbar and its scale: Users can customize the colormap, size, and position parameters of the Colorbar, as well as the numerical values, labels, colors, and font parameters of the scale to meet different data analysis needs.

[0033] The thickness of the coastline: Users can customize the thickness of the coastline to meet different map display needs.

[0034] The colors of land and sea ice: Users can select different color, texture, and transparency parameters to meet different map display needs.

[0035] Projection method: Users can select different projection methods, such as azimuthal projection and cylindrical projection, to meet different map display needs.

[0036] Title and subtitle text and style: Users can customize the title text, font, size, and color parameters to meet different image display requirements;

[0037] Subtitle text and style: Users can customize the subtitle text, font, size, and color parameters to meet different image display requirements;

[0038] Step 6:

[0039] The system can analyze and compare the forecast results, with the aim of showing the accuracy of the forecast products and the analysis results to the user;

[0040] The commonly used forecasting methods are: inertial forecast and climatological forecast. The climatological forecast takes the average value of the time at that time in previous years as the forecast result, and adds the anomaly value at the initial time to the climatological average value at the forecast time as the inertial forecast.

[0041] At the same time, the system also provides an Arctic sea ice intelligent forecast product based on deep learning, which performs data analysis on the forecast results to calculate the Pearson correlation coefficient PCC, root mean square error RMSE, anomaly correlation coefficient ACC, mean absolute error MAE, and forecast skill score SS evaluation indicators, and objectively analyzes and evaluates the forecast product:

[0042] RMSE calculation formula:

[0043]

[0044] PCC calculation formula:

[0045]

[0046] ACC calculation formula:

[0047]

[0048] MAE calculation formula:

[0049]

[0050] SS calculation formula:

[0051]

[0052] Where N is the number of test samples, MM is the number of spatial grid points, Truei and Predicti are the observed value and predicted value of the i-th sample, the anomaly true value AT and the anomaly predicted value AP are the actual value and predicted value after removing the climate state, and Rereference is the comparison reference field; in addition, when SS > 0, it indicates that the prediction result is better than the reference field; when SS = 1, it indicates a prediction without error; SS < 0 indicates that the prediction result is worse than the reference field.

[0053] When the forecast time is 100 d, the root mean square error RMSE, the correlation coefficient PCC, and the anomaly correlation coefficient ACC of the product are 0.2, 0.77, and 0.74 respectively; compared with the prediction results of the hybrid coordinate ocean model, SS is always greater than 0, and the average SS for 9 days is 0.7172, which has considerable reference value.

[0054] The present invention has the following beneficial effects:

[0055] Provide a scientific basis for climate change and ecological protection: The visualization system for analyzing sea ice concentration forecast results can help scientists better understand the changing trend of sea ice concentration and provide a scientific basis for climate change and ecological protection. For example, by observing the 100-day sea ice concentration forecast result image, scientists can better understand the form, distribution, and changes of sea ice, so as to better grasp the future sea ice change trend and provide a scientific basis for coping with climate change and ecological protection.

[0056] Promote scientific communication and cooperation: The visualization system for analyzing ice concentration forecast results can help scientists better display and share their research results, and promote scientific communication and cooperation. Scientists can generate visualization images of ice concentration forecast results analysis with good visibility and operability through this system and share them with other scientists, thus promoting scientific cooperation and joint research.

[0057] Improve resource development decisions: The visualization system for analyzing ice concentration forecast results can help resource development decision-makers better understand the distribution and change law of sea ice concentration, so as to better plan and implement resource development plans. For example, with the visualization system for analyzing ice concentration forecast results, decision-makers can better understand the distribution and change law of sea ice, provide a scientific basis for resource development decisions, and avoid damaging the marine ecological environment during resource development.

[0058] Providing services for ocean engineering and maritime transportation: The sea ice concentration forecast result analysis visualization system can help relevant personnel in ocean engineering and maritime transportation better understand the distribution and variation rules of sea ice concentration, so as to better plan and implement ocean engineering and maritime transportation plans. For example, when navigating at sea or carrying out ocean engineering construction, the sea ice concentration visualization system can help personnel better understand the distribution and variation trend of sea ice, thus avoiding damage to ships or facilities. Brief Description of the Drawings

[0059] Figure 1 It is the system composition and flowchart of the present invention.

[0060] Figure 2 It is the interface diagram of the data visualization preview module and the image personalized modification module of the present invention. Detailed Description of the Invention

[0061] The present invention will be described in detail below with reference to the drawings and specific embodiments, but the protection scope of the present invention is not limited.

[0062] Embodiment 1

[0063] Figure 1 It is the system composition and flowchart of the present invention. A visualization system based on the analysis of sea ice concentration forecast results includes a data acquisition module, a region selection module, a data storage module, a visualization preview module, a forecast display module, and a data analysis module; the data acquisition module, the region selection module, the data storage module, and the visualization preview module are connected in sequence, and the forecast display module and the data analysis module are respectively connected to the visualization preview module;

[0064] The data acquisition module: It is used to support the acquisition of multiple sea ice data sources, including satellite remote sensing data, ocean observation station data, simulation data, and user-defined uploads;

[0065] The region selection module: It is used to support multiple region selection methods, including manual selection, automatic selection, and import selection; at the same time, this module is also used to edit and save regions for the convenience of subsequent use by users; the data information of the selected region must be included in the dataset;

[0066] The data storage module: It is used to store the finally generated images in the user-defined storage path and can choose whether to generate them in batches; this module supports the storage of multiple image formats, including PNG, JPEG, and SVG; at the same time, this module is also used to compress and encrypt the images to ensure the security and privacy of the images;

[0067] The visualization preview module: It is used to generate an image according to the dataset selected by the user according to the system default recommended parameters and display it to the user; this module supports the following types of images: Spatial distribution map: This type of image can show the distribution of sea ice in different regions; Time series graph: This type of image can show the change trend of sea ice in different time periods, reflecting the sea ice coverage rate and sea ice drift;

[0068] The forecast display module: It is used to support a variety of image editing and processing tools, including cropping, rotating, scaling, and filters; at the same time, this module also supports a variety of interaction methods, including gestures and keyboard shortcuts;

[0069] The data analysis module: It is used to perform a forecast after selecting a dataset, sea area, and the user's self-selected forecast method, obtain the forecast result and analyze it, and provide a forecast product or a forecast product evaluation to the user.

[0070] Example 2

[0071] The implementation plan and technical route of the system described in Example 1 are as follows:

[0072] The first step:

[0073] After the user selects to connect their own forecast mode to the system, in the data acquisition module, the user can select the required dataset through simple operations; in addition to the sea ice concentration datasets of the European Centre, the National Snow and Ice Data Center, and the University of Bremen in Germany provided by default in the system, this module also supports online datasets and local sea ice concentration datasets uploaded by the user;

[0074] The second step:

[0075] In the area selection module, the user can select the sea area range for which they want to perform a forecast analysis; specifically, the user needs to input the boundary values of the sea area, including the north-south boundary values and the east-west boundary values; among them, the north-south boundary values can customize the specific degrees of south latitude or north latitude, and the east-west boundary values can customize the specific degrees of east longitude or west longitude;

[0076] The user can select different sea area ranges according to actual needs to more accurately study the distribution and changes of sea ice; at the same time, this module also supports the user to select multiple sea area ranges for comparison and analysis to more comprehensively understand the spatio-temporal distribution law of sea ice;

[0077] After selecting the sea area range, this module can also check and verify the sea area range to ensure that the selected sea area range meets the actual situation and research needs; at the same time, this module can also provide a visual sea area range display function to facilitate the user to more intuitively understand and select the sea area range;

[0078] Step 3:

[0079] When saving an image, the user can select the image format and quality to be saved and customize the local storage path for saving; meanwhile, this module also supports the user to save images in batches to facilitate more efficient data analysis and research;

[0080] When sharing an image, the user can select the platform and method of sharing and customize the content and format of sharing; meanwhile, this module also supports the user to share images in batches to facilitate more efficient data analysis and research;

[0081] Step 4:

[0082] In the visualization preview module, the system will generate a visualization image of the selected area and the number of forecast days according to the default parameters and store it in the specified path; meanwhile, the generated image will be displayed in the interface, and the image supports the user to perform zoom-in, zoom-out, and drag-and-drop interaction operations to facilitate the user to more intuitively understand and analyze the spatio-temporal distribution law of sea ice concentration;

[0083] When generating an image, the system will automatically generate a high-quality, visual image according to the sea area range and other parameters selected by the user; meanwhile, the system can also adjust the parameters and algorithms according to the user's needs and feedback to improve the accuracy and interpretability of the image;

[0084] When displaying an image, the system will provide a variety of interaction operations and tools to facilitate the user to more conveniently conduct data analysis and research; the user can view the sea ice concentration in different areas through zoom-in, zoom-out, and drag-and-drop operations; the user can also use tools such as annotation and measurement for data analysis and comparison; meanwhile, the system also supports the user to customize the style and parameters of the image to more personalized display and analyze data;

[0085] Step 5:

[0086] If the user is not satisfied with the image generated by the system by default or has a more detailed modification intention, the data visualization preview module also supports further customization; the user can modify various parameters of the image to meet different research needs and personalized requirements;

[0087] The parameters that can be modified include:

[0088] The style of the Colorbar and its scale: The user can customize the colormap, size, and position parameters of the Colorbar, as well as the numerical value, label, color, and font parameters of the scale to meet different data analysis needs;

[0089] The thickness of the coastline: The user can customize the thickness of the coastline to meet different map display needs;

[0090] Land and sea ice colors: Users can select different colors, textures, and transparency parameters to meet different map display requirements;

[0091] Projection mode: Users can choose different projection modes, such as azimuthal projection and cylindrical projection, to meet different map display requirements;

[0092] Title and subtitle text and style: Users can customize the title text, font, size, and color parameters to meet different image display requirements;

[0093] Subtitle text and style: Users can customize the subtitle text, font, size, and color parameters to meet different image display requirements;

[0094] Step 6:

[0095] The system can analyze and compare the forecast results, with the aim of showing the accuracy of the forecast products and the analysis results to the user;

[0096] The commonly used forecasting methods are: inertial forecast and climatological forecast. The climatological forecast takes the average value of the time at that time in previous years as the forecast result, and adds the anomaly value at the initial time to the climatological average value at the forecast time as the inertial forecast.

[0097] At the same time, the system also provides an Arctic sea ice intelligent forecast product based on deep learning, which performs data analysis on the forecast results to calculate the Pearson correlation coefficient PCC, root mean square error RMSE, anomaly correlation coefficient ACC, mean absolute error MAE, and forecast skill score SS evaluation indicators, and objectively analyzes and evaluates the forecast product:

[0098] RMSE calculation formula:

[0099]

[0100] PCC calculation formula:

[0101]

[0102] ACC calculation formula:

[0103]

[0104]

[0105] SS calculation formula:

[0106]

[0107] Wherein, N is the number of test samples, MM is the number of spatial grid points, Truei and Predicti are the observed value and predicted value of the i-th sample, the abnormal true value AT and abnormal predicted value AP are the actual value and predicted value after deducting the climate state, and Rereference is the comparison reference field; in addition, when SS > 0, it indicates that the prediction result is better than the reference field; when SS = 1, it indicates error-free prediction; SS < 0 indicates that the prediction result is worse than the reference field.

[0108] When the forecast time is 100d, the root mean square error RMSE, correlation coefficient PCC, and abnormal correlation coefficient ACC of this product are 0.2, 0.77, and 0.74 respectively; compared with the prediction results of the hybrid coordinate ocean model, SS is always greater than 0, and the average SS for 9 days is 0.7172 ， It has considerable reference value.

[0109] The above-described embodiments are only the preferred embodiments of the present invention, rather than all the feasible embodiments of the present invention. For those of ordinary skill in the art, any obvious changes made without departing from the principle and spirit of the present invention should be considered to be included within the protection scope of the claims of the present invention.

Claims

1. A visualization system based on the analysis of sea ice concentration forecast results, characterized in that, It includes a data acquisition module, a region selection module, a data storage module, a visualization preview module, a forecast display module, and a data analysis module; the data acquisition module, the region selection module, and the data storage module are connected in sequence, and the forecast display module and the data analysis module are respectively connected to the visualization preview module; The data acquisition module: is used to support the acquisition of multiple sea ice data sources, including satellite remote sensing data, ocean observation station data, simulation data, and user-defined uploads; The region selection module: is used to support multiple region selection methods, including manual selection, automatic selection, and import selection; at the same time, this module is also used to edit and save regions for the convenience of subsequent use by users; The data storage module: is used to store the finally generated images in the user-defined storage path and can choose whether to generate them in batches; this module supports the storage of multiple image formats, including PNG, JPEG, and SVG; at the same time, this module is also used to compress and encrypt images to ensure the security and privacy of the images; The visualization preview module: is used to generate images according to the selected data set by the user in accordance with the system default recommended parameters and display them to the user; this module supports the following several types of images: Spatial distribution map: This type of image can show the distribution of sea ice in different regions; Time series graph: This type of image can show the change trend of sea ice in different time periods, reflecting the sea ice coverage rate and sea ice drift; The forecast display module: is used to support multiple image editing and processing tools, including cropping, rotating, scaling, and filters; at the same time, this module also supports multiple interaction methods, including gestures and keyboard shortcuts; The data analysis module: is used to perform forecasting after selecting the data set, sea area, and the user's self-selected forecasting method, obtain the forecasting results and conduct analysis, and provide forecasting products or forecasting product evaluations to users.

2. The visualization system based on the analysis of sea ice concentration forecast results according to claim 1, characterized in that, In the region selection module, the data information of the selected region must be included in the data set.

3. A visualization system based on the analysis of sea ice concentration forecast results according to claim 1 or 2, characterized in that The implementation plan and technical route of the system are as follows: The first step: When the user chooses to connect their own forecasting mode to the system, in the data acquisition module, the user can select the required data set through simple operations; in addition to the sea ice concentration data sets of the European Centre, the National Snow and Ice Data Center, and the University of Bremen in Germany provided by default in the system, this module can also support online data sets and locally uploaded sea ice concentration data sets by users; The second step: In the region selection module, the user can select the sea area range for which they want to conduct forecasting analysis; specifically, the user needs to input the boundary values of the sea area, including the north-south boundary values and the east-west boundary values; among them, the north-south boundary values can be customized with the specific degrees of south latitude or north latitude, and the east-west boundary values can be customized with the specific degrees of east longitude or west longitude; The user can select different sea area ranges according to actual needs to more accurately study the distribution and changes of sea ice; at the same time, this module can also support the user to select multiple sea area ranges for comparison and analysis to more comprehensively understand the spatio-temporal distribution law of sea ice; After selecting the sea area range, this module can also check and verify the sea area range to ensure that the selected sea area range meets the actual situation and research requirements. At the same time, this module can also provide a visual display function of the sea area range to facilitate users to more intuitively understand and select the sea area range. Step 3: When saving an image, the user can select the image format and quality to be saved and customize the local storage path for saving. At the same time, this module can also support the user to save images in batches to facilitate more efficient data analysis and research. When sharing an image, the user can select the sharing platform and method and customize the content and format of the sharing. At the same time, this module also supports the user to share images in batches to facilitate more efficient data analysis and research. Step 4: In the visual preview module, the system will generate a visual image of the selected area and the forecast days according to the default parameters and store it in the specified path. At the same time, the generated image is displayed in the interface, and the image supports users to perform interactive operations such as zooming in, zooming out, and dragging to facilitate users to more intuitively understand and analyze the spatio-temporal distribution law of sea ice concentration. When generating an image, the system will automatically generate a high-quality, visual image according to the sea area range and other parameters selected by the user. At the same time, the system can also adjust the parameters and algorithms according to the user's needs and feedback to improve the accuracy and interpretability of the image. When displaying an image, the system will provide a variety of interactive operations and tools to facilitate users to more conveniently conduct data analysis and research. Users can view the sea ice concentration in different areas through zooming in, zooming out, and dragging operations. Users can also use tools such as annotation and measurement for data analysis and comparison. At the same time, the system also supports users to customize the style and parameters of the image to more personalized display and analyze data. Step 5: If the user is not satisfied with the image generated by the system by default or has a more detailed modification intention, the data visualization preview module also supports further customization. The user can modify a variety of parameters of the image to meet different research needs and personalized requirements. The parameters that can be modified include: The style of the Colorbar and its scale: The user can customize the colormap, size, and position parameters of the Colorbar, as well as the numerical value, label, color, and font parameters of the scale to meet different data analysis needs. The thickness of the coastline: The user can customize the thickness of the coastline to meet different map display needs. The colors of land and sea ice: The user can select different color, texture, and transparency parameters to meet different map display needs. Projection method: The user can select different projection methods, including azimuthal projection and cylindrical projection, to meet different map display needs. The text and style of the title and subtitle: The user can customize the text, font, size, and color parameters of the title to meet different image display needs. The text and style of the subtitle: The user can customize the text, font, size, and color parameters of the subtitle to meet different image display needs. Step 6: The system can analyze and compare the forecast results, with the aim of showing the accuracy of the forecast products and the analysis results to the user; The commonly used forecasting methods are: inertial forecast and climatological forecast. The climatological forecast takes the average value of the time at that time in previous years as the forecast result, and adds the anomaly value at the initial time to the climatological average value at the forecast time as the inertial forecast. At the same time, the system also provides an Arctic sea ice intelligent forecast product based on deep learning, which performs data analysis on the forecast results to calculate the Pearson correlation coefficient PCC, root mean square error RMSE, anomaly correlation coefficient ACC, mean absolute error MAE, and forecast skill score SS evaluation indicators, and objectively analyzes and evaluates the forecast product: RMSE calculation formula: PCC calculation formula: ACC calculation formula: MAE calculation formula: SS calculation formula: Where N is the number of test samples, MM is the number of spatial grid points, Truei and Predicti are the observed value and predicted value of the i-th sample, the anomaly true value AT and the anomaly predicted value AP are the actual value and predicted value after deducting the climate state, and Rereference is the comparison reference field. In addition, when SS>0, it means that the prediction result is better than the reference field; when SS=1, it means there is no error prediction; SS<0 means that the prediction result is worse than the reference field.

4. A visualization system based on the analysis of sea ice concentration forecast results according to claim 3, characterized in that, When the forecast time is 100 days, the forecast root mean square error RMSE, correlation coefficient PCC and anomaly correlation coefficient ACC of the product are 0.2, 0.77 and 0.74 respectively; compared with the prediction results of the hybrid coordinate ocean model, the SS is always greater than 0, and the 9-day average SS is 0.7172, which has considerable reference value.

Images