180 results about "Sea surface temperature" patented technology

The invention discloses a near-shore sea surface temperature fusion method based on deep learning driven variational analysis, and belongs to the technical field of data processing, and the method specifically comprises the steps: inputting historical sea surface temperature data after variational analysis into a deep learning sea surface temperature prediction model, outputting a predicted sea surface temperature in a future specified time period as a predicted background field of current variation analysis; based on a deep learning sea surface temperature prediction model, establishing a relationship between a multi-step sea surface temperature prediction difference value and a prediction error, generating a background error variance during fusion, and fusing a spatial distance function and a short-term time correlation function to construct a background error covariance model during variational analysis; assimilating sea surface temperature observation data collected in real time to the prediction background field, solving an optimal analysis field and adjusting the resolution by combining the background error covariance model and the observation error weight, and outputting a near-shore area sea surface temperature analysis field; according to the invention, a sea surface temperature analysis field with high precision and fine scale characteristics is provided for a near-shore area.

A method for studying interannual variation of marine heatwaves in the South Indian Ocean includes: investigating spatial distribution features and linear variation trend features of marine heatwaves in the South Indian Ocean according to obtained data; analyzing the relationship between an intensity of marine heatwaves in the South Indian Ocean and an El Niño-Southern Oscillation (ENSO) process; analyzing a spatial evolution of marine heatwaves in the South Indian Ocean and sea surface temperature (SST) anomalies during an El Niño event; analyzing energy contribution of ocean and atmosphere to marine heatwave events during the El Niño event; investigating physical mechanisms of the marine heatwaves in the South Indian Ocean; and determining whether the marine heatwaves in the South Indian Ocean are influenced by shortwave radiation and latent heat modulated by the ENSO process. The present disclosure reveals that marine heatwaves occur during the El Niño event in the South Indian Ocean.

The invention discloses an ocean three-dimensional temperature and salt field prediction method based on remote sensing data driving, and the method comprises the steps: firstly constructing a satellite remote sensing sea surface temperature data set, a sea surface height anomaly data set and an ocean three-dimensional reanalysis data set which are needed by an experiment; then, on the basis of a convolutional neural network and a Transform module, a basic framework of the model is constructed according to the temporal and spatial variation characteristics of the remote sensing data and the temporal and spatial variation characteristics of the three-dimensional temperature and salt field; the model adopts an end-to-end design mode, and adapts to planar spatial-temporal variation of historical remote sensing data through a two-dimensional convolutional coding structure; the three-dimensional spatial-temporal change of a future three-dimensional temperature and salt field is concerned through a three-dimensional convolution decoding structure; an attention mechanism and a feature evolution module are designed according to the spatial-temporal features of the two kinds of data, and the change relation of the remote sensing data and the temperature and salt field data under different spatial-temporal dimensions is learned. And finally, inputting a mode of combining satellite remote sensing temperature of historical n days and sea surface height anomaly into the trained model to obtain three-dimensional temperature and salt field data of future n days.

The invention belongs to the technical field of sea temperature complementation, and discloses a sea temperature complementation method and system based on an asynchronous diffusion Schrodinger bridge, and the method comprises the steps: firstly, generating a weight anomal which reflects the abnormal degree of a pixel through a preprocessing step S1, so as to guide a subsequent diffusion process; s2, establishing a bidirectional diffusion path between the initial complementation image and the real image based on a diffusion Schrodinger bridge theory, and dynamically adjusting a diffusion coefficient according to anomay to generate an intermediate state xt; predicting a score function pred through a U-Net network S3, and reconstructing a current image x0 for updating a state or calculating loss; model parameters are optimized through iteration during training, multi-round denoising reconstruction is carried out during inference, and finally a complete high-quality sea surface temperature image SSTrecon is output. According to the invention, local details are fully reserved, and the accuracy of image completion is improved.

The invention belongs to the technical field of image processing, and particularly relates to a time sequence frequency domain feature enhanced diffusion-based sea temperature image completion method and system, and the method comprises the following steps: inputting a damaged sea temperature image, an initialized cloud mask, a weekly average sea temperature image and a historical sequence sea temperature image into a time sequence frequency domain feature extraction module for processing; and mapping into fused frequency domain features, and outputting a complex frequency domain condition vector. A real SST image on the current day is coded into an initial latent variable through a latent space enhancement diffusion module, a complete noisy latent variable is generated through forward diffusion sampling, a denoised latent variable is obtained through backward stable diffusion sampling, after the denoised latent variable is decoded into a pixel field through an output reconstruction module, constraint post-processing is carried out in combination with a mask and a damaged sea temperature image, and a real SST image is obtained. And finally outputting a reconstructed image. And the damaged sea temperature image completion precision and the time sequence continuity are improved.

The invention discloses a fishery sea area monitoring system based on satellite remote sensing, and belongs to the technical field of fishery monitoring, and the system specifically comprises a remote sensing data acquisition module which obtains and preprocesses sea surface temperature, chlorophyll concentration and sea surface height abnormal data; the buoy data acquisition module synchronously acquires temperature-salinity profile data of the Argo buoy array, and the layout density of the Argo buoy array is dynamically adjusted along with historical fish catch and ocean frontal surface intensity; the space-time alignment module divides a time window through an orbital period, corrects space offset in combination with a buoy drift trajectory, sets a time buffer area based on a georotation flow rate and a satellite image resolution, and generates a space-time unified three-dimensional marine environment data set; the fishery suitability calculation module constructs a biophysical coupling model, correlates surface chlorophyll and subsurface thermohaline structures, and calculates the suitability index of each water layer; the fishery division module intelligently defines a core area and an edge area according to the index three-dimensional distribution characteristics; according to the invention, accurate fishery identification based on multi-source data fusion is realized.

The invention provides a space-time intelligent ocean satellite internet of things perception information screening method and device, and belongs to the technical field of data processing. Marine environment data space-time prediction is carried out based on a block time sequence converter, marine hot spot area mining is carried out based on a density clustering algorithm of mean shift, and marine environment data comprises sensing data of sea surface temperature, salinity, wave height and the like; marine environment risk assessment is carried out based on a BP neural network, information screening feedback control is carried out based on multi-agent reinforcement learning, the efficient information processing requirement of a complex marine environment detection scene is met, and the low-redundancy and high-precision data acquisition and transmission adaptive feedback control effect is achieved. The method can be directly applied to an ocean observation and detection facility networking system in the ocean Internet of Things, can provide powerful support in the aspects of military application and civil application, and has wide and important application prospects and value.

The invention belongs to the technical field of image processing, and particularly relates to a sea temperature complementation method and system based on recursive double-flow Mama, and the method comprises the following steps: splicing a damaged SST image and a weekly average SST image as input, and outputting a predicted weekly average SST image and a predicted abnormal SST image through N times of recursive iteration of N same recursive hierarchical Mama blocks, and the two output images are added to obtain a final complemented SST image. According to the method, a recursive hierarchical Mama block for a sea surface temperature completion task is set, two parallel layers, namely a stable information representation module and an abnormal information representation module, are integrated in each block, features related to stability and features related to anomalies are extracted respectively, long-range dependency relationship modeling under large-area deficiency is achieved, and completion accuracy is improved.

The invention provides a cloud condition adaptive geostationary satellite sea surface temperature inversion method based on deep learning, and relates to the technical field of sea surface temperature inversion, and the method specifically comprises the steps: obtaining geostationary satellite observation data, geographic information data, polar orbit satellite sea surface temperature and cloud products, and analyzing the sea surface temperature and atmospheric background data; performing standardization preprocessing on the obtained stationary satellite observation data; constructing a cloud detection data set; a cloud detection model of deep learning is constructed and trained; constructing a clear sky data set; a clear sky inversion model of deep learning is constructed and trained; constructing a data set for under-cloud inversion model training; constructing an under-cloud inversion model of deep learning and training the under-cloud inversion model; and integrating the cloud detection model, the clear sky inversion model and the under-cloud inversion model, adaptively calling the corresponding model according to a real-time cloud detection result, and generating a final sea surface temperature product. According to the technical scheme, the problem that in the prior art, the accuracy of a reconstructed product on the spatial structure cannot be comprehensively reflected is solved.

The invention discloses a sea surface temperature prediction method and system. The method comprises the steps of obtaining first historical data, and performing first preprocessing on the first historical data; performing first decomposition on the first historical data after the first preprocessing to obtain first decomposed data; and taking the first decomposition data as the input of a pre-trained first prediction model, and performing sea surface temperature prediction according to the output of the first prediction model. The accuracy and efficiency of sea surface temperature prediction can be improved through advanced data processing and analysis technologies. By using the deep learning model, a complex mode in historical data can be learned, and the future sea surface temperature change can be accurately predicted. In addition, the method can adapt to specific conditions of different sea areas, provides customized prediction services, and meets the requirements of different users.

The invention discloses a sea surface temperature prediction method and device based on a multivariable time series, and belongs to the field of meteorological ocean data processing and prediction.The sea surface temperature prediction method comprises the steps that sea surface temperature and meteorological factor data are collected from multi-source observation data of a target area, the collected data are preprocessed, and cleaned time series data are obtained; building a sea surface temperature prediction model based on a multivariable time sequence, wherein the sea surface temperature prediction model comprises a variable embedding layer, a lag feature embedding layer, a Transform encoder, a linear modeling layer, an output layer and a RevIN module for performing normalization and de-normalization processing on non-stationary data; training the established sea surface temperature prediction model by using the preprocessed data; and predicting the day-by-day sea surface temperature of the future 30 days by adopting the trained sea surface temperature prediction model. According to the invention, the nonlinear dynamic change of the sea surface temperature can be effectively captured, and the prediction precision of the sea surface temperature is significantly improved.

The invention belongs to the technical field of ocean observation, and discloses an underwater sound velocity distribution super-resolution construction method based on multi-modal data fusion. According to the method, a sound velocity distribution forecasting model is constructed, the model carries out temperature preliminary reconstruction on original space area temperature data through a Unet neural network, and sea surface temperature features are extracted through a Mama neural network; correcting the preliminarily reconstructed temperature data through sea surface temperature features to obtain high-spatial-resolution temperature data; and finally, obtaining high-spatial-resolution sound velocity distribution data through a sound velocity distribution calculation module. According to the method, the underwater temperature and the sea surface temperature are subjected to data fusion, so that the sound velocity distribution of the whole space area is quickly and accurately forecasted under the condition of original low-spatial-resolution data, and the problem of insufficient spatial resolution of ocean sound velocity distribution estimation is solved.

The invention relates to the technical field of sea surface temperature prediction, and discloses a sea surface temperature prediction method based on multi-scale frequency domain enhancement, which comprises the following steps: preprocessing a sea surface temperature data set obtained by ERA5 reanalysis data, and dividing the sea surface temperature data set into a training set, a verification set and a test set; inputting data of the training set into a deep learning model MFE-ConvLSTM for training, adjusting parameters by using the verification set, and finally performing evaluation by using the test set; the deep learning model MFE-ConvLSTM is composed of a multi-scale frequency domain enhancement module and a convolutional long and short term memory network module; and preprocessing sea surface temperature data of a sea area to be predicted, and inputting the sea surface temperature data into the qualified model for sea surface temperature prediction. According to the method disclosed by the invention, the complex spatial-temporal change characteristics of the sea-land junction are effectively captured through a multi-scale frequency domain enhancement mechanism, so that the prediction stability in the whole sea area range is remarkably improved while the local prediction precision is improved.

The invention discloses a seawater parameter correlation analysis method and device, electronic equipment and a storage medium, and relates to the technical field of ocean monitoring. The method comprises the following steps: acquiring satellite remote sensing data of a target sea area and performing space-time quality screening; preprocessing the satellite remote sensing data to generate correction data for temperature inversion; analyzing tidal dynamic characteristics of the target sea area; inverting a sea surface temperature field based on the pre-processed data, and performing enhancement analysis by fusing tidal dynamic characteristics; establishing a coupling strength quantitative index of the tidal power and the temperature field; collecting and complementing historical observation data of the power plant; the relevance between the temperature and the liquid level parameters is counted and analyzed; and the correlation of the seawater parameters is judged by integrating the coupling strength quantitative index and the data statistical result. The invention provides a two-channel decision-making mechanism with cooperation of a physical mechanism and operation data statistics, and the mechanism not only improves the operation efficiency of the cooling system of the power plant, but also reduces the misjudgment rate of abnormal working conditions, and provides an enforceable optimization strategy for warm drainage management.

The invention discloses a sea surface temperature image diffusion completion method and system based on dual-stage fusion constraint, and the method comprises the steps: firstly constructing sea area multi-source priori knowledge features to guide the reconstruction of sea surface temperature through integrating three types of heterogeneous data of geography, space-time and power; deep fusion is carried out on the sea area multi-source priori knowledge features through a cross-modal feature fusion module to obtain fused multi-source priori knowledge, and the fused multi-source priori knowledge is fused with structural knowledge to form comprehensive condition features used for guiding the generation process of a conditional diffusion model; and finally, realizing refined reconstruction of the missing region based on a conditional diffusion model, and outputting a completion result. And obtaining a final complete image according to the complementation result, the sea surface temperature image to be complemented and the missing mask. According to the invention, the accuracy of sea surface temperature image completion is improved.

The invention provides a method for cooperatively regulating and controlling chlorophyll-a concentration variability through vortex-upwelling, and relates to the technical field of ocean remote sensing and environmental monitoring. A multi-source satellite and reanalysis data set is obtained, and a stationary time sequence is constructed; identifying a dominant variation period in chlorophyll-a concentration, sea surface temperature and sea surface height anomalies, and extracting a monthly abnormal value sequence; performing complex empirical orthogonal function analysis on the sea surface height anomaly, extracting a dominant spatial variation mode of the sea surface height anomaly, and identifying a vortex distribution position and form; classifying and screening target vortexes according to vortex types and seasons, carrying out normalization processing on the chlorophyll-a concentration, aligning to vortex centers, and analyzing a spatial response relationship between the vortexes and the chlorophyll-a concentration; and drawing a density time-depth profile map and a CHL-SLA composite map by combining surface chlorophyll-a concentration, sea surface height anomaly and vertical Argo buoy data, and analyzing an influence mechanism of vortex-upwelling on chlorophyll-a horizontal distribution.

The invention relates to a three-dimensional sea temperature monitoring method, and the method comprises the steps: predicting a residual error through a sea surface temperature reconstruction model based on low-resolution sea temperature background data, meteorological data and geographic feature data of a to-be-monitored sea area, superposing the residual error with the low-resolution sea temperature background data, and obtaining a high-resolution sea temperature parameter of each region of a sea surface; and the precision and the physical consistency of the sea surface temperature parameters are improved. In deep sea temperature inversion, based on biochemical environment data and high-resolution surface sea temperature parameters, sea temperature parameters of all depth areas of a sea area to be monitored are predicted through a deep sea temperature prediction model, and a heat exchange rule is reflected more accurately. According to the method, surface reconstruction and deep inversion are combined, an association connection mechanism is established, information collaborative integration is achieved, complete and coherent three-dimensional sea temperature data are output, and reliable support is provided for applications such as ocean heat transfer calculation.

The invention relates to the technical field of ocean remote sensing monitoring, in particular to a remote sensing image-based warm discharge water reference temperature self-adaptive extraction method, which specifically comprises the steps of S01, data acquisition and preprocessing, S02, cumulative histogram generation, S03, curve fitting and smoothing, S04, second derivative calculation and S05, reference temperature determination. According to the method, the reference temperature is determined by analyzing the statistical distribution characteristics of the sea surface temperature data field, namely the form of the accumulative histogram, the physical demarcation point of the background water body and the warm drainage area on the temperature distribution is automatically positioned by using the second derivative, the process is completely driven by data, the interference of subjective factors is eliminated, and the accuracy of temperature distribution is improved. The method can automatically adapt to water areas with different geographical forms and different warm drainage diffusion modes, and objectivity and universality of reference temperature extraction are achieved.

The invention discloses a method, system and device for constructing a sea surface temperature multi-source data completion fusion model based on SwinGAN, and belongs to the technical field of data processing, and the method achieves the high-precision downscaling of sea surface temperature through the integration of field measured data, numerical reanalysis data and satellite data. An improved Swindow-UNet is adopted as a generator, and a window attention mechanism of the improved Swindow-UNet is utilized to effectively capture multi-scale spatial-temporal characteristics of the sea temperature field; and the discriminator introduces conditional adversarial learning and gradient penalty strategies to ensure the physical consistency of the generated data and the high-resolution target. According to the method, the RMSE of the generated data and the real data in the 2.5-time downscaling multi-source data fusion task in the Atlantic Ocean sea area reaches 1.1, and the spatial-temporal characteristics of the ocean temperature can be accurately reproduced. According to the method, the multi-source sea surface temperature data are fused, the physical rationality and time-space continuity of the data are ensured, and higher-quality data support is provided for the fields of marine environment monitoring, climate prediction, fishery resource management and the like.

The present invention provides a method for extracting ocean vortex sea surface temperature and salinity anomalies, comprising the following steps: Step 1, data collection and preprocessing; Step 2, identifying vortex features; Step 3, constructing the background fields of vortex sea surface temperature and salinity; Step 4, extracting the anomaly fields of vortex sea surface temperature and salinity. The present invention extracts the observed sea surface temperature and salinity data within the vortex environment area, and conducts regression analysis in combination with the climatological sea surface temperature and salinity data to construct a more reasonable background temperature and salinity field. This method can adapt to the temperature and salinity changes in different seasons and different regions, improve the self-adaptability of the background temperature and salinity field, and avoid the errors caused by simply using the multi-year average climatological data.

The invention relates to the technical field of temperature forecasting and machine learning, and provides a sea surface temperature forecasting statistical correction method based on machine learning, and the method comprises the following steps: obtaining sea surface temperature forecasting data of a target area and sea surface temperature observation data of the target area; constructing a statistical correction model; a neural network used for calculating model parameters of a statistical correction model is constructed, features reflecting the position of a target area and the sea surface temperature of the target area are input into the neural network, and the neural network outputs the model parameters of the statistical correction model; the sea surface temperature forecast data, the sea surface temperature observation data and the model parameters are input into the statistical correction model, and the statistical correction model calculates a sea surface temperature forecast statistical correction result based on the model parameters, the sea surface temperature forecast data and the sea surface temperature observation data; by adopting the method, the sea surface temperature forecasting error can be reduced, and the forecasting precision is improved.

The invention provides a Bohai sea temperature prediction model based on a fusion model, the previous research mainly carries out prediction through single models, and the single models have corresponding limitations more or less. In order to solve the problem, a Bohai sea temperature prediction model based on a fusion model is provided. According to the method, through fusion of an Informer model and an LSTM model, and through feature optimization and XGBoost correction, the model obtains better prediction performance. According to the model, the prediction precision of the sea surface temperature of the semi-closed sea area is improved, and high-reliability data support is provided for ocean decisions such as storm surge early warning, fishing ground scheduling and ecological management.

The invention relates to the field of image processing, discloses a marine environment change monitoring method and device based on a remote sensing image and a medium, and provides a powerful tool for decision support by visually displaying risk distribution in a pseudo-color image form. Comprising the following steps: acquiring an original sea surface temperature image, performing fractional order differential enhancement processing, generating an enhanced temperature gradient map, extracting a frontal surface topological feature matrix, extracting a chlorophyll concentration map by using a multispectral remote sensing image, constructing a multiband phase coherent field, generating an ecological feature tensor field, and generating a multi-scale fusion image. And extracting a feature contour line based on the image, calculating a curvature gradient, generating a thermodynamic diagram, and outputting a marine environment dynamic risk map. According to the method, multi-source remote sensing data are fused, comprehensive, accurate and real-time monitoring of marine environment changes is realized through multi-scale analysis and feature enhancement, and powerful technical support is provided for marine environment management, disaster early warning and ecological protection.

Systems and methods for predicting decrease in average sea surface temperature (SST) are disclosed. The method includes setting a value of a first Boolean flag when at least one of an eclipse season start date and an eclipse season end date in a particular calendar year occurs in a period of thirty days preceding a timestamp for a particular partition in a plurality of partitions segmenting the particular calendar year. The method includes inputting to a trained machine learning model the first Boolean flag. The trained machine learning model is trained using a plurality of time series data using lunar orbit characteristics. The method includes predicting, using the trained machine learning model, a probability of decrease in the average SST for a predefined time interval in the particular calendar year. The method includes invoking a weather forecasting system to generate weather forecast in dependence on the predicted probability.

The invention belongs to the technical field of ocean data reanalysis space forecasting and image super-resolution, and particularly relates to a sea surface temperature space statistical downscaling method based on a U-shaped convolutional neural network, which comprises the following steps of: 1, acquiring statistical data to be analyzed and performing normalization preprocessing; 2, constructing a data set of network training, and dividing a training set and a test set; 3, constructing a U-shaped convolutional neural network; training the U-shaped convolutional neural network through the training set and the test set, and outputting the trained U-shaped convolutional neural network; 4, judging the downscaling effect of the U-shaped convolutional neural network; and step 5, acquiring real-time low-resolution ocean temperature data, preprocessing the real-time low-resolution ocean temperature data, and outputting downscaled high-resolution ocean temperature data by using the trained U-shaped convolutional neural network. According to the method, details and nonlinear features in the spatial information can be well extracted and restored, and the precision of statistical downscaling can be improved.

This invention relates to the field of meteorological services and provides a method and system for correcting subseasonal model precipitation based on climate zoning and sea surface temperature (SST) background constraints. The method includes: climate zoning, adding SST background constraints, reanalysis data modeling, model historical return data modeling, real-time prediction model matching, and prediction result correction. The system includes: a data preprocessing unit, a climate zoning unit, an SST background constraint unit, a reanalysis data modeling unit, a model historical return data modeling unit, and a real-time model matching and prediction result correction unit. This invention considers the importance of regional differences in precipitation and SST background constraints, solving the problems of significant regional differences in daily summer precipitation in my country, the tendency for overfitting in single-point modeling, and the lack of physical constraints that make correction methods unsuitable. It effectively utilizes physically meaningful predictability sources to modify the cumulative probability density correction method based on percentile mapping, thereby improving my country's summer subseasonal precipitation prediction capabilities.

The invention discloses an ocean sea surface temperature prediction method based on optimal interpolation in combination with ConvLSTM, and relates to the technical field of ocean remote sensing detection. The invention provides a sea surface temperature prediction method fusing an OI method and a ConvLSTM model, the OI method solves the problem of spatial integrity of data and provides high-quality input for ConvLSTM, and the ConvLSTM model exerts the learning ability of the ConvLSTM model on spatio-temporal characteristics on the basis of the data processed by the OI method and realizes accurate time sequence prediction; the sea surface temperature prediction method and the sea surface temperature prediction device are matched with each other, so that the method can not only ensure the continuity and the physical rationality of sea surface temperature data in space, but also effectively predict the change of the sea surface temperature data in a time sequence, and a more reliable and more accurate solution is provided for sea surface temperature prediction under the conditions that the marine environment is complex and the data is difficult to acquire.

The invention provides a training method and device of a north pole sea fog model, a storage medium and electronic equipment. The electronic equipment obtains the sea fog visibility of the research area at a plurality of historical time points by using a pre-constructed PWRF model; for each historical time point, weighting the original meteorological data, the original sea surface temperature data and the original sea ice data of the plurality of meteorological models at the historical time point to obtain sample meteorological data, sample sea surface temperature data and sample sea ice data at the historical time point; and training the to-be-trained model through the synthesized data to obtain the north pole sea fog model. Therefore, the problem that sea fog visibility data with long time sequence, large range area and high spatial resolution cannot be obtained at present is solved, and input data formed by fusing multiple models has high confidence, so that the trained north pole sea fog model can effectively capture complex mechanisms such as humidity threshold effect, interaction of wind speed and sea ice and the like; and the sea fog prediction precision is improved.

The invention discloses an ocean nearshore dissolved oxygen estimation method, system and equipment based on remote sensing data. The method comprises the following steps: S1, acquiring measured data of a dissolved oxygen concentration site of a sea area to be measured, ocean remote sensing product data and ocean numerical mode HYCOM reanalysis data; s2, performing data reconstruction on missing data of four elements including chlorophyll, granular organic carbon, suspended solids and sea surface temperature in the marine remote sensing product data by adopting a DINCAE algorithm to obtain complete marine remote sensing product data; s3, performing correlation analysis and variance expansion factor analysis on the five ocean elements related to the dissolved oxygen, selecting characteristic variables, and constructing a remote sensing estimation model of the dissolved oxygen; and S4, inverting the sea surface dissolved oxygen concentration of the sea area to be measured by adopting the optimal dissolved oxygen remote sensing estimation model, and drawing the spatial distribution of the dissolved oxygen concentration. According to the invention, the problem that the dissolved oxygen as a non-optical active substance has no characteristic response on a remote sensing spectrum and cannot be inversed directly by satellite remote sensing is solved.

The invention discloses a sea surface temperature and sea level anomaly joint prediction method based on cross attention, and the method comprises the steps: constructing a deep neural network joint prediction model, taking collected sea surface temperature SST data and sea level anomaly height SLA data as input, and predicting future SST data and SLA data; the prediction process comprises the following steps: (1) training a prediction model by adopting a training data set; (2) after each training iteration is finished, verifying the prediction model by adopting a verification data set, and storing prediction model parameters with optimal performance; and (3) testing the trained prediction model by adopting the test data set. The prediction model adopts a double-branch structure to extract spatial-temporal characteristics of SST and SLA through a convolutional long-short-term memory network, and dynamic information interaction fusion between two variables is realized by using a cross attention mechanism, so that a coupling relationship between thermodynamics and dynamics is described. The model carries out end-to-end training through a unified loss function, and continuously outputs high-precision and high-consistency prediction results in a long period.