Land water reserve dynamic attribution evaluation method and device based on time lag effect

Abstract

The invention discloses a land water reserve dynamic attribution evaluation method and device based on a time lag effect. The method comprises the following steps: obtaining a land water reserve data set and a related driving factor data set of a target area in a historical time period, and unifying grid resolution through grid preprocessing and resampling; analyzing the spatial-temporal dynamic state of the land water reserves based on the obtained grid set, and outputting the change trend and degree of the land water reserves of the target area in the research time period; decomposing the seasonal trend to decompose the land water reserves of the target area into a trend term, a seasonal term and a residual term, and evaluating the relative contribution of the three components; calculating the correlation coefficient of the driving factor of each grid in the grid set, performing significance test on the correlation coefficient, screening out the driving factors without obvious time-delay effect, and generating a key driving factor set; and finally, evaluating the interpretation degree of the key driving factors on the land water reserve dynamic state and the combined effect thereof.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrological monitoring, and in particular to a method and device for dynamic attribution assessment of terrestrial water reserves based on time lag effect. Background Art

[0002] Terrestrial water storage is defined as the total amount of water stored on land, and related changes represent the dynamics of available freshwater resources at regional and national scales, which are more closely related to natural ecosystems and socio-economic development.

[0003] Since 2002, the Gravity Recovery and Climate Experiment (GRACE) satellite and its follow-up satellite (GRACE-FO) have been able to provide monthly land water storage anomalies with unprecedented accuracy by measuring the distance changes between the binary satellites. In particular, the GRACE satellites enable TWSA to conduct investigations at multiple temporal (from seasonal to decadal) and spatial (from regional to global) scales.

[0004] Climate change and human activities are the main factors affecting the dynamic changes of terrestrial water reserves. At the same time, the relationship between driving factors and changes in terrestrial water reserves is mostly analyzed by simple correlation. There is a lack of quantitative analysis of the contribution of climate factors to changes in terrestrial water reserves, and then an assessment of the impact of human activities on it. In addition, a large amount of evidence shows that the dynamic changes in terrestrial water reserves generally lag behind changes in climate indicators, and the lag time varies for different indicators or in different regions.

[0005] Therefore, there is currently a lack of a method that can accurately and quickly calculate the impact of climate change and human activities on water reserves based on time lag effects. Summary of the invention

[0006] In order to overcome the shortcomings of the prior art, the present invention provides a dynamic attribution assessment method for terrestrial water reserves based on time lag effect, which can accurately and quickly calculate the impact of climate change and human activity factors on water reserves based on time lag effect.

[0007] Another object of the present invention is to provide a dynamic attribution assessment device for terrestrial water reserves based on time lag effect.

[0008] The technical solution of the present invention to solve the above technical problems is:

[0009] A dynamic attribution assessment method for terrestrial water storage based on time lag effect includes the following steps:

[0010] Step S1: obtaining a terrestrial water storage dataset of a target area in a historical period and a dataset of related driving factors of the target area;

[0011] Step S2: Perform grid preprocessing and resampling on the terrestrial water storage dataset and the related driving factor dataset, and output a grid set of terrestrial water storage and driving factors at a uniform temporal and spatial resolution within the study period;

[0012] Step S3: Analyze the spatiotemporal dynamics of terrestrial water storage based on the grid set obtained in step S2, and output the trend and degree of terrestrial water storage changes in the target area within the study period;

[0013] Step S4: By decomposing the seasonal trend, the terrestrial water storage in the target area is decomposed into a trend term, a seasonal term, and a residual term, and the relative contributions of these three components are evaluated to reveal the dominant pattern of the terrestrial water storage in the target area;

[0014] Step S5: Calculate the correlation coefficient of the driving factor of each grid in the grid set by using the cross-correlation analysis algorithm, and perform a significance test on it, so as to screen out the driving factors without obvious time lag effect, and then generate a set of key driving factors;

[0015] Step S6: Combine the partial least squares regression model with the geographic detector to evaluate the degree to which the key driving factors in the time lag effect explain the dynamics of terrestrial water storage and their combined effects.

[0016] As one of the preferred solutions of the present invention, the specific process of step S2 is:

[0017] Step S201: gridding the terrestrial water storage data set and the driving factor data set, wherein the gridding processing includes data format conversion, missing value filling and outlier processing;

[0018] Step S202: resampling each raster data based on a unified spatial resolution. The resampling process interpolates or aggregates the terrestrial water storage dataset and the driving factor dataset in the same spatial grid according to a predetermined spatial grid scale to obtain raster data with a unified spatial resolution.

[0019] Step S203: Time-aligning the terrestrial water storage dataset and the driving factor dataset to ensure that the terrestrial water storage dataset and the driving factor dataset have the same time step according to the time resolution of the research period;

[0020] Step S204: Integrate the terrestrial water storage dataset and driving factor dataset that have undergone raster preprocessing, resampling and time alignment into a raster set with a unified spatiotemporal resolution, and output the raster set as a data file in a specified format for subsequent time lag effect analysis and attribution evaluation.

[0021] As one of the preferred solutions of the present invention, the specific process of step S3 is:

[0022] Step S301: Calculate the change trend of terrestrial water storage in the study area using the Theil-Sen median estimation method;

[0023] Step S302: using the Mann-Kendall trend test method to perform a significance level analysis on the change trend of terrestrial water reserves;

[0024] Step S303: Output the change trend and significance level of the terrestrial water storage of each grid point in the target area within the study period, and classify them according to the direction of the change trend and the degree of significant change, wherein the degree of significant change includes but is not limited to significant increase and significant decrease;

[0025] Step S304: Visualize the trend and significance results of terrestrial water storage, and output a trend spatial distribution map of the target area, showing the degree of increase or decrease of terrestrial water storage in different areas of the target area and their significance distribution.

[0026] As one of the preferred solutions of the present invention, the specific process of step S4 is:

[0027] Step S401: using a local regression analysis algorithm to fit the terrestrial water storage data of the target area, decomposing the original time series into three parts: trend term, seasonal term and residual term;

[0028] Step S402: extracting a trend term, which reflects the long-term change direction and amplitude of the terrestrial water storage in the target area; extracting a seasonal term, which reflects the periodic change characteristics of the terrestrial water storage in the target area, and the seasonal term includes the amplitude and frequency of annual seasonal fluctuations; after removing the trend term and the seasonal term from the original terrestrial water storage data, a residual term is obtained; the residual term represents randomness or unexplainable noise;

[0029] Step S403: Calculate the contribution rate of trend term, seasonal term and residual term to the total change of terrestrial water storage, and analyze the contribution rate of trend term, seasonal term and residual term to reveal the dominant change pattern of terrestrial water storage in the target area, and determine whether the trend term or seasonal term is dominant in a specific area.

[0030] As one of the preferred solutions of the present invention, the specific process of step S5 is:

[0031] Step S501: Perform cross-correlation analysis on the terrestrial water storage time series and the driving factor time series of each grid point in the grid set, calculate the correlation coefficients at different lag times, and perform significance tests to determine whether the correlation at each lag time is statistically significant, so as to identify the time-lag association between terrestrial water storage and each driving factor;

[0032] Step S502: At each lag time point, the correlation coefficients in the grid set are counted, and the number of grid points whose correlation coefficients are within the predetermined interval [a, b] is counted; the lag time with the largest number of grid points whose correlation coefficients are within the predetermined interval [a, b] within the lag time range is selected as the key lag response time of the driving factor.

[0033] As one of the preferred solutions of the present invention, the specific process of step S6 is:

[0034] Step S601: performing regression analysis on the selected key driving factor set based on the partial least squares regression model, and calculating the degree of explanation of each key driving factor on the dynamics of terrestrial water storage;

[0035] Step S602: Calculate the individual contribution of each key driving factor to terrestrial water storage under the time lag effect;

[0036] Step S603: Analyze the spatial heterogeneity of key driving factors and their spatial correlation with terrestrial water storage in combination with the geographic detector algorithm, and evaluate the combined effect of key driving factors; quantitatively analyze the combined effect of key driving factors based on the geographic detector, and evaluate the interactive effects between different key driving factors, so as to identify the synergistic or inhibitory effects of key driving factors on terrestrial water storage in specific spatial regions;

[0037] Step S604: Output the analysis results of the partial least squares regression model and the geographic detector, which include the contribution, combined effect and spatial distribution characteristics of each key driving factor, and finally generate a comprehensive terrestrial water storage dynamic attribution assessment report to reflect the comprehensive explanatory power of each key driving factor under the time lag effect.

[0038] A dynamic attribution assessment device for terrestrial water storage based on time lag effect, comprising:

[0039] Data acquisition unit: obtain terrestrial water storage data, driving factor data, and time and space coordinate information;

[0040] Data processing unit: perform data preprocessing, cleaning, format conversion and resampling;

[0041] Trend analysis unit: Conduct trend analysis on the processed data and use Theil-Sen median estimation method and Mann-Kendall test method to evaluate the long-term trend and significant changes of terrestrial water storage;

[0042] Seasonal trend decomposition unit: decomposes the terrestrial water storage data into trend term, seasonal term and residual term based on the local regression analysis algorithm, and evaluates the relative contribution of potential term, seasonal term and residual term;

[0043] Cross-correlation analysis unit: perform cross-correlation analysis, calculate the correlation coefficient of the driving factors at each grid point, and perform significance test to screen out key driving factors with significant time lag effects;

[0044] Attribution contribution assessment unit: Based on the partial least squares regression model, the screened key driving factors are regressed and analyzed to calculate the degree of explanation of each key driving factor for terrestrial water storage. The geographic detector algorithm is used to evaluate the spatial heterogeneity of key driving factors and their combined effects, and the comprehensive impact of different key driving factors on terrestrial water storage is analyzed.

[0045] An electronic device comprises a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the dynamic attribution assessment method of terrestrial water reserves based on time lag effect.

[0046] A computer-readable storage medium stores a computer program implemented according to the dynamic attribution assessment method for terrestrial water reserves based on time lag effect in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the dynamic attribution assessment method for terrestrial water reserves based on time lag effect are executed.

[0047] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0048] The dynamic attribution assessment method of terrestrial water reserves based on time lag effect of the present invention can more deeply study the changing trend and degree of terrestrial water reserves and the correlation between terrestrial water reserves and various driving factors. Combined with the partial least squares regression model and geographic detector, it can better understand the spatial heterogeneity of key driving factors and their combined effects, so as to more effectively analyze the changing mechanism of terrestrial water reserves. This is expected to provide more precise evidence support for optimal management of water resources, and for coping with challenges in various fields under the background of strong interaction between climate change and human activities, and provide practical reference for the formulation of future sustainable development and water resources management policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flow chart of the dynamic attribution assessment method of terrestrial water reserves based on time lag effect of the present invention.

[0050] Figure 2 It is a structural block diagram of the dynamic attribution evaluation device for terrestrial water reserves based on time lag effect of the present invention.

[0051] Figure 3 A schematic diagram of the structure of a computer device. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0053] Example 1

[0054] See also Figure 1 The dynamic attribution assessment method of terrestrial water reserves based on time lag effect of the present invention comprises the following steps:

[0055] Step S1: obtaining a terrestrial water storage dataset of a target area in a historical period and a dataset of related driving factors of the target area;

[0056] Specifically, the second edition of the CSR_GRACE_GRACE-FO_RL06 gravity field model data recently published by the Space Institute of the University of Texas can be used. This product can effectively overcome the leakage error of traditional spherical harmonic coefficients caused by filtering, and shows higher accuracy than previous traditional spherical harmonic coefficient products.

[0057] Specifically, a preset latitude and longitude range can be selected as the target area, that is, the area to be studied; the image data in the original area measured by satellite can be cropped, for example, the satellite observation data within a certain country can be cropped based on the global satellite observation data. As for the time scale, because the lag response is generally within 0-3 months, the terrestrial water storage data set and driving factor data set of the target area in the target period can be obtained as the monthly time series data in the target area.

[0058] In this embodiment, the driving factors include climate change and human activities. For example, the climate change includes ground temperature, near-ground air pressure, near-ground air specific humidity, near-ground total wind speed, downward short-wave radiation from the ground, downward long-wave radiation from the ground, and ground precipitation; and human activities can include population, per capita GDP, night light data, cultivated land area, etc.

[0059] Step S2: Perform grid preprocessing and resampling on the terrestrial water storage dataset and the related driving factor dataset, and output the grid set of terrestrial water storage and driving factors at a uniform temporal and spatial resolution within the study period, specifically:

[0060] Step S201: gridding the terrestrial water storage data set and the driving factor data set; the gridding processing includes data format conversion (i.e., conversion from the commonly used netcdf4 file to the tif format suitable for image processing), missing value filling and outlier processing, such as linear interpolation processing for historical data, so as to ensure the integrity and consistency of the data;

[0061] Step S202: interpolating or aggregating the terrestrial water storage dataset and the driving factor dataset in the same spatial grid according to a predetermined spatial grid scale to obtain raster data with uniform spatial resolution;

[0062] Step S203: Time-align the terrestrial water storage dataset and the driving factor dataset, and ensure that the terrestrial water storage dataset and the driving factor dataset have the same time step according to the time resolution of the research period;

[0063] Step S204: Integrate the data sets (including the terrestrial water storage data set and the driving factor data set) that have undergone raster preprocessing, resampling and time alignment into a raster set with a uniform spatiotemporal resolution, and output the obtained raster set as a data file in a specified naming format for subsequent batch processing.

[0064] Step S3, based on the grid set obtained in step S204, the spatiotemporal dynamics of terrestrial water storage are analyzed, and the changing trend and degree of terrestrial water storage in the target area within the study period are output, specifically:

[0065] Step S301: Calculate the change trend of terrestrial water storage in the study area based on the Theil-Sen median estimation method; wherein the Theil-Sen median estimation method calculates the median of the rate of change of each pair of time steps to obtain the long-term trend of terrestrial water storage, thereby avoiding the influence of extreme values ​​on trend estimation;

[0066] Step S302: performing a significance level analysis on the change trend of terrestrial water reserves based on the Mann-Kendall trend test method; wherein the Mann-Kendall trend test method is used to test the monotonic trend of the terrestrial water reserves time series and calculate the significance level of the trend to assess whether there is a significant increasing trend or decreasing trend;

[0067] Step S303: outputting the terrestrial water storage change trend and its significance level of each grid point in the target area within the study period, and classifying them according to the direction of the change trend and the degree of significant change, wherein the degree of significant change includes but is not limited to significant increase and significant decrease;

[0068] Step S304: visualize the trend and significance results of terrestrial water storage, output the trend spatial distribution map of the target area or use a more intuitive line graph to show the time change, and display the degree of increase or decrease of terrestrial water storage in different regions and its significance distribution.

[0069] Step S4: Through seasonal trend decomposition, the terrestrial water storage in the target area is decomposed into trend term, seasonal term and residual term, and the relative contribution of the three components is evaluated to reveal the dominant pattern of terrestrial water storage in the target area, which specifically includes the following steps:

[0070] Step S401: fitting the terrestrial water storage data of the target area based on the local regression analysis algorithm, and decomposing the original time series into three parts: long-term trend, seasonal fluctuation and random residual;

[0071] Specifically, the local regression analysis algorithm defines a local window around each time point and uses weighted regression to fit the local trend to smooth the seasonal fluctuations and trend changes in the time series. The seasonal trend decomposition is a time series decomposition method that uses robust local weighted regression as a smoothing method. It has a good effect on decomposing periodic data and is suitable for gravity satellite data.

[0072] X o =X long +X sea +X res ;

[0073] Where: X o is the original time series; X long Represents the long-term trend of the time series; X sea Represents the periodic change law in the time series; X res is the residual term.

[0074] Extracting a long-term trend term from the local regression analysis, the long-term trend term reflects the long-term change direction and amplitude of the terrestrial water storage in the target area; extracting a seasonal term, the seasonal term reflects the periodic change characteristics of the terrestrial water storage in the target area, including the amplitude and frequency of annual seasonal fluctuations; after removing the trend term and the seasonal term from the original terrestrial water storage data, obtaining a residual term, the residual term represents randomness or unexplainable noise;

[0075] Step S402: Calculate the contribution rates of trend terms, seasonal terms and residual terms to the total change of terrestrial water storage, and analyze the contribution rates of trend terms, seasonal terms and residual terms to reveal the dominant change pattern of terrestrial water storage in the target area and determine whether long-term trends or seasonal fluctuations dominate in a specific area.

[0076] Step S5: Calculate the correlation coefficient of the driving factors of each grid in the grid set through the cross-correlation analysis algorithm, and perform a certain level of significance test on it to screen out the driving factors without obvious time lag effect and generate a set of key driving factors; specifically:

[0077] Step S501: Perform cross-correlation analysis on the terrestrial water storage time series and the driving factor time series of each grid point in the grid set, calculate the correlation coefficient r at different lag times, and perform a significance test to determine whether the correlation at each lag time is statistically significant, so as to identify the time-lag association between terrestrial water storage and each driving factor;

[0078] The calculation method of the correlation coefficient r of each grid is as follows:

[0079]

[0080] Where: k is the number of lag months; n is the number of data; G t and Q t-k is the time series data of TWSA and climate indicator; σ G and σ Q are the variances of terrestrial water storage and driving factors, respectively; a 5% significance level was selected to conduct a significance test on the correlation coefficient.

[0081] Step S502: At each lag time point, the correlation coefficients in the grid set are counted, and the number of grid points with correlation coefficients within a predetermined interval [a, b] is counted; the lag time with the largest number of grid points with correlation coefficients within the predetermined interval [a, b] within the lag time range (e.g., 0 to 3 months) is selected as the key lag response time of the driving factor;

[0082] In this embodiment, whether the time lag effect is statistically significant is determined by analyzing the p-value. An insignificant time lag effect may indicate that there is no stable time lag relationship between the driving factor and the response variable.

[0083] Step S6: Combine the partial least squares regression model with the geographic detector to evaluate the degree to which the key driving factors under the time lag effect explain the dynamics of terrestrial water storage and their combined effects. Specifically,

[0084] Step S601: According to the selected key driving factors, the time steps of the dependent variables and independent variables are checked again to prepare for the subsequent evaluation; the partial least squares regression model (PLSR) is used to perform regression analysis on the selected key driving factor set to calculate the degree of explanation of each key driving factor on the dynamics of terrestrial water storage;

[0085] Among them, the partial least squares algorithm mainly includes the following steps: first, obtain the element matrix of the independent variable and the data matrix of the dependent variable, perform standardization processing, and obtain the independent variable matrix X and the dependent variable matrix Y; combine the typical correlation analysis and principal component analysis ideas to find the first pair of principal components, and establish regression models of Y and X for the first pair of principal components respectively; then replace X and Y with the residuals of the regression model; repeat the above steps, and finally establish the following partial least squares regression equation:

[0086]

[0087] Where: k is the axis vector of the principal component; β k The regression coefficient vector for the kth principal component to model the dependent variable principal component regression; k is the number of principal components. Each element in this vector represents the weight of the corresponding principal component in constructing the regression model of the dependent variable.

[0088] For each grid, the terrestrial water storage is used as the dependent variable Y of the partial least squares algorithm, the driving factor considering the optimal lag time is used as the independent variable X of the model, and the determination coefficient R is used. 2 To evaluate the simulation effect of the model and leave-one-out cross-validation to finally determine the optimal model structure. In addition to using the partial least squares algorithm to evaluate the predictive ability of driving factors on terrestrial water storage, the regression coefficient of each driving factor in the partial least squares algorithm describes the weight of each component in the algorithm that depends on the original variable, which can be used to evaluate the contribution of different driving factors to the dynamic changes of terrestrial water storage.

[0089] Combined with the geographic detector algorithm, the spatial heterogeneity of driving factors and their spatial correlation with terrestrial water storage are analyzed to evaluate the combined effect of driving factors; based on the geographic detector, the combined effect of driving factors is quantitatively analyzed to evaluate the interactive effects between different driving factors and identify the synergistic or inhibitory effects of key driving factors on terrestrial water storage in specific spatial regions;

[0090] In this embodiment, the interaction detection function in the geographic detector is mainly used to explore the interaction by comparing the q values ​​of two single variables and their interaction; the formula of the q value is as follows:

[0091]

[0092] Where N v and is the number and variance of observations in the entire study area, N v,j and is the number and variance of observations of variable v in subregion j.

[0093] Five types of interaction relationships between two spatial variables can be explained by the interaction detectors (e.g., nonlinear weakening, univariate weakening, bivariate enhancement, independence, and nonlinear enhancement), as shown in Table 1;

[0094]

[0095] Step S602: Output the partial least squares regression model and the geographic detector analysis results, including the contribution, combined effect and spatial distribution characteristics of each driving factor, and finally generate a comprehensive terrestrial water storage dynamic attribution assessment report to reflect the comprehensive explanatory power of each factor under the time lag effect.

[0096] Example 2

[0097] See also Figure 2 The terrestrial water storage dynamic attribution evaluation device based on time lag effect of the present invention comprises a data acquisition unit, a data processing unit, a trend analysis unit, a seasonal trend decomposition unit, a cross-correlation analysis unit and an attribution contribution evaluation unit, wherein:

[0098] The data acquisition unit is used to obtain terrestrial water storage data, driving factor data, and time and space coordinate information;

[0099] The data processing unit is used for preprocessing, cleaning, format conversion and resampling of data;

[0100] The trend analysis unit is used to perform trend analysis on the processed data, and use the Theil-Sen median estimation method and the Mann-Kendall test method to evaluate the long-term trend and significant changes of terrestrial water storage;

[0101] The seasonal trend decomposition unit is used to decompose the terrestrial water storage data into trend terms, seasonal terms and residual terms using a local regression analysis algorithm, and to evaluate the relative contributions of the potential terms, seasonal terms and residual terms;

[0102] The cross-correlation analysis unit is used to perform cross-correlation analysis, calculate the correlation coefficient of the driving factor of each grid point, and perform a significance test to screen out key driving factors with significant time lag effects;

[0103] The attribution contribution assessment unit performs regression analysis on the screened key driving factors based on the partial least squares regression model, calculates the degree of explanation of each key driving factor for terrestrial water reserves, uses the geographic detector algorithm to evaluate the spatial heterogeneity of the key driving factors and their combined effects, and analyzes the comprehensive impact of different key driving factors on terrestrial water reserves.

[0104] Example 3

[0105] See also Figure 3This embodiment also provides an electronic device, which can be implemented by a computer device, such as Figure 3 As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions, and the database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a dynamic attribution assessment method for terrestrial water reserves based on a time lag effect. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor can execute the dynamic attribution assessment method for terrestrial water reserves based on a time lag effect of the present application. The network interface of the computer device is used to connect and communicate with a terminal.

[0106] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0107] In this embodiment, the processor is used to execute Figure 2 The memory stores the program code and various data required to execute the above modules or submodules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the rice field weed detection device of the present application, and the server can call the program code and data of the server to execute the functions of all submodules.

[0108] This embodiment also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the dynamic attribution assessment method of terrestrial water reserves based on time lag effect described in any embodiment of the present application.

[0109] This embodiment also provides a computer program product, including a computer program / instruction, which, when executed by one or more processors, implements the steps of the dynamic attribution assessment method of terrestrial water reserves based on time lag effect described in any embodiment of the present application.

[0110] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0111] The above is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited to the above content. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A dynamic attribution assessment method for terrestrial water reserves based on time lag effect, characterized in that: The following steps are involved: Step S1: obtaining a terrestrial water storage dataset of a target area in a historical period and a dataset of related driving factors of the target area; Step S2: Perform grid preprocessing and resampling on the terrestrial water storage dataset and the related driving factor dataset, and output a grid set of terrestrial water storage and driving factors at a uniform temporal and spatial resolution within the study period; Step S3: Analyze the spatiotemporal dynamics of terrestrial water storage based on the grid set obtained in step S2, and output the change trend and degree of terrestrial water storage in the target area within the study period; Step S4: By decomposing the seasonal trend, the terrestrial water storage in the target area is decomposed into a trend term, a seasonal term, and a residual term, and the relative contributions of these three components are evaluated to reveal the dominant pattern of the terrestrial water storage in the target area; Step S5: Calculate the correlation coefficient of the driving factor of each grid in the grid set by using the cross-correlation analysis algorithm, and perform a significance test on it, so as to screen out the driving factors without obvious time lag effect, and then generate a set of key driving factors; Step S6: Combine the partial least squares regression model with the geographic detector to evaluate the degree to which the key driving factors in the time lag effect explain the dynamics of terrestrial water storage and their combined effects.

2. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 1 is characterized in that: The specific process of step S2 is: Step S201: gridding the terrestrial water storage data set and the driving factor data set, wherein the gridding processing includes data format conversion, missing value filling and outlier processing; Step S202: resampling each raster data based on a unified spatial resolution. The resampling process interpolates or aggregates the terrestrial water storage dataset and the driving factor dataset in the same spatial grid according to a predetermined spatial grid scale to obtain raster data with a unified spatial resolution. Step S203: Time-aligning the terrestrial water storage dataset and the driving factor dataset to ensure that the terrestrial water storage dataset and the driving factor dataset have the same time step according to the time resolution of the research period; Step S204: Integrate the terrestrial water storage dataset and driving factor dataset that have undergone raster preprocessing, resampling and time alignment into a raster set with a unified spatiotemporal resolution, and output the raster set as a data file in a specified format for subsequent time lag effect analysis and attribution evaluation.

3. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 2 is characterized in that: The specific process of step S3 is: Step S301: Calculate the change trend of terrestrial water storage in the study area using the Theil-Sen median estimation method; Step S302: using the Mann-Kendall trend test method to perform a significance level analysis on the change trend of terrestrial water reserves; Step S303: output the change trend and significance level of the terrestrial water storage of each grid point in the target area within the study period, and classify them according to the direction of the change trend and the degree of significant change; Step S304: Visualize the trend and significance results of terrestrial water storage, and output a trend spatial distribution map of the target area, showing the degree of increase or decrease of terrestrial water storage in different areas of the target area and their significance distribution.

4. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 3 is characterized in that: The specific process of step S4 is as follows: Step S401: using a local regression analysis algorithm to fit the terrestrial water storage data of the target area, decomposing the original time series into three parts: trend term, seasonal term and residual term; Step S402: extracting trend items, which reflect the long-term change direction and amplitude of the terrestrial water storage in the target area; extracting seasonal items, which reflect the periodic change characteristics of the terrestrial water storage in the target area, including the amplitude and frequency of annual seasonal fluctuations; after removing the trend items and seasonal items from the original terrestrial water storage data, obtaining the residual items; The residual term represents randomness or unexplainable noise; Step S403: Calculate the contribution rate of trend term, seasonal term and residual term to the total change of terrestrial water storage, and analyze the contribution rate of trend term, seasonal term and residual term to reveal the dominant change pattern of terrestrial water storage in the target area, and determine whether the trend term or seasonal term is dominant in a specific area.

5. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 4 is characterized in that: The specific process of step S5 is as follows: Step S501: Perform cross-correlation analysis on the terrestrial water storage time series and the driving factor time series of each grid point in the grid set, calculate the correlation coefficients at different lag times, and perform significance tests to determine whether the correlation at each lag time is statistically significant, so as to identify the time-lag association between terrestrial water storage and each driving factor; Step S502: At each lag time point, the correlation coefficients in the grid set are counted, and the number of grid points whose correlation coefficients are within the predetermined interval [a, b] is counted; the lag time with the largest number of grid points whose correlation coefficients are within the predetermined interval [a, b] within the lag time range is selected as the key lag response time of the driving factor.

6. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 5 is characterized in that: The specific process of step S6 is as follows: Step S601: performing regression analysis on the selected key driving factor set based on the partial least squares regression model, and calculating the degree of explanation of each key driving factor on the dynamics of terrestrial water storage; Step S602: Calculate the individual contribution of each key driving factor to terrestrial water storage under the time lag effect; Step S603: Analyze the spatial heterogeneity of key driving factors and their spatial correlation with terrestrial water storage in combination with the geographic detector algorithm to evaluate the combined effect of key driving factors; Quantitatively analyze the combined effects of key driving factors based on geographic detectors and evaluate the interactive effects between different key driving factors, so as to identify the synergistic or inhibitory effects of key driving factors on terrestrial water storage in specific spatial regions; Step S604: Output the analysis results of the partial least squares regression model and the geographic detector, which include the contribution, combined effect and spatial distribution characteristics of each key driving factor, and finally generate a comprehensive terrestrial water storage dynamic attribution assessment report to reflect the comprehensive explanatory power of each key driving factor under the time lag effect.

7. The dynamic attribution assessment method for terrestrial water reserves based on time lag effect according to claim 1 is characterized in that: In step S1, the driving factors include climate change and human activities.

8. A device for implementing the method for dynamic attribution assessment of terrestrial water reserves based on time lag effect as described in any one of claims 1 to 7, characterized in that: include: Data acquisition unit: obtain terrestrial water storage data, driving factor data, and time and space coordinate information; Data processing unit: perform data preprocessing, cleaning, format conversion and resampling; Trend analysis unit: Conduct trend analysis on the processed data and use the Theil-Sen median estimation method and the Mann-Kendal l test method to evaluate the long-term trend and significant changes of terrestrial water storage; Seasonal trend decomposition unit: decomposes the terrestrial water storage data into trend term, seasonal term and residual term based on the local regression analysis algorithm, and evaluates the relative contribution of potential term, seasonal term and residual term; Cross-correlation analysis unit: perform cross-correlation analysis, calculate the correlation coefficient of the driving factor at each grid point, and perform significance test to screen out key driving factors with significant time lag effects; Attribution contribution assessment unit: Based on the partial least squares regression model, the screened key driving factors are regressed and analyzed to calculate the degree of explanation of each key driving factor for terrestrial water storage. The geographic detector algorithm is used to evaluate the spatial heterogeneity of key driving factors and their combined effects, and the comprehensive impact of different key driving factors on terrestrial water storage is analyzed.

9. An electronic device, comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the dynamic attribution assessment method of terrestrial water reserves based on time lag effect as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: It stores a computer program implemented according to the dynamic attribution assessment method for terrestrial water reserves based on time lag effect as described in any one of claims 1 to 7 in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the dynamic attribution assessment method for terrestrial water reserves based on time lag effect are executed.

Images