High time resolution water level retrieval method based on multi-source height measurement data fusion

Abstract

The application discloses a high-time-resolution water level inversion method based on multi-source height measurement data fusion. The method significantly improves the accuracy and time resolution of the water level sequence through dynamic fusion of multiple satellite data. First, the multi-source satellite data is used to invert the target water body water level time sequence, and the systematic error and noise are eliminated through benchmark unification and abnormal value elimination. Then, combined with Kalman filtering technology, a dynamic weighting mechanism based on Helmert variance component estimation is introduced, the noise covariance is accurately evaluated, and the water level state estimation is optimized. The method effectively improves the water level monitoring accuracy and time resolution, enhances the anti-interference ability of low-quality data, provides a powerful tool for hydrological research and water resources management, and in practical application, the water level accuracy is equivalent to that of the authoritative data set (DAHITI / Hydroweb), and the time resolution is significantly improved.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing hydrological monitoring technology, and in particular relates to a high temporal resolution water level inversion method based on multi-source altimetry data fusion. Background Technology

[0002] Inland lake water level changes are among the most sensitive indicators in the Earth's water cycle system, reflecting the combined effects of regional climate change, water resource balance, and ecological environment evolution. Long-term changes in lake water levels directly impact the structure of surrounding ecosystems, vegetation distribution, and wetland extent, and indirectly affect regional climate regulation, groundwater recharge, and surface runoff processes, making them crucial for water resource management and disaster prevention and mitigation. Continuous, accurate, and long-term lake water level monitoring is a vital foundation for studying global change, watershed water resource management, and ecological security assessment.

[0003] However, constrained by factors such as complex geographical environments, harsh climatic conditions, and economic limitations, the number of surface hydrological observation stations worldwide is decreasing year by year. Particularly in high-altitude, frigid, and uninhabited areas, the high costs of constructing and maintaining hydrological stations result in extremely uneven spatial coverage and discontinuities in time series of measured water level data, making it difficult to meet the need for continuous monitoring of lake dynamics. This insufficient observational capacity has become a significant bottleneck restricting research on regional and even global hydrological processes and climate change.

[0004] With the rapid development of satellite remote sensing technology, satellite altimetry has provided a new approach for obtaining large-scale, long-term lake water level data. Compared to traditional ground-based measurements, satellite altimetry can achieve rapid monitoring of water elevation globally, offering advantages such as independence from ground-based observation networks, wide spatial coverage, and stable observation cycles. After decades of development, existing in-orbit altimetry satellites can provide water level data with centimeter-level accuracy, providing crucial data support for global hydrological and climate monitoring.

[0005] Existing satellite altimetry methods employ radar or laser altimeters. Using a single radar altimeter to retrieve lake levels is typically limited by temporal resolution and spatial coverage, making it difficult to retrieve levels in small to medium-sized lakes. The ICESat-2 satellite, equipped with a laser altimeter, has a smaller footprint diameter than radar altimeters, allowing it to monitor smaller bodies of water, but its temporal resolution is lower. Combining the two methods can achieve complementary advantages.

[0006] There are three main existing fusion methods:

[0007] (1) Reference conversion method: This method directly fuses data by eliminating the reference bias of different satellite data, but the accuracy of this method is low.

[0008] (2) Morphological modeling method: a fusion model is constructed by taking into account parameters such as water body shape and length. Although this method can improve the time resolution, the model has weak generalization ability, insufficient adaptability to irregular water bodies, and high computational complexity.

[0009] (3) Mainstream datasets: DAHITI achieves global coverage through multi-satellite fusion but lacks data on small and medium-sized lakes, while Hydroweb focuses on river linkages but has poor timeliness. The time resolution of both is usually two weeks or months or even lower, making it difficult to capture short-term dynamic changes in lake water levels.

[0010] It is evident that the commonly used multi-source height measurement data fusion methods have their own drawbacks, making it difficult to construct water level time series with high temporal resolution and high accuracy. Summary of the Invention

[0011] To overcome the shortcomings of the existing technologies, this invention introduces the Helmert-Kalman dynamic fusion method (HKDF) and proposes a high temporal resolution water level inversion method based on the fusion of multi-source altimetry data. This method does not rely on water body morphology parameters, constructs a unified benchmark water level time series based on multi-source altimetry data, and performs bias estimation and weighted fusion techniques based on this to finally form a high-precision, high-temporal-resolution water level series.

[0012] According to one aspect of the present invention, a high temporal resolution water level inversion method based on multi-source altimetry data fusion is provided, comprising:

[0013] Acquire observation data from multiple altimetry satellites;

[0014] By using lake water masking to filter the transit trajectories of each satellite, the effective echo signals within the lake area are extracted. Combined with waveform retracking and water surface elevation inversion methods, the time series of lake water levels for each satellite are obtained.

[0015] The obtained lake water level time series from each satellite are standardized and outliers are removed to obtain a multi-source water level time series for fusion.

[0016] A state-space model is constructed based on multi-source water level time series. The state noise covariance is determined by using the water level change rate and standard deviation retrieved from satellites. The weights of the observation noise of each satellite and the observation noise covariance are dynamically determined by using Helmert variance components. A fused high-temporal-resolution water level time series is generated through recursive iteration of Kalman filtering.

[0017] As a further technical solution, the state noise covariance is determined using satellite-retrieved water level change rate and standard deviation, including:

[0018] Obtain the time series of effective measured water levels around each hydrological station obtained from satellite inversion;

[0019] Based on the acquired time series, calculate the water level changes at adjacent time points;

[0020] Based on the water level changes at adjacent time points, the standard deviation of the rate of change is calculated, and then the state noise covariance is calculated.

[0021] As a further technical solution, the weights of the observation noise of each satellite and the observation noise covariance are dynamically determined using Helmert variance component estimation, including:

[0022] Based on the observations from different satellites corresponding to each lake, the least squares method is used to calculate the estimated value and the residual vector;

[0023] Based on the Helmert variance component estimation formula, the unit weight variance of each satellite is calculated using the residual vector;

[0024] Update the weights of each satellite based on the unit weight variance of each satellite;

[0025] Repeat the weight update process until the iteration termination condition is met;

[0026] Calculate the observation noise covariance for each satellite based on its weight.

[0027] As a further technical solution, after constructing the state-space model, the first value of the multi-source water level time series is used as the initial value, and the initial estimated covariance is set to zero.

[0028] As a further technical solution, the obtained time series of lake water levels from various satellites are standardized in terms of benchmarks and outlier removal is performed, including:

[0029] The obtained time series of lake water levels from various satellites were unified to the same reference standard;

[0030] Based on data unified to the same reference benchmark, systematic deviations between different satellite missions are analyzed and corrected to construct a water level time series with a unified benchmark.

[0031] A support vector regression algorithm is introduced to identify and remove outliers in the benchmark unified water level time series.

[0032] As a further technical solution, when introducing the support vector regression algorithm to identify outliers, abnormal water levels are eliminated by using a confidence interval of twice the rate of change of the water level time series.

[0033] According to one aspect of the present invention, a high temporal resolution water level inversion system based on multi-source altimetry data fusion is provided, for implementing the aforementioned high temporal resolution water level inversion method based on multi-source altimetry data fusion, comprising:

[0034] The first main module is used to acquire observation data from multiple altimetry satellites;

[0035] The second main module is used to filter the transit trajectories of each satellite using the lake water mask, extract the effective echo signals in the lake area, and obtain the lake water level time series of each satellite by combining waveform retracking and water surface elevation inversion methods.

[0036] The third main module is used to unify the benchmark and remove outliers from the lake water level time series obtained from each satellite, so as to obtain a multi-source water level time series for fusion.

[0037] The fourth main module is used to construct a state-space model based on multi-source water level time series, determine the state noise covariance by using the water level change rate and standard deviation retrieved from satellites, dynamically determine the weight of each satellite observation noise and the observation noise covariance by using Helmert variance component estimation, and generate a fused high temporal resolution water level time series through Kalman filtering recursively iteratively.

[0038] According to one aspect of the present invention, a high temporal resolution water level inversion device based on multi-source altimetry data fusion is provided, comprising a memory and a processor, wherein the memory stores program instructions that are executed by the processor, and the processor invokes the program instructions to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion.

[0039] According to one aspect of the present invention, a non-transitory computer-readable storage medium is provided, the non-transitory computer-readable storage medium storing computer instructions that cause the computer to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] The method of this invention provides high-precision, long-term, and high-frequency lake water levels by systematically fusing multi-source satellite data, without relying on water morphology parameters, thus significantly improving the temporal resolution of water level series. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the high temporal resolution water level inversion method based on multi-source altimetry data fusion provided in an embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram of the water level inversion result from a single altimeter satellite provided in an embodiment of the present invention, where a, b, c, d, and e are the accuracy comparisons between the water level inverted by the satellite and the measured water level, respectively.

[0045] Figure 3 The present invention provides a comparison of the fused water level time series and accuracy, wherein a is the time series of water level, water level product and in-situ water level before and after satellite fusion of a certain lake; b, c, d and e are the accuracy comparisons of Hydroweb, DAHITI, water level before fusion and water level after fusion with measured water level, respectively. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined to form new technical solutions. Such combinations are not bound by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0047] This invention proposes a multi-source altimetry data fusion method independent of water body morphology parameters, solving the problems of insufficient accuracy and low spatiotemporal resolution of single satellite altimetry data. This method significantly improves the accuracy and temporal resolution of water level sequences through dynamic fusion of data from multiple satellites. First, the time series of water levels in the target water body is retrieved using multi-source satellite data, and systematic errors and noise are eliminated through benchmark unification and outlier removal. Then, combined with Kalman filtering technology, a dynamic weighting mechanism based on Helmert variance component estimation is introduced to accurately evaluate noise covariance and optimize water level state estimation. This method effectively improves the accuracy and temporal resolution of water level monitoring and enhances the anti-interference ability of low-quality data, providing a powerful tool for hydrological research and water resource management. In practical applications, the water level accuracy is comparable to authoritative datasets (DAHITI / Hydroweb), and the temporal resolution is significantly improved.

[0048] Please see Figure 1 The implementation process of the method provided by this invention is as follows:

[0049] Step (1): Acquisition of observation data from multi-mission altimeter satellites.

[0050] Step (2): Use the lake water mask to screen the transit trajectories of each satellite, extract the effective echo signals in the lake area, and combine waveform retracking and water surface elevation inversion methods to obtain the lake water level time series of each satellite.

[0051] Step (3) unifies the satellite water level time series obtained in step (2) to the same reference benchmark: the elevation system is converted to be consistent with the ground measured water level benchmark, and the time system is converted to be unified.

[0052] Step (4): Based on the data obtained in step (3), analyze and correct the systematic deviations between different satellite missions, establish a unified time series benchmark, and ensure the consistency of multi-source water level data in overall trend and amplitude.

[0053] Step (5): Based on step (4), the support vector regression (SVR) algorithm is introduced to identify and remove outliers in the water level time series.

[0054] Step (6) involves dynamically fusing multi-source water level sequences based on the water level data obtained in step (5) using Kalman filtering. The state noise covariance is determined using the satellite water level change rate and standard deviation. To dynamically adjust the observation noise weights, Helmert variance component estimation is employed to evaluate the observation noise and iteratively optimize the weights until convergence, thereby determining the observation noise covariance. Recursive estimation is then performed based on the constructed state-space model to finally obtain the fused high-temporal-resolution water level time series data.

[0055] Kalman filtering (KF) is a widely used algorithm for recursive estimation and data fusion. It achieves optimal state estimation by integrating observations, dynamic models, and their respective uncertainties, minimizing the mean square error. Specifically, in the water level estimation of this invention, KF achieves optimal fusion of multi-source measurement data through prediction and update steps, including the following steps:

[0056] Step S1: Construct the state-space model.

[0057] The core of Kalman filtering includes state equations and measurement equations. This invention uses the following linear state equations and measurement equations to perform water level inversion.

[0058] ,

[0059] In the formula, , Let represent the posterior state estimates of the water level at time k and time (k+1), respectively. is the optimal estimate of the state at that time. This invention performs water level inversion. Let be the water level inversion value at time k. A represents the state transition model, assuming the water level follows a random walk process, i.e., A is the identity matrix. It is the satellite-inverted water level value at the k-th time after benchmark unification, and H is the observation model that maps the real state space to the observation space. Here, H=1 is taken. For process noise, These are observation noises, assumed to be independent Gaussian white noises that follow a normal distribution, with covariances satisfying: , .

[0060] Step S2, prediction.

[0061] Before Kalman filtering begins, an initial state estimate x0 and an initial estimated covariance matrix P0 are required. Through the steps described above, the predicted (prior) state estimate can be obtained from the prediction step. and prior estimate of covariance :

[0062] .

[0063] Step S3, update.

[0064] Obtain the Kalman gain from the update step. posterior state estimate and posterior estimation of covariance :

[0065] .

[0066] Step S4: State noise and observation noise estimation.

[0067] KF simulates the uncertainties of the system and observations through process noise and measurement noise. Therefore, accurate estimation of the state noise covariance Q and the observation noise covariance R is crucial. This invention proposes a method to accurately determine Q and R by using satellite inversion of water level.

[0068] The state noise Q is determined by the rate of change and standard deviation of the water level retrieved from the satellite, as follows:

[0069] ① Obtain the time series of effective measured water levels around each hydrological station. i = 1, 2, ..., n, where n is the length of the time series;

[0070] ② Calculate the water level changes at adjacent time points : ;

[0071] ③ Calculate the standard deviation of the rate of change : ;

[0072] ④ Calculate the state noise covariance: .

[0073] To dynamically adjust the observation noise weights of satellites, this invention proposes the Helmert variance component estimation method to estimate the observation noise R of each satellite. The specific steps are as follows:

[0074] ① Assuming a lake has observations from m different satellites, its function model is:

[0075] ,

[0076] In the formula, It is a set of column vectors representing observations of the same lake from different satellites. It is the corresponding coefficient matrix. It is a column vector composed of unknowns.

[0077] ② Calculate the estimated value using the least squares method. Then calculate the residual vector. ;

[0078] ,

[0079] In the formula, Let B represent the weight of satellite i (i=1,2,...,m), and let B be the coefficient matrix with the initial weight set to 1.

[0080] ③ Based on the rigorous Helmert variance component estimation formula, the unit weight variance of each satellite is calculated using the residual vector. The formula is as follows:

[0081] , ,

[0082] In the formula, tr represents the estimated variance of the unit weight, and tr represents the rank of the matrix. , , These represent the residual vector, normal matrix, and number of observations for the i-th group, respectively.

[0083] ④ Update weights Where c is a constant, and in this invention, c = ;

[0084] ,

[0085] ⑤ Repeat steps ①-④ until the iteration termination condition is met: ;

[0086] ⑥ Calculate the observed noise covariance: .

[0087] As a preferred embodiment, this invention takes a lake as an example to introduce the implementation steps of the high temporal resolution water level inversion method based on multi-source altimetry data fusion:

[0088] Step (1): Download data from Sentinel-3, HY-2B, Cryosat-2, Jason-3, ICEsat-2, and GEDI satellites from January 2019 to December 2020 for a certain lake;

[0089] Step (2): Use lake mask data to filter the transit trajectory points of each altimeter satellite, extract the effective echo signals in the lake area, and calculate the water level time series of each satellite;

[0090] Step (3) is to perform a unified reference transformation on the water level time series obtained in step (2): the reference ellipsoid is unified to the WGS84 ellipsoid, the elevation system is converted to the water level reference of the surface hydrological station, and the time system is converted to the unified simplified Julian day.

[0091] Step (4): Based on the unified benchmark water level time series obtained in step (3), the Sentinel-3 satellite water level series is selected as the reference standard, and the systematic deviation between each mission is calculated and corrected to obtain a relatively consistent multi-source water level time series.

[0092] Step (5): For the multi-source water level sequence obtained in step (4), use the support vector regression algorithm to identify abnormal observation points, and use twice the rate of change of the water level time series as the confidence interval to remove abnormal water levels, thereby improving the stability and reliability of the water level sequence.

[0093] Step (6): Based on the water level data obtained in step (5), a state-space model is constructed to achieve dynamic data fusion. The first value of the satellite water level time series is used as the initial value, and the initial estimated covariance is set to 0. The variance component estimation method is used to dynamically determine the weight of each satellite observation noise, and each state noise is determined by observing the water level. Finally, a fused high-time-resolution, high-precision water level time series is generated through recursive iteration of Kalman filtering.

[0094] Figure 2 A schematic diagram of the water level inversion results from a single altimeter satellite is presented. In the diagram, a, b, c, d, and e represent the accuracy comparison between the water level inverted by the satellite and the measured water level, respectively. Figure 3 A comparison of the accuracy of the fused water level time series is presented. Figure a shows the time series of water level, water level products, and in-situ water level before and after satellite fusion; b, c, d, and e show the accuracy comparisons of water level with measured water levels from Hydroweb, DAHITI, before fusion, and after fusion, respectively. The accuracy of the fused water level time series is significantly higher than that before fusion, and comparable to that of the Hydroweb and DAHITI databases. The temporal resolution of the fused water level time series is significantly higher than that of the two databases. Table 1 compares the capabilities of lake water level data fusion methods.

[0095] Table 1 Comparison of the capabilities of lake water level data fusion methods

[0096] .

[0097] The implementation of the various embodiments of the present invention is based on programmed processing by a device with processor functionality. Therefore, in practical engineering, the technical solutions and functions of the various embodiments of the present invention are encapsulated into various modules. Based on this reality, and building upon the above embodiments, the embodiments of the present invention provide a high temporal resolution water level inversion system based on multi-source altimetry data fusion. This system is used to execute a high temporal resolution water level inversion method based on multi-source altimetry data fusion from the above method embodiments.

[0098] The system comprises: a first main module for acquiring observation data from multiple altimetry satellites; a second main module for filtering the transit trajectories of each satellite using a lake water mask, extracting effective echo signals within the lake area, and obtaining the lake water level time series of each satellite by combining waveform retracking and water surface elevation inversion methods; a third main module for standardizing the lake water level time series of each satellite and removing outliers to obtain a multi-source water level time series for fusion; and a fourth main module for constructing a state-space model based on the multi-source water level time series, determining the state noise covariance using the inverted water level change rate and standard deviation from the satellites, dynamically determining the weights of the observation noise of each satellite and the observation noise covariance using Helmert variance component estimation, and generating a fused high temporal resolution water level time series through Kalman filtering recursively iteratively.

[0099] This invention provides a high temporal resolution water level inversion system based on multi-source altimetry data fusion. Addressing the shortcomings of several commonly used multi-source altimetry data fusion methods, which struggle to construct high temporal resolution and high accuracy water level time series, this invention employs several aforementioned modules and introduces the Helmert-Kalman dynamic fusion method to propose a high temporal resolution water level inversion method based on multi-source altimetry data fusion. This method does not rely on water body morphology parameters, constructs a unified benchmark water level time series based on multi-source altimetry data, and performs bias estimation and weighted fusion techniques based on this, ultimately forming a high-accuracy, high temporal resolution water level series.

[0100] It should be noted that the system embodiments provided by the present invention are used not only to implement the methods in the above method embodiments, but also to implement the methods in other method embodiments provided by the present invention. The only difference is that corresponding functional modules are set. The principle is basically the same as that of the above system embodiments provided by the present invention. As long as those skilled in the art can improve the modules in the above system embodiments by referring to the specific technical solutions in other method embodiments and combining technical features to obtain corresponding technical means and technical solutions composed of these technical means, on the basis of the above system embodiments, and on the premise of ensuring the practicality of the technical solutions, they can obtain corresponding system-like embodiments for implementing the methods in other method-like embodiments.

[0101] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides a high temporal resolution water level inversion device based on multi-source altimetry data fusion, including a memory and a processor. The memory stores program instructions that are executed by the processor, and the processor calls the program instructions to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion.

[0102] In embodiments of the present invention, the memory can be non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it can be volatile memory, such as random-access memory (RAM). Memory is any other medium capable of carrying or storing desired program code having an instruction or data structure form and accessible by a computer, but is not limited thereto. The memory in embodiments of the present invention can also be a circuit or any other device capable of implementing a storage function for storing program instructions and / or data.

[0103] In this embodiment of the invention, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in this embodiment of the invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in this embodiment of the invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0104] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides a non-transitory computer-readable storage medium storing computer instructions. These computer instructions instruct the computer to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion, with the following steps:

[0105] Acquire observation data from multiple altimetry satellites;

[0106] By using lake water masking to filter the transit trajectories of each satellite, the effective echo signals within the lake area are extracted. Combined with waveform retracking and water surface elevation inversion methods, the time series of lake water levels for each satellite are obtained.

[0107] The obtained lake water level time series from each satellite are standardized and outliers are removed to obtain a multi-source water level time series for fusion.

[0108] A state-space model is constructed based on multi-source water level time series. The state noise covariance is determined by using the water level change rate and standard deviation retrieved from satellites. The weights of the observation noise of each satellite and the observation noise covariance are dynamically determined by using Helmert variance components. A fused high-temporal-resolution water level time series is generated through recursive iteration of Kalman filtering.

[0109] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0111] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0112] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0113] In summary, this invention significantly improves the accuracy and temporal resolution of water level series through the dynamic fusion of data from multiple satellites. First, the time series of water levels in the target water body is retrieved using multi-source satellite data, and systematic errors and noise are eliminated through benchmark unification and outlier removal. Then, combining Kalman filtering technology, a dynamic weighting mechanism based on Helmert variance component estimation is introduced to accurately evaluate noise covariance and optimize water level state estimation. This method effectively improves the accuracy and temporal resolution of water level monitoring and enhances the anti-interference capability of low-quality data, providing a powerful tool for hydrological research and water resource management. In practical applications, the water level accuracy is comparable to authoritative datasets (DAHITI / Hydroweb), and the temporal resolution is significantly improved.

[0114] The terms “comprising” and “having”, and any variations thereof, in the specification, claims, and accompanying drawings of this invention are intended to cover a non-exclusive inclusion, such as a process, method, system, product, or apparatus that includes a series of steps or units, not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A high temporal resolution water level inversion method based on multi-source height measurement data fusion, characterized in that, include: Acquire observation data from multiple altimetry satellites; By using lake water masking to filter the transit trajectories of each satellite, the effective echo signals within the lake area are extracted. Combined with waveform retracking and water surface elevation inversion methods, the time series of lake water levels for each satellite are obtained. The obtained lake water level time series from each satellite are standardized and outliers are removed to obtain a multi-source water level time series for fusion. A state-space model is constructed based on multi-source water level time series. The state noise covariance is determined by using the water level change rate and standard deviation retrieved from satellites. The weight of the observation noise of each satellite is dynamically determined by using Helmert variance component estimation, and the observation noise covariance is calculated accordingly. A fused high temporal resolution water level time series is generated through Kalman filtering recursively iteratively. The method of determining the state noise covariance by using satellite-retrieved water level change rate and standard deviation includes: acquiring the time series of effective measured water levels around each hydrological station obtained by satellite inversion; calculating the water level change at adjacent time points based on the acquired time series; calculating the standard deviation of the change rate based on the water level change at adjacent time points, and then calculating the state noise covariance. The method of dynamically determining the weights of observation noise for each satellite using Helmert variance component estimation and calculating the observation noise covariance accordingly includes: calculating the estimated value and residual vector using the least squares method based on the observation values ​​of different satellites corresponding to each lake; calculating the unit weight variance of each satellite using the residual vector according to the Helmert variance component estimation formula; updating the weights of each satellite based on the unit weight variance of each satellite; repeating the weight update process until the iteration termination condition is met; and calculating the observation noise covariance of each satellite based on its weight.

2. The high temporal resolution water level inversion method based on multi-source altimetry data fusion according to claim 1, characterized in that, After constructing the state-space model, the first value of the multi-source water level time series is used as the initial value, and the initial estimated covariance is set to zero.

3. The high temporal resolution water level inversion method based on multi-source altimetry data fusion according to claim 1, characterized in that, The obtained time series data of lake water levels from various satellites were standardized to a benchmark and outlier removed, including: The obtained time series of lake water levels from various satellites were unified to the same reference standard; Based on data unified to the same reference benchmark, systematic deviations between different satellite missions are analyzed and corrected to construct a water level time series with a unified benchmark. A support vector regression algorithm is introduced to identify and remove outliers in the benchmark unified water level time series.

4. The high temporal resolution water level inversion method based on multi-source altimetry data fusion according to claim 3, characterized in that, When introducing support vector regression algorithm to identify outliers, abnormal water levels are removed using a confidence interval of twice the rate of change of the water level time series.

5. A high temporal resolution water level inversion system based on multi-source altimetry data fusion, used to implement the high temporal resolution water level inversion method based on multi-source altimetry data fusion as described in any one of claims 1 to 4, characterized in that, include: The first main module is used to acquire observation data from multiple altimetry satellites; The second main module is used to filter the transit trajectories of each satellite using the lake water mask, extract the effective echo signals in the lake area, and obtain the lake water level time series of each satellite by combining waveform retracking and water surface elevation inversion methods. The third main module is used to unify the benchmark and remove outliers from the lake water level time series obtained from each satellite, so as to obtain a multi-source water level time series for fusion. The fourth main module is used to construct a state-space model based on multi-source water level time series, determine the state noise covariance by using the water level change rate and standard deviation retrieved from satellites, dynamically determine the weight of each satellite observation noise by using Helmert variance component estimation and calculate the observation noise covariance accordingly, and generate a fused high temporal resolution water level time series through Kalman filtering recursively iteratively.

6. A high temporal resolution water level inversion device based on multi-source altimetry data fusion, characterized in that, The method includes a memory and a processor, wherein the memory stores program instructions that are executed by the processor, and the processor invokes the program instructions to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions that cause the computer to execute the high temporal resolution water level inversion method based on multi-source altimetry data fusion as described in any one of claims 1 to 4.

Images