29 results about "Satellite altimetry" patented technology

The invention relates to the technical field of satellite altimetry, in particular to a height measurement method based on a feature-enhanced multi-base integrated re-tracking model, which is characterized in that the feature-enhanced multi-base integrated re-tracking model FMERM is constructed to optimize the position of a reflection waveform re-tracking point, and the FMERM model executes the following processing processes: 1, extracting input variable features by using PCA, and extracting the input variable features by using PCA; combining with an original input variable to form an enhanced feature set; secondly, optimizing the hyper-parameters of the XGBoost model by using a grid search method, so that the hyper-parameters of the XGBoost model are optimal; thirdly, utilizing the trained model to accurately calculate a re-tracking point normalization power value of the reflected signal waveform; fourthly, the sea surface height is inverted through the geometrical relationship and direct reflection signal time delay and atmospheric correction; an effective means is provided for solving the problem that a reflection waveform re-tracking method is inaccurate, and powerful support is provided for high-precision satellite-borne GNSS-R sea surface height measurement.

The invention relates to the technical field of satellite altimetry, in particular to an altimetry method based on a residual network multi-mode deep learning model, which can remarkably improve the robustness of SSH inversion under sea condition noise interference, and is characterized in that a fusion self-attention mechanism and a residual network multi-mode deep learning model ViTResNetMDL is established, and the model ViTResNetMDL is applied to improve the robustness of the SSH inversion under sea condition noise interference. A satellite-borne GNSS-R sea surface height inversion geometric physical model is established, a ViTResNetMDL model is composed of ResNet50, ViT and MHL-NN, the first part adopts a ResNet50 convolution structure to extract effective features from an original DDM, the second part adopts a ViT layer which captures global features in an effective scattering area DDM based on a self-attention mechanism in a Transform module, the third part is composed of linear layers, and the fourth part is composed of linear layers. And using the MHL-NN to retrieve the sea surface height.

The invention provides a prior information constraint-based Antarctic ice cover material balance inversion method and device, and the method comprises the steps: obtaining satellite height measurement data, and extracting an elevation parameter of an Antarctic ice cover region; preprocessing the gravity satellite data; performing inversion material balance of physical priori constraint, including dividing the Antarctic ice cover into Mascon units which are geographically aligned with the ice flow basin, constructing a regularization constraint matrix of space differentiation based on elevation variation variance of satellite altimetry data, dynamically optimizing regularization parameters in combination with generalized cross validation criterion, and solving mass change; and outputting the Antarctic ice cover material balance result. According to the method, regularization is carried out by using the uncertainty difference of the quantized area of the height measurement data, so that the inversion precision and the spatial resolution are remarkably improved, and GIA model errors and data noise interference are effectively inhibited.

The invention discloses a satellite height measurement data optimal interpolation fusion method based on cross calibration and spectrum analysis, and the method comprises the steps: carrying out the sea surface height anomaly calculation and power spectrum analysis, and calculating the sea surface height anomaly of a satellite altimeter according to the precise orbit determination coordinates of a satellite and the distance measurement value of the altimeter; the related scale of adjacent observation points of the to-be-fused satellite altimeter is obtained through the power spectrum analysis of the sea surface height anomaly of the to-be-fused satellite altimeter; cross calibration: performing cross calibration on sea surface height anomaly of the fusion satellite altimeters by using the sea surface height anomaly of the reference satellite altimeter to obtain system deviation in the to-be-fused satellite altimeters; performing self-crossing calibration by using sea surface height anomaly of the satellite altimeter to be fused to obtain a diagonal element of an observation error covariance matrix; constructing a covariance matrix, wherein a non-diagonal element of the observation error covariance matrix is a product of a standard deviation at the two measurement points and a correlation function of the two measurement points; and fusing the fused satellite altimeter and the background field data by using an optimal interpolation method.

The present invention discloses a high-temporal-spatial-resolution water level inversion method that integrates satellite altimetry data and GNSS-R. First, the target area is determined, and the GNSS data, satellite altimetry data, and water body mask data of the target area are obtained; the satellite altimetry data is processed to obtain the observed water level in the study area, and the system offset is eliminated by comparing the average water level deviation during the overlapping period of different altimetry satellites to obtain the multi-source altimetry satellite fusion water level; the GNSS data is processed to construct a mathematical model between the signal-to-noise ratio data and the water surface height, and the GNSS-R water level data is obtained using a nonlinear fitting method; finally, the satellite altimetry data and the GNSS-R data are combined to construct a high-temporal-spatial-resolution water level inversion model based on XGBoost. By combining satellite altimetry data and GNSS-R data, the present invention can achieve water level monitoring with higher temporal-spatial-resolution, significantly improve the accuracy and reliability of water level inversion, and has important applications in disaster warning and water resources management.

The present application relates to the technical field of satellite altimetry and ocean surveying and mapping, and in particular to a GNSS-R sea surface height measurement method based on variable step line segment intermediate frequency data processing, which can effectively improve the measurement accuracy and efficiency, comprising obtaining satellite-borne GNSS-R original intermediate frequency data, calculating mirror reflection points through a new variable step line segment (NVSLSM), generating a delay Doppler map (DDM), and performing sea surface height inversion. Compared with the prior art, the NVSLSM is used to process the mirror reflection points, the calculation time is increased by about 2.04 times and 1.47 times compared with the line segment bisection method and the QH method, and on the premise of ensuring the convergence accuracy, the stable solution can be reached through fewer iteration steps, thereby reducing the calculation cost per sample; in the delay Doppler map generation link, the overall calculation efficiency is increased by more than about 90%.

The present application relates to the technical field of satellite altimetry and ocean surveying and mapping, and in particular to a GNSS-R sea surface height measurement method based on variable step line segment intermediate frequency data processing, which can effectively improve the measurement accuracy and efficiency, comprising obtaining satellite-borne GNSS-R original intermediate frequency data, calculating mirror reflection points through a new variable step line segment (NVSLSM), generating a delay Doppler map (DDM), and performing sea surface height inversion. Compared with the prior art, the NVSLSM is used to process the mirror reflection points, the calculation time is increased by about 2.04 times and 1.47 times compared with the line segment bisection method and the QH method, and on the premise of ensuring the convergence accuracy, the stable solution can be reached through fewer iteration steps, thereby reducing the calculation cost per sample; in the delay Doppler map generation link, the overall calculation efficiency is increased by more than about 90%.

This invention discloses a method for verifying reservoir capacity curves based on satellite altimetry and remote sensing imaging technology, belonging to the fields of water conservancy and hydropower engineering and hydrological and water resources management. The method includes: inverting and correcting the reservoir's water surface elevation using altimetry data from satellite remote sensing monitoring; identifying the reservoir's water body extent and statistically analyzing its area sequence based on remote sensing images; constructing a reservoir capacity calculation model based on the reservoir's concurrent remote sensing water level and area sequences; fitting remote sensing water level-area and water level-capacity curves and evaluating their accuracy; and realizing the process of reconstructing reservoir capacity curves using the same satellite mission by inverting concurrent water level and area data. This method avoids the shortcomings of previous methods that reconstructed reservoir characteristic curves using satellite remote sensing data due to asynchronous water level and area measurements. It also overcomes the limitations of traditional laser or synthetic aperture radar altimetry satellites, which can only acquire sub-satellite point cloud elevations and have long step sizes. This significantly improves the accuracy of the reconstructed characteristic curves and has significant advantages in reconstructing characteristic curves for small reservoirs in remote areas and over large areas.

The invention discloses a satellite height measurement and depth gauge combined submarine topography settlement monitoring method, which belongs to the technical field of ocean measurement, is used for submarine topography settlement monitoring, and comprises the following steps: acquiring time sequence satellite height measurement data of a monitored sea area and depth measurement data of a submarine depth gauge, and preprocessing the time sequence satellite height measurement data and the depth measurement data of the submarine depth gauge; calculating the amplitude and lag angle of astronomical partial tide by utilizing a tidal harmonic analysis method, and eliminating periodic tidal fluctuation to obtain a long-term average sea surface; obtaining an instantaneous sea surface height and a long-term average sea surface height corresponding to the position of the depth gauge through bilinear interpolation; uniformly reducing sounding data to be under a long-term average sea surface reference by calculating a tide level deviation; and constructing a seabed settlement space-time sequence model, and generating a seabed settlement rate map and an accumulated settlement volume curved surface. According to the method, time sequence satellite height measurement data and depth gauge single-point high-precision data are subjected to space-time matching and fusion, vertical reference unification and high-precision transmission from the sea surface to the seabed are achieved, and the cost of traditional ship measurement or dense arrangement of in-situ observation stations is reduced.

The present application provides a gravity inversion accuracy evaluation system and method considering the self-noise of shipborne measurements. The system includes: a shipborne gravity data verification module, which is used to refine the shipborne gravity data, then use the shipborne gravity data to verify the altimetry gravity inversion product to obtain the shipborne verification accuracy of the altimetry gravity inversion product, and use the shipborne gravity data to verify the gravity field model to obtain the shipborne verification accuracy of the gravity field model; a global gravity field model comparison module, which is used to intercept the global gravity field model for comparison to obtain the gravity field model comparison accuracy of the altimetry gravity inversion product; a gravity inversion accuracy evaluation module, which is used to jointly calculate the self-noise of the shipborne verification data and deduct the shipborne self-noise to obtain the accurate satellite altimetry gravity inversion accuracy. The advantages of the present application are: it can effectively remove the influence of the self-noise of shipborne gravity data on the altimetry gravity inversion accuracy and evaluate the accurate altimetry gravity inversion accuracy.

The application discloses a kind of lake satellite radar altimetry data screening processing method, device and electronic equipment.Therein, the method includes: obtaining random sample data from global lake satellite radar altimetry data;Wherein, the basic data required by random sample data includes global satellite altimetry data and lake boundary data;Random sample data is processed based on the preset graphical interface tool, and the echo waveform and multi-source auxiliary information corresponding to random sample data are shown;Wherein, multi-source auxiliary information includes multi-source footprint characteristic value and geographic location information;Random sample data is marked according to the category based on echo waveform and multi-source auxiliary information;Machine learning model training is carried out using random sample data of marked type, and the automatic classification model of lake satellite radar altimetry data is obtained.The application solves the technical problem of low sampling classification efficiency of lake satellite radar altimetry data.

The application provides a sea ice surface snow depth estimation method combined with laser satellite altimetry and radar satellite altimetry, which comprises satellite altimetry data preprocessing, satellite altimetry sampling point matching, obtaining a joint observation track of two satellites according to the time and spatial position of the orbit of the laser altimetry satellite and the orbit of the radar altimetry satellite, taking the sampling point of the radar altimetry satellite along the orbit as a reference point, searching for the sampling point of the laser altimetry satellite in a preset space-time range to obtain a joint observation sampling point set, obtaining the total freeboard value of the laser altimetry satellite corresponding to the radar freeboard of the radar altimetry satellite in each search range by using the inverse distance weighting method to obtain a sea ice freeboard matching data set, and obtaining the sea ice surface snow depth estimation result and outputting the same based on the obtained sea ice freeboard matching data set and considering the penetration characteristics of the radar signal on the snow. The application is free from the dependence on field observation by combining the laser satellite altimetry data and the radar satellite altimetry data, and the reliability of the sea ice surface snow depth estimation is improved.

The present invention discloses a method and system for satellite altimetry bathymetric correction based on VGGNet, which relates to the technical field of deep learning and includes: collecting shipborne multibeam sonar bathymetric data and satellite altimetry bathymetric data as original data; preprocessing the original data to obtain a training data set; constructing a bathymetric correction model based on a convolutional neural network, and inputting the training data set into the bathymetric correction model for model training until the loss function converges to obtain an optimal bathymetric correction model; using the optimal bathymetric correction model to perform water depth correction on satellite altimetry sounding bathymetric data to obtain corrected satellite altimetry water depth data. The present invention uses a deep learning method to correct satellite altimetry water depth data based on multibeam sonar water depth data, providing a new technical means possibility for high-precision and high-real-time bathymetric measurement of the global sea area.

The invention relates to the technical field of ocean geophysical exploration and surveying and mapping, and provides a multi-source and multi-physics field gravity data super-resolution reconstruction method. The method comprises the following steps: acquiring preprocessed low-resolution satellite altimetry gravity data and preprocessed high-resolution ship load data, and constructing a training data set; constructing a dual regression deep learning model; training the dual regression deep learning model by using the training data set to obtain a trained dual regression deep learning model; and inputting preprocessed low-resolution satellite gravity data of a to-be-reconstructed target area into the trained dual regression deep learning model to obtain super-resolution reconstructed gravity data of the reconstructed target area. Through dual network closed-loop constraint and adaptive degradation modeling, dependence on high-resolution shipborne load data is significantly reduced, and reconstruction precision and generalization ability are improved at the same time.

The invention discloses a method for resolving mass change of an ice cover with high spatial resolution based on a three-dimensional acceleration point mass method. The method comprises the following steps: determining a ground point and a spatial point; calculating perturbation acceleration of a certain point in space caused by mass change of all ground points; calculating ice cover mass change data solved by the satellite height measurement data; and by taking the calculated ice cover mass change data as a limiting condition, constructing indirect adjustment with the limiting condition according to the calculated perturbation acceleration, and calculating a mass change vector of the ground point to be solved. According to the invention, the three-dimensional information observed by the satellite is fully fused with the satellite height measurement data, so that the mass change of the ice cover is calculated, and the spatial resolution of the ice cover is improved. According to the method, more accurate scientific data and theoretical support are provided for more precise estimation of ice cover mass change and analysis of global climate change.

The invention discloses a lake hydro-fluctuation belt terrain construction method based on SWOT satellite data, and the method comprises the steps: extracting a month-scale water body range through lake optical remote sensing image data, and constructing a month-by-month hydro-fluctuation belt region based on a difference set of the month-by-month water body range and the maximum water body range; taking the month-by-month hydro-fluctuation belt region as a spatial constraint to cut out a corresponding region from the height measurement data; spatial filtering is carried out on the multi-temporal SWOT satellite altimetry data of the cutting area, the satellite laser altimetry data is used as a true value to evaluate the precision of the corresponding temporal SWOT satellite altimetry data and determine a fusion weight, then weighted average is carried out on the hourly-temporal SWOT satellite altimetry data to obtain fusion elevation information, and the hydro-fluctuation belt terrain is constructed based on the fusion elevation information. The method is particularly suitable for natural lakes and artificial reservoirs with interannual atrophy or large seasonal change, and the lake hydro-fluctuation belt terrain constructed based on the method can support large-area long-time-sequence lake water level and water quantity change monitoring.

The application discloses a gravity anomaly calculation method, device, equipment, storage medium and product, relates to the satellite altimetry data processing technical field, and the gravity anomaly calculation method comprises: obtaining sea surface height data observed by an altimetry satellite; according to the sea surface height data, a plurality of residual geoid gradients in different directions are calculated; the plurality of residual geoid gradients in different directions are fused by a Bayesian triangular cap method, and an east-west direction component and a south-north direction component of the residual geoid gradient are obtained; and based on the east-west direction component and the south-north direction component of the residual geoid gradient, the marine gravity anomaly of a target region is calculated by a gravity anomaly inversion algorithm. The application solves the technical problems of low utilization rate of multi-direction gradient information of a single altimetry satellite and insufficient component extraction precision, realizes mining and optimization of multi-direction gradient information of a single satellite, and improves the inversion precision of the marine gravity anomaly.

This invention relates to a high-precision tropospheric wet delay estimation method based on optimal interpolation, used for wet delay correction of wide-swath satellite altimetry SWOT observation data over land. The technical approach is as follows: First, initial wet delay values ​​for the swath calculation point and GNSS station are obtained through spatiotemporal interpolation using ECMWF reanalysis data, and the deviation between the initial values ​​and measured values ​​is calculated at the GNSS station. Then, assuming that the water vapor observations at the GNSS station are of equal precision, and in the absence of other external observations, the weighting function is simplified to a weight determined by the equivalent wet distance from the GNSS station to the swath calculation point. Finally, the optimal interpolation method is used to complete the wet delay correction estimation for the swath calculation point. The advantage of this method is that it can transfer the high temporal resolution and accuracy of the measured values ​​from the GNSS station, improve the initial values ​​to obtain a more accurate optimal swath interpolation, and significantly improve the estimation accuracy of wet delay.

The application discloses a kind of submarine topography subsidence monitoring methods of satellite altimetry and depth gauge combination, belong to marine surveying technical field, for submarine topography subsidence monitoring, including obtaining the time series satellite altimetry data and submarine depth gauge sounding data of monitoring sea area are preprocessed;Utilize tidal harmonic analysis method to calculate the amplitude and lag angle of astronomical component, eliminate periodic tidal fluctuation, obtain long-term average sea surface;The instantaneous sea surface height and long-term average sea surface height corresponding to the depth gauge position are obtained by bilinear interpolation;By calculating tidal deviation, the sounding data is uniformly reduced to long-term average sea surface datum;Submarine subsidence spatiotemporal sequence model is constructed, and submarine subsidence rate graph and cumulative subsidence amount surface are generated.The application is spatiotemporal matching and fusion by time series satellite altimetry data and depth gauge single-point high-precision data, realizes the vertical datum uniformity and high-precision transmission from sea surface to submarine, reduces the cost of traditional ship measurement or dense arrangement in-situ observation station.

The application relates to the technical field of marine geophysical exploration and mapping, and provides a multi-source and multi-physical-field gravity data super-resolution reconstruction method.The method comprises the following steps: acquiring preprocessed low-resolution satellite altimetry gravity data and preprocessed high-resolution shipborne gravity data, and constructing a training data set; constructing a dual regression deep learning model; training the dual regression deep learning model by using the training data set, so as to obtain a trained dual regression deep learning model; inputting preprocessed low-resolution satellite gravity data of a target area to be reconstructed into the trained dual regression deep learning model, and acquiring super-resolution reconstruction gravity data of the target area to be reconstructed.Through dual network closed-loop constraint and adaptive degradation modeling, the application can significantly reduce the dependence on high-resolution shipborne gravity data, and improve the reconstruction accuracy and generalization ability.

The invention relates to the field of positioning data processing methods, in particular to a self-adaptive fusion method for ocean multi-source gravity data, which comprises the following steps: step 1, arranging obtained gravity data of a target area; step 2, carrying out mass classification on the ship measurement gravity data; step 3, determining weight coefficients of the ship measurement gravity data of different mass categories for correcting the satellite height measurement gravity data, circulating the weight coefficients of different mass categories according to constraint conditions, and stopping circulating when an end condition is reached; 4, correcting and fusing the satellite altimetry gravity data according to a preset rule by using the ship-measured gravity data and the weight coefficient; and step 5, calculating and correcting a root-mean-square error between the gravity anomaly reference map obtained by fusion and the high-precision ship measurement gravity data which does not participate in the fusion process, taking a weight coefficient when the root-mean-square error is minimum as a fusion parameter combination in the step 3, and finally obtaining a high-fineness gravity reference map. According to the invention, the accuracy of underwater navigation can be improved.

The invention relates to the technical field of satellite height measurement data processing, and discloses a height measurement echo waveform redetermination method and system based on pole symmetry mode decomposition, and the method comprises the steps: introducing a pole symmetry mode decomposition method, and carrying out the multi-mode decomposition of an echo waveform signal of satellite height measurement data, so as to complete the waveform denoising; constructing a dual-mode fusion adaptive waveform analysis model, and reconstructing a complete height measurement waveform sequence; and through a Bayesian algorithm, waveform parameters are optimized in a combined manner, a neural network model is combined, adaptive echo waveform redetermination is carried out, and waveform classification and processing are completed. According to the method, a pole symmetry mode decomposition method is introduced, multi-scale intrinsic mode decomposition is carried out on echo waveform signals, high-frequency noise modes are accurately separated, high-fidelity reconstruction of the signals is achieved, the method has excellent nonlinear signal processing capacity and local adaptivity, the denoising effect is remarkably improved, and by combining adaptive echo waveform reconstruction, the denoising effect is greatly improved. The waveform reconstruction precision in complex terrains and severe environments is improved, and the high-precision satellite height measurement data processing requirement is met.

The invention discloses a sea surface height anomaly (SLA) multi-step space-time prediction method based on combination of self-attention Transform and a recurrent neural network, and the method comprises a data preprocessing module, a space-time feature extraction module, a prediction generation module and a space reconstruction module, and comprises the following steps: inputting SLA data obtained by satellite height measurement into the data preprocessing module; performing blocking and normalization processing to generate a standardized input sequence; inputting the standardized sequence into a spatio-temporal feature extraction module, extracting global spatial dependency characteristics by using a self-attention Transform module, and extracting dynamic change characteristics of time dimensions through a recurrent neural network; recursively generating SLA prediction data of a plurality of time steps by using a prediction generation module to form a multi-step prediction sequence; and finally, the multi-step prediction sequence is restored to the original spatial resolution through a spatial reconstruction module, and a final prediction result is obtained. According to the method, the self-attention Transform and the recurrent neural network are innovatively combined, the limitation of a traditional method in modeling of the complex spatial-temporal dynamic characteristics of the SLA is broken through, the multi-step prediction precision is remarkably improved, and effective technical support is provided for ocean mesoscale vortex monitoring and environment forecasting.

The present invention discloses a method for calculating depth reference epochs based on GNSS and satellite altimetry, which belongs to the field of marine surveying and mapping technology. The method is used for calculating depth references, including data processing of a time series of vertical crustal deformation, calculating the linear rate of vertical crustal deformation, and data processing of a time series of absolute sea level height to obtain the rate of change of absolute sea level height; calculating a depth reference value based on tide gauge data, performing epoch reduction on the depth reference value, and verifying the epoch reduction results. The present invention more comprehensively captures the dynamic characteristics of vertical crustal deformation and absolute sea level height changes; on a long-term time scale, it can more accurately reflect the trend of sea level height changes; provides a new theoretical basis for depth reference epoch reduction, dynamically adjusts the depth reference value, avoids the problem of reference surface drift caused by sea level changes and crustal movement, and thus ensures the long-term stability and scientific nature of the depth reference.

The invention discloses a gravity inversion method based on a physical information neural network, and belongs to the technical field of ocean gravity inversion, and the method specifically comprises the steps: preprocessing submarine topography, ship load and satellite height measurement data; a physical reference model is constructed based on the consideration of an inverse Vening Meinesz formula, the precision of the ship load and the physical reference model is evaluated, and hyper-parameters of fusion data and physical information are quantitatively determined; constructing a physical information neural network model, and determining a loss function; by training the model, high-precision ocean gravity anomaly inversion is finally realized; according to the method, a physical mechanism is used as constraint to be embedded into neural network training, the problems that a traditional inversion method is low in precision and an existing neural network is poor in generalization ability in a data sparse region are effectively solved, and the method is particularly suitable for high-precision gravity field construction of a marine area with sparse ship load distribution.

The invention discloses a gravity anomaly calculation method, device and equipment, a storage medium and a product, and relates to the technical field of satellite altimetry data processing, and the gravity anomaly calculation method comprises the steps: obtaining sea surface height data observed by an altimetry satellite; according to the sea surface height data, a plurality of residual geoid gradients in different directions are obtained through calculation; performing fusion processing on the residual geoid gradients in the different directions through a Bayesian triangular cap method to obtain east-west direction components and south-north direction components of the residual geoid gradients; and based on the east-west direction component and the south-north direction component of the residual geoid gradient, calculating through a gravity anomaly inversion algorithm to obtain the ocean gravity anomaly of the target area. According to the method, the technical problems of low utilization rate of multi-directional gradient information and insufficient component extraction precision of a single altimetry satellite are solved, mining and optimization of the multi-directional gradient information of the single satellite are realized, and the inversion precision of ocean gravity anomaly is improved.

The invention relates to the cross technical field of satellite altimetry, satellite navigation and the like, in particular to a method for improving the accuracy of an altimetry satellite track cross point based on multi-factor constraint, which comprises the following steps of: pre-screening data to reduce invalid calculation; according to the sea surface height data calculation method, a rapid exclusion and straddling test mechanism is introduced, and the calculation efficiency is improved through strategies such as constraint of an accurate intersection calculation window and the like; according to the method, large-area and high-precision satellite data provided by an SWOT satellite task is used for calculation, the capability of efficiently processing the satellite data and obtaining a high-precision intersection point result is achieved, high precision of the intersection point position can be effectively guaranteed, sea surface height calculation verification of different areas is carried out based on the algorithm, and the calculation efficiency is improved. The obtained sea level height has good precision in different latitude areas.

The present invention discloses a data-deficient river and lake water level prediction method based on multi-source remote sensing information and a machine learning algorithm. The method utilizes multiple satellite altimetry products to obtain historical river and lake water level data, multiple remotely sensed precipitation products to obtain historical precipitation data for river and lake basins, and multiple remotely sensed evapotranspiration products to obtain historical evapotranspiration data for river and lake basins. A machine learning model is introduced, using remotely sensed inverted river and lake water level data, precipitation data, and evapotranspiration data as the primary training data. With water level as the prediction variable and precipitation and evapotranspiration as the primary predictive factors, an intelligent river and lake water level prediction model is established. The model then uses corrected future precipitation forecast data to predict future river and lake water level changes. The advantages of this method are: By using big data methods to learn the driving mechanisms between basin meteorological factors and river and lake water levels, the method reduces the errors and uncertainties caused by the inaccurate depiction of physical processes by traditional mechanism models, thereby further improving the accuracy of river and lake water level predictions.

The present application relates to the technical field of satellite altimetry, in particular to a GNSS-R altimetry method based on attention mechanism feature enhancement residual multi-modal, which can significantly improve the altimetry precision. First, a new attention mechanism feature enhancement residual multi-modal FERM model is constructed, and then the FERM model is used to realize altimetry. The FERM model is composed of a feature enhancement residual FER network based on an attention mechanism and a fully connected neural network FCNN. The FER uses a channel attention module and a spatial attention module to distinguish and enhance the DDM channel and spatial effective features extracted by the residual structure, so as to enhance the features from two aspects of improving the DDM information utilization rate and giving the feature attention weight, and then improve the SSH inversion precision of the FCNN from the input quality.