Root zone soil moisture estimation method based on remote sensing driving and moisture hysteresis characteristic simulation, electronic device and medium

Abstract

The application discloses a root zone soil moisture estimation method based on remote sensing driving and moisture hysteresis characteristic simulation, an electronic device and a medium. The method can comprise: acquiring remote sensing data with the same time resolution and spatial resolution, including remote sensing surface soil moisture data, remote sensing evapotranspiration data, potential evapotranspiration data, rainfall data and saturated soil humidity; calculating root zone soil humidity based on the remote sensing surface soil moisture data, and then calculating an estimated value of the evaporation ratio based on the root zone soil humidity and the saturated soil humidity; calculating the remote sensing evaporation ratio based on the remote sensing evapotranspiration data and the potential evapotranspiration data; establishing a target function to calculate an error, and when the error is less than a set threshold, outputting the root zone soil humidity. The application introduces the remote sensing evaporation ratio to exert physical constraints to embody the influence of vegetation and improve physical rationality, and constructs a multi-constraint target function optimization model parameter to improve the estimation accuracy and reliability of the remote sensing evaporation ratio.

Description

Technical Field

[0001] This invention relates to the field of remote sensing hydrology, and more specifically, to a method, electronic device, and medium for estimating root zone soil moisture based on remote sensing-driven and water lag characteristic simulation. Background Technology

[0002] Remote sensing methods for estimating root zone soil moisture can be broadly categorized into empirical methods, semi-empirical methods, physical mechanism-based methods, and machine learning methods. Empirical methods primarily rely on statistical models to establish a mathematical relationship between surface soil moisture (SSM) or remote sensing-derived indices and root zone soil moisture (RZSM) using historical observation data. Common empirical methods include statistical regression, cross-correlation analysis, and cumulative distribution function matching (CDF matching). Statistical regression methods mainly utilize drought indices extracted from remote sensing data (such as Temperature Vegetation Drought Index (TVDI), Vegetation Water Supply Index (VSWI), and Vegetation Temperature Condition Index (VTCI)) to predict RZSM through regression analysis. Cross-correlation analysis is used to study the lag relationship between SSM and RZSM, determining the degree of influence of surface moisture signals on root zone soil moisture. The CDF method estimates RZSM by matching the cumulative probability distributions of SSM and RZSM, also relying on the correlation between them. It is evident that these methods are highly dependent on the correlation between SSM and RZSM and historical observation data, and struggle to characterize the soil moisture transport mechanism within the profile, exhibiting limited generalization ability under conditions of data scarcity or climatic change.

[0003] Semi-empirical methods combine empirical statistical methods with simplified physical processes to improve the physical plausibility of the model. The Exponential Filter (ExpF) is a typical example of a semi-empirical method. Semi-empirical methods strike a balance between computational efficiency and physical plausibility, effectively estimating RZSM in regional-scale applications. While not as accurate as fully physical models, they reduce computational complexity by simplifying physical processes and can be adjusted and optimized using observational data and machine learning methods, thus providing reasonable estimation results under various environmental conditions. However, the application of the Exponential Filter still heavily relies on measured RZSM data, and because it does not consider the impact of vegetation transpiration on soil moisture, it may introduce biases in areas with significant vegetation regulation or complex climatic conditions.

[0004] Physically based modeling methods primarily solve Richard's equations or unsaturated hydrodynamic equations for soil moisture to determine root zone soil moisture. Many models also incorporate remotely sensed surface soil moisture or land surface temperature data into the physical model to improve estimation accuracy. While the physical processes are relatively robust, the models suffer from high computational costs due to numerous parameters, demanding input data, and difficulties in calibration. They are also highly sensitive to data quality and initial conditions, leading to certain uncertainties in regional-scale applications.

[0005] With the rapid development of artificial intelligence technology, machine learning methods have been widely used in RZSM estimation in recent years. Machine learning can capture the complex nonlinear relationship between soil moisture and all relevant predictors (such as meteorological variables, land cover, and soil parameters). Common models include Random Forest (RF), Artificial Neural Network (ANN), Long Short-Term Memory Network (LSTM), and Extreme Gradient Boosting (XGBoost). However, machine learning methods are highly dependent on ground-based measured data as training targets, while ground-based RZSM observation sites are scarce and data is difficult to obtain, thus limiting the widespread application of machine learning methods.

[0006] Therefore, it is necessary to develop a root zone soil moisture estimation method, electronic equipment, and media based on remote sensing and simulation of water lag characteristics.

[0007] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0008] This invention proposes a root zone soil moisture estimation method, electronic device, and medium based on remote sensing-driven and water lag characteristic simulation. It can improve the physical rationality by introducing the remote sensing evaporation ratio to apply physical constraints to reflect the impact of vegetation, and construct a multi-constraint objective function to optimize model parameters, thereby improving the estimation accuracy and reliability of the remote sensing evaporation ratio.

[0009] In a first aspect, embodiments of this disclosure provide a root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation, including: Acquire remote sensing data with the same temporal and spatial resolution, including remote sensing surface soil moisture data, remote sensing evapotranspiration data, potential evapotranspiration data, rainfall data, and saturated soil moisture; Based on remote sensing surface soil moisture data, the root zone soil moisture is calculated, and then the evaporation ratio is estimated based on the root zone soil moisture and saturated soil moisture. The remote sensing evapotranspiration ratio is calculated based on remote sensing evapotranspiration data and potential evapotranspiration data. Establish an objective function to calculate the error. When the error is less than a set threshold, output the soil moisture in the root zone.

[0010] Preferably, based on remote sensing surface soil moisture data, the initial value of the exponential filtering method is set to 1, and the soil moisture in the root zone is calculated.

[0011] Preferably, the estimated value of the evaporation ratio is:

[0012] in, This is an estimated value for the evaporation ratio. This represents saturated soil moisture, and a and b are empirical coefficients of the EF-RZSM relationship. This refers to the soil moisture in the root zone.

[0013] Preferably, the remote sensing evaporation ratio is:

[0014] Where ET represents remotely sensed evapotranspiration data, PET represents potential evapotranspiration data, and EF represents potential evapotranspiration data. rs This is the remote sensing evaporation ratio.

[0015] Preferably, the objective function is:

[0016] Where J is the error, Here, P represents the estimated remotely sensed evapotranspiration ratio, ET represents the rainfall data, and Var(·) represents the remotely sensed evapotranspiration data. This represents the second time difference between the estimated values ​​of soil moisture in the root zone. , Let N be the soil moisture in the root zone at time t, and N be the number of samples. It is the first-order time difference of soil moisture in the root zone, representing the short-term changes in soil moisture. It is the second-order time difference of surface soil moisture. This indicates the accuracy constraint for fitting the evaporation ratio. This indicates a constraint on the consistency of water balance trends. This indicates a temporal smoothing constraint on soil moisture in the root zone.

[0017] Preferably, if the error is not less than the set threshold, the initial value of the exponential filtering method and the empirical coefficient of the EF-RZSM relationship are adjusted, and the above steps are repeated until the error is less than the set threshold, and the root zone soil moisture is output.

[0018] Preferably, if the temporal resolution and / or spatial resolution of the acquired remote sensing data are different, resampling is performed.

[0019] Secondly, embodiments of this disclosure also provide an electronic device, the electronic device comprising: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement the root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation.

[0020] Thirdly, embodiments of this disclosure also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned root zone soil moisture estimation method based on remote sensing-driven and water hysteresis characteristic simulation.

[0021] Its beneficial effects are as follows: This invention does not require the support of measured RZSM data, overcoming the limitation of the original ExpF method which relies on measured data to calibrate T-values. Simultaneously, by introducing physical constraints through the remotely sensed evaporation ratio (EF) as a diagnostic variable, it effectively characterizes the consumption and regulatory effects of vegetation transpiration on soil moisture, compensating for the shortcomings of the exponential filtering method in not considering the influence of vegetation, thereby improving the physical rationality of RZSM estimation. Furthermore, by constructing an objective function that considers the fitting accuracy of evaporation ratio, the consistency of water balance trends, and the temporal smoothness of root zone soil moisture, multi-constraint optimization of model parameters is achieved, improving the reliability and stability of parameter calibration, thereby further enhancing the accuracy of RZSM estimation.

[0022] The methods and apparatus of the present invention have other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0023] The above and other objects, features and advantages of the present invention will become more apparent from the more detailed description of exemplary embodiments of the invention in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same parts.

[0024] Figure 1 A flowchart illustrating the steps of a root zone soil moisture estimation method based on remote sensing-driven simulation of water lag characteristics according to an embodiment of the present invention is shown. Detailed Implementation

[0025] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention are described below, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0026] To facilitate understanding of the solutions and effects of the embodiments of the present invention, three specific application examples are given below. Those skilled in the art should understand that these examples are merely for the purpose of understanding the present invention, and any specific details therein are not intended to limit the present invention in any way.

[0027] Example 1

[0028] Figure 1 A flowchart illustrating the steps of a root zone soil moisture estimation method based on remote sensing-driven simulation of water lag characteristics according to an embodiment of the present invention is shown.

[0029] like Figure 1 As shown, the root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation includes: Step 101: Acquire remote sensing data with the same temporal and spatial resolution, including remote sensing surface soil moisture data, remote sensing evapotranspiration data, potential evapotranspiration data, rainfall data, and saturated soil moisture. Step 102: Calculate the root zone soil moisture based on remote sensing surface soil moisture data, and then calculate the estimated value of the evaporation ratio based on the root zone soil moisture and saturated soil moisture. Step 103: Calculate the remote sensing evapotranspiration ratio based on remote sensing evapotranspiration data and potential evapotranspiration data; Step 104: Establish the objective function to calculate the error. When the error is less than the set threshold, output the soil moisture in the root zone.

[0030] In one example, based on remote sensing surface soil moisture data, the initial value of the exponential filtering method is set to 1 to calculate the soil moisture in the root zone.

[0031] In one example, the estimated evaporation ratio is:

[0032] in, This is an estimated value for the evaporation ratio. This represents saturated soil moisture, and a and b are empirical coefficients of the EF-RZSM relationship. This refers to the soil moisture in the root zone.

[0033] In one example, the remotely sensed evaporation ratio is:

[0034] Where ET represents remotely sensed evapotranspiration data, PET represents potential evapotranspiration data, and EF represents potential evapotranspiration data. rs This is the remote sensing evaporation ratio.

[0035] In one example, the objective function is:

[0036] Where J represents the error, P represents the rainfall data, ET represents the remote sensing evapotranspiration data, and Var(·) represents the variance. This represents the second time difference between the estimated values ​​of soil moisture in the root zone. , Let N be the soil moisture in the root zone at time t, and N be the number of samples. It is the first-order time difference of soil moisture in the root zone, representing the short-term changes in soil moisture. It is the second-order time difference of surface soil moisture. This indicates the accuracy constraint for fitting the evaporation ratio. This indicates a constraint on the consistency of water balance trends. This indicates a temporal smoothing constraint on soil moisture in the root zone.

[0037] In one example, if the error is not less than the set threshold, the initial value of the exponential filtering method and the empirical coefficient of the EF-RZSM relationship are adjusted, and the above steps are repeated until the error is less than the set threshold, and the root zone soil moisture is output.

[0038] In one example, if the temporal and / or spatial resolutions of the acquired remote sensing data are different, resampling is performed.

[0039] Specifically, the root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation of the present invention includes the following steps: 1. Collect remote sensing data products with the same temporal and spatial resolution, including: ① The remote sensing surface soil moisture data product SSM is used to drive the Expof method, that is, it serves as the input to the Expof method; ② Remote sensing evapotranspiration data product ET and ③ Potential evapotranspiration data product PET are used to calculate the evaporation ratio (EF). rs ) as a diagnostic variable; ④ Rainfall data P is used to calculate P-ET, which is the difference between rainfall and evapotranspiration, as the available surface water. ⑤ Saturated soil moisture .

[0040] When the temporal and / or spatial resolutions of the above data are different, resampling is performed to ensure that the temporal and spatial resolutions are the same.

[0041] 2. Calculation of root zone soil moisture based on the Exponential Filtering (ExpF) method: Using remote sensing surface soil moisture data product SSM as input, the initial value of the parameter (T value) in the ExpoF method is set to 1, and the root zone soil moisture RZSM is estimated using the ExpoF method.

[0042] 3. Estimating EF based on EF-RZSM relationship es: Using the root zone soil moisture RZSM, the estimated value of EF (EF) was obtained based on the relationship between EF and RZSM. es ):

[0043] in This represents saturated soil moisture. a and b are empirical coefficients of the EF-RZSM relationship, with initial values ​​set to 1 and 0.421, respectively. This refers to the soil moisture in the root zone.

[0044] 4. Calculate the remotely sensed evaporation ratio EF rs : The remotely sensed evaporation ratio EF is calculated from the remotely sensed evaporation data product ET and the potential evaporation data product PET. rs : .

[0045] 5. Calculate the estimation error based on the objective function:

[0046] The objective function consists of three dimensionless terms, corresponding to the evaporation ratio fitting accuracy constraint, the water balance trend consistency constraint, and the root zone soil moisture temporal smoothing constraint, respectively. All three terms are standardized by variance to avoid manual weight setting. Var(·) represents the variance, and P is the rainfall data. This represents the second-order time difference of the estimated soil moisture values ​​in the root zone, where N is the sample size. It is the first-order time difference of soil moisture in the root zone, representing the short-term changes in soil moisture. It is the second-order time difference of surface soil moisture. This indicates the accuracy constraint for fitting the evaporation ratio. This indicates a constraint on the consistency of water balance trends. This indicates a temporal smoothing constraint on soil moisture in the root zone.

[0047] 6. Parameter tuning based on estimation error. The calculated... The smaller the value, the higher the EF. es The closer to the 'true value', the more... EF es The high accuracy indicates that the RZSM estimated in step 3 also has high accuracy, and the output RZSM is taken as the final RZSM estimation result. When EF es The error is large, requiring parameter readjustment, including the parameter T value of the exponential filtering method in step 2 and the empirical coefficients a and b of the EF-RZSM relationship in step 3. Parameter tuning methods can include particle swarm optimization algorithms and genetic algorithms. Repeat steps 2 to 6 until… The RZSM obtained at this point is the final estimation result, which is then output and saved.

[0048] This invention integrates remote sensing evaporation ratio with an improved exponential filtering method to construct a root zone soil moisture estimation method based on remote sensing and simulation of water lag characteristics. Its main advantages are: ① By using the evaporation ratio that can be obtained by remote sensing as the objective function, the existing methods are freed from their dependence on measured RZSM data, thus overcoming the bottleneck problem of existing methods in estimating RZSM. ② By introducing physical constraints using remote sensing evaporation ratio (EF) as a diagnostic variable, the consumption and regulation of soil moisture by vegetation transpiration can be effectively characterized, making up for the shortcomings of the exponential filtering method in not considering the influence of vegetation, thereby improving the physical rationality of RZSM estimation.

[0049] ③ This method has physical rationality and strong operability. The coupling relationship between evaporation ratio and RZSM reflects the consumption of soil moisture by vegetation transpiration, and the exponential filtering method characterizes the transmission process of surface soil moisture and deep soil moisture. Therefore, this method has physical support. At the same time, using remote sensing data products as input greatly improves the operability of the model and enables spatiotemporally continuous RZSM estimation.

[0050] Example 2

[0051] This disclosure provides an electronic device, comprising: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the aforementioned root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation.

[0052] An electronic device according to an embodiment of the present disclosure includes a memory and a processor.

[0053] This memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0054] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory.

[0055] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0056] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0057] Example 3

[0058] This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the root zone soil moisture estimation method based on remote sensing-driven and water hysteresis characteristic simulation.

[0059] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods described in the foregoing embodiments of the present disclosure are performed.

[0060] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0061] Those skilled in the art should understand that the above description of the embodiments of the present invention is only intended to illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any of the examples given.

[0062] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for estimating root zone soil moisture based on remote sensing and simulation of water lag characteristics, characterized in that, include: Acquire remote sensing data with the same temporal and spatial resolution, including remote sensing surface soil moisture data, remote sensing evapotranspiration data, potential evapotranspiration data, rainfall data, and saturated soil moisture; Based on remote sensing surface soil moisture data, the root zone soil moisture is calculated, and then the evaporation ratio is estimated based on the root zone soil moisture and saturated soil moisture. The remote sensing evapotranspiration ratio is calculated based on remote sensing evapotranspiration data and potential evapotranspiration data. Establish an objective function to calculate the error. When the error is less than a set threshold, output the soil moisture in the root zone.

2. The root zone soil moisture estimation method based on remote sensing and water lag characteristic simulation according to claim 1, wherein, Based on remote sensing surface soil moisture data, the initial value of the exponential filtering method was set to 1, and the soil moisture in the root zone was calculated.

3. The root zone soil moisture estimation method based on remote sensing and water lag characteristic simulation according to claim 2, wherein, The estimated value of the evaporation ratio is: in, This is an estimated value for the evaporation ratio. This represents saturated soil moisture, and a and b are empirical coefficients of the EF-RZSM relationship. This refers to the soil moisture in the root zone.

4. The root zone soil moisture estimation method based on remote sensing and water lag characteristic simulation according to claim 3, wherein, The remote sensing evaporation ratio is: Where ET represents remotely sensed evapotranspiration data, PET represents potential evapotranspiration data, and EF represents potential evapotranspiration data. rs This is the remote sensing evaporation ratio.

5. The root zone soil moisture estimation method based on remote sensing and water lag characteristic simulation according to claim 4, wherein, The objective function is: Where J represents the error, P represents the rainfall data, ET represents the remote sensing evapotranspiration data, and Var(·) represents the variance. This represents the second time difference between the estimated values ​​of soil moisture in the root zone. , Let N be the soil moisture in the root zone at time t, and N be the number of samples. It is the first-order time difference of soil moisture in the root zone, representing the short-term changes in soil moisture. It is the second-order time difference of surface soil moisture. This indicates the accuracy constraint for fitting the evaporation ratio. This indicates a constraint on the consistency of water balance trends. This indicates a temporal smoothing constraint on soil moisture in the root zone.

6. The root zone soil moisture estimation method based on remote sensing and water lag characteristic simulation according to claim 5, wherein, If the error is not less than the set threshold, adjust the initial value of the exponential filtering method and the empirical coefficient of the EF-RZSM relationship, repeat the above steps until the error is less than the set threshold, and output the root zone soil moisture.

7. The root zone soil moisture estimation method based on remote sensing-driven and water lag characteristic simulation according to any one of claims 1-6, wherein, If the temporal and / or spatial resolutions of the acquired remote sensing data are different, resampling is performed.

8. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement the root zone soil moisture estimation method based on remote sensing-driven and water hysteresis characteristic simulation as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the root zone soil moisture estimation method based on remote sensing-driven and water hysteresis characteristic simulation as described in any one of claims 1-7.

Images