Vegetation evapotranspiration change dynamic monitoring and driving force identification method and system

Abstract

The invention provides a vegetation evapotranspiration change dynamic monitoring and driving force identification method and system, and the method comprises the steps: obtaining a multi-time-sequence remote sensing image which comprises remote sensing image data and environment influence factors; the obtained data is preprocessed; carrying out evapotranspiration inversion by utilizing remote sensing image data based on a remote sensing evapotranspiration model; smoothing the evapotranspiration time sequence data by adopting a local weighted scatter smoothing algorithm, removing noise and revealing a long-term trend; the Hurst index of the evapotranspiration time sequence is calculated, the vegetation evapotranspiration change durability is evaluated, and the future trend is judged; and performing evapotranspiration analysis based on the evapotranspiration data and the influence factors, calculating the contribution rate of each factor to the vegetation evapotranspiration change, and identifying main driving factors of the vegetation evapotranspiration change. According to the method, the dynamic change characteristics of vegetation evapotranspiration can be accurately captured, the influence of different environmental factors is effectively distinguished, and technical support is provided for regional ecological system monitoring, environmental governance and scientific decision making.

Description

Technical Field

[0001] The invention relates to the technical field of vegetation monitoring, and in particular to a method and system for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces. Background Art

[0002] Evapotranspiration is an important process of material and energy exchange between ecosystems and the atmosphere, and is one of the important indicators for assessing regional water cycles, carbon cycles, and dynamic changes in ecosystems. Accurately grasping the dynamic characteristics and main driving forces of vegetation evapotranspiration will help us to deeply understand the response and adaptation of ecosystems to environmental factors, and provide a scientific basis for water resources management, ecological environmental protection, and agricultural production regulation. However, traditional evapotranspiration monitoring methods usually rely on ground observation sites, with limited data coverage and difficulty meeting the needs of large-scale, long-term dynamic monitoring.

[0003] The development of remote sensing technology has provided a new way to monitor vegetation evapotranspiration at regional and even global scales, and can achieve high temporal and spatial resolution inversion of the surface evapotranspiration process, significantly improving the breadth and accuracy of monitoring. However, in the dynamic monitoring of remote sensing evapotranspiration, how to eliminate time series noise, reveal long-term trends, and distinguish the relative contribution of various environmental factors to evapotranspiration changes are still hot topics and difficulties in current research.

[0004] Existing studies usually use simple correlation analysis or regression analysis to identify the driving forces of evapotranspiration changes. This method is difficult to fully reveal the comprehensive impact of complex interactions of multiple factors on evapotranspiration. In addition, the comprehensive driving effects of environmental factors on ecosystems are often nonlinear, spatiotemporally heterogeneous, and lagging, and traditional statistical analysis is difficult to fully evaluate their relative contributions. Therefore, it is of great scientific significance and application value to develop a new method that integrates multiple analysis methods to accurately monitor the dynamic changes of evapotranspiration and quantitatively identify driving forces. Summary of the invention

[0005] In view of the technical problems existing in the prior art, the present invention provides a method and system for dynamically monitoring vegetation evapotranspiration changes and identifying driving forces, so as to obtain more accurate and detailed spatiotemporal evolution characteristics of vegetation evapotranspiration changes, and to quantitatively analyze the specific contribution of various environmental factors and driving forces to vegetation changes.

[0006] According to a first aspect of the present invention, the present invention provides a method for dynamically monitoring changes in vegetation evapotranspiration and identifying driving forces, comprising the following steps: Acquire multi-time series remote sensing images, including: remote sensing image data and environmental impact factor data; Preprocess the acquired remote sensing image data and environmental impact factor data; Based on the remote sensing evapotranspiration model, evapotranspiration inversion is performed using remote sensing image data; The local weighted scatter point smoothing algorithm is used to smooth the time series data of evapotranspiration to remove noise and reveal long-term trends; Calculate the Hurst index of the evapotranspiration time series to assess the persistence of vegetation evapotranspiration changes and determine future trends; Evapotranspiration analysis was conducted based on evapotranspiration data and influencing factors, the contribution rate of each factor to the change of vegetation evapotranspiration was calculated, and the main driving factors of the change of vegetation evapotranspiration were identified.

[0007] Based on the above technical solution, the present invention can also make the following improvements.

[0008] Preferably, the remote sensing image data include MODIS and Landsat images, and the environmental influencing factor data include precipitation and solar radiation meteorological data.

[0009] Preferably, the preprocessing includes denoising, resampling, projection correction and normalization.

[0010] Preferably, the remote sensing evapotranspiration model is a method for calculating surface evapotranspiration by an energy balance method, wherein the surface evapotranspiration is calculated as follows:

[0011] In the formula, is the 24-hour latent heat flux, w / m 2 , is the 24-hour evaporation ratio; is the 24-hour net radiation, is the latent heat of evaporation coefficient.

[0012] Preferably, the smoothing process of the time series data of evapotranspiration using a local weighted scatter point smoothing algorithm comprises: Select a local neighborhood, for each prediction point , select the K data points closest to it from the data set as the local neighborhood, where K is an odd number; Calculate neighborhood data With prediction point The distance between And convert the distance into weight ; Perform weighted fitting according to the weights , for data points in the neighborhood Weighted fitting was performed using the least squares method.

[0013] Preferably, the Hurst index is calculated using the following formula:

[0014] Where: Indicates the length of the time series or the number of data points; Indicates that when the time series length is The rescaled range at , that is, the range of the cumulative sum of the deviation sequence; Indicates that when the time series length is The standard deviation at time , used to standardize the rescaled range ; is the Hurst exponent, when When , it indicates that the KNDVI time series shows persistence, that is, the future change trend is likely to continue the past trend; when When , the kernel normalized difference vegetation index time series shows randomness, indicating that the future trend has nothing to do with the past; When This indicates that the KNDVI time series has anti-persistence, that is, the future trend may be opposite to the past. when The closer it is to 1, the stronger the persistence of the time series. Conversely, the closer it is to 0, the more significant the anti-persistence is.

[0015] Preferably, the evapotranspiration analysis is performed based on the evapotranspiration data and the influencing factors, and the contribution rate of each factor to the change of vegetation evapotranspiration is calculated, which includes: Normalize each set of data and calculate different units of different variables together; Vegetation evapotranspiration and environmental factors were input into the ridge regression model; The relative and absolute contributions of each factor to evapotranspiration were calculated using the ridge regression coefficients and the standardized trends of environmental factors.

[0016] Preferably, the following formula is used to calculate the relative and absolute contribution of each factor to evapotranspiration:

[0017]

[0018]

[0019] in, is the relative contribution of the first environmental factor, is the coefficient, The detrended value of the first environmental factor, The relative contribution ratio of the first environmental factor, is the relative contribution of the second environmental factor, is the relative contribution of the third environmental factor, is the relative contribution of the nth environmental factor, is the absolute contribution, The non-trend component of vegetation evapotranspiration, Trend component of vegetation evapotranspiration.

[0020] According to a second aspect of the present invention, a system for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces is provided, comprising: Data acquisition module, used to obtain multi-time series remote sensing images, including: remote sensing image data and environmental impact factor data; Data processing module, used for preprocessing the collected data, including denoising, resampling, projection correction and normalization; Evapotranspiration inversion module, used to calculate vegetation evapotranspiration based on remote sensing evapotranspiration model; Trend analysis module, which is used to smooth the evapotranspiration time series using the local weighted scatter point smoothing method to reveal the long-term trend; The persistence assessment module is used to analyze the evapotranspiration time series through the Hurst index and assess the persistence of its changes and future trends; The driving force identification module is used to analyze the relative and absolute contributions of environmental factors to evapotranspiration changes through ridge regression and identify the main driving factors; The display and output module is used to output the analysis results in the form of graphics and reports to facilitate visual interpretation and scientific decision-making.

[0021] According to a third aspect of the present invention, an electronic device is provided, comprising a memory and a processor, wherein the processor is configured to implement steps of a method for dynamically monitoring changes in vegetation evapotranspiration and identifying driving forces when executing a computer program stored in the memory.

[0022] Technical effects and advantages of the present invention: The present invention provides a method and system for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces. The method collects multi-source information including MODIS, Landsat images, and environmental data such as precipitation and solar radiation; and performs refined preprocessing on the collected data to ensure the quality and uniformity of the data; accurately calculates the evapotranspiration of vegetation based on the evapotranspiration inversion module; uses local weighted scatter point smoothing technology to smooth the evapotranspiration time series data to reveal its long-term change trend; evaluates the persistence of evapotranspiration changes and its future trend through Hurst index analysis; uses ridge regression analysis method to quantify the contribution of environmental factors to evapotranspiration changes and identify key driving factors; and intuitively displays the analysis results in the form of charts and reports, which is convenient for users to make visual interpretations and scientific decisions. The present invention can accurately capture the dynamic change characteristics of vegetation evapotranspiration and effectively distinguish the influence of different environmental factors, providing important technical support for regional ecosystem monitoring, environmental governance and scientific decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of a method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces provided by an embodiment of the present invention; Figure 2 is a flow chart of a method for estimating evapotranspiration based on a remote sensing evapotranspiration model provided by an embodiment of the present invention; Figure 3 is a schematic diagram of the dynamic monitoring results of evapotranspiration changes in the Loess Plateau agricultural area provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the identification results of the driving force of evapotranspiration changes in the Loess Plateau agricultural area provided by an embodiment of the present invention; Figure 5 This is a schematic diagram of the dominant factor results of annual evapotranspiration changes provided by an embodiment of the present invention; Figure 6 is a block diagram of an exemplary system for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces provided by an embodiment of the present invention; Figure 7 It is an example block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] It is understandable that based on the defects in the background technology, the embodiment of the present invention proposes a method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces, specifically as follows: Figure 1 As shown, the following steps are included: S1. Acquire multi-time series remote sensing images, including: remote sensing image data and environmental influencing factor data; the remote sensing image data includes MODIS and Landsat images, and the environmental influencing factor data includes precipitation and solar radiation meteorological data; S2. Preprocessing the acquired remote sensing image data and environmental influencing factor data; the preprocessing includes: denoising, resampling, and correction processing to ensure the temporal and spatial consistency of the data; S3, evapotranspiration inversion using remote sensing image data based on remote sensing evapotranspiration model; The remote sensing evapotranspiration SEBAL (Surface Energy Balance Algorithm for Land) model is a method for calculating surface evapotranspiration by energy balance method. The surface energy balance formula is as follows:

[0026] in, ET is latent heat flux (w / m 2 ); is the latent heat evaporation coefficient, taken as 2.49×10 6 W / (m 2 .mm); is the net radiation flux of the surface (w / m 2 ); is the soil heat flux (w / m 2 ); is the sensible heat flux (w / m 2 ).

[0027] Where: R n The calculation formula of the net radiation flux on the surface is:

[0028] in, Solar shortwave radiation; is the atmospheric long-wave radiation; It is the upward longwave radiation from the surface.

[0029] In the SEBAL model, soil heat flux Mainly related to the normalized vegetation index, surface temperature, surface albedo and net radiation:

[0030] In the formula, is the surface temperature; is the surface albedo; is the Normalized Difference Vegetation Index.

[0031] The calculation formula of sensible heat flux is as follows:

[0032] In the formula, is the air density, which is taken as 1.29 kg / m 3 ; is the specific heat capacity of air at constant pressure, which is 1004 J / (kg·K); is the evaporation surface temperature; is the air temperature at the reference altitude; is the aerodynamic impedance.

[0033] Latent heat flux is the water vapor and heat flux lost due to evapotranspiration. Therefore, estimating daily evaporation should start with latent heat flux calculation. According to the surface energy balance equation, the latent heat flux calculation formula is as follows:

[0034] Instantaneous evaporation The calculation formula is as follows:

[0035] The SEBAL model assumes that the evaporation ratio remains constant throughout the day, so the evaporation ratio can be calculated as follows:

[0036] In the formula, is the instantaneous evaporation ratio; is the 24-hour evaporation ratio; is the 24-hour net radiation, and the calculation formula is as follows; is the 24-hour soil heat flux.

[0037]

[0038]

[0039]

[0040] In the formula, is the distance between the Earth and the Sun; is the geographical latitude; is the solar declination.

[0041] The 24-hour evapotranspiration is calculated as follows:

[0042] S4, using the local weighted scatter point smoothing algorithm to smooth the evapotranspiration time series data, remove noise and reveal long-term trends; The method of smoothing the time series data of evapotranspiration using a local weighted scatter point smoothing algorithm comprises the following steps: First, select a local neighborhood, for each prediction point , select the K data points closest to it from the data set as the local neighborhood, where K is usually an odd number.

[0043] Secondly, calculate the weights and calculate these neighborhood data With prediction point The distance between And convert the distances to weights:

[0044] Where: It is the bandwidth parameter that controls the width of the weight distribution. The weight is determined by methods such as Gaussian kernel function.

[0045] Finally, weighted fitting is performed, according to these weights , for data points in the neighborhood Perform a weighted fit, usually using the least squares method:

[0046] Where: It is the prediction point The estimated value of .

[0047] S5. Calculate the Hurst index of the evapotranspiration time series to assess the persistence of vegetation evapotranspiration changes and determine future trends; The Hurst exponent is calculated using the following formula:

[0048] Where: is the Hurst exponent; Indicates the length of the time series or the number of data points; Indicates that when the time series length is The rescaled range at , that is, the range of the cumulative sum of the deviation sequence; Indicates that when the time series length is The standard deviation at time , used to standardize the rescaled range .

[0049] In the above formula, when When , it indicates that the kNDVI time series shows persistence, that is, the future trend of change is likely to continue the past trend; when When , the kNDVI time series shows randomness, indicating that the future trend has nothing to do with the past; When When , it indicates that the kNDVI time series has anti-persistence, that is, the future trend of change may be opposite to the past; The closer it is to 1, the stronger the persistence of the time series. Conversely, the closer it is to 0, the more significant the anti-persistence is.

[0050] It should be noted that kNDVI (Kernel Normalized Difference Vegetation Index) stands for kernel normalized difference vegetation index, which is a vegetation index calculated based on remote sensing image data and used to study vegetation coverage and growth status. It is improved on the basis of NDVI, and uses the kernel density estimation method to smooth NDVI to reduce noise interference and enhance spatial continuity, thereby improving the accuracy of vegetation index. S6. Conduct evapotranspiration analysis based on evapotranspiration data and influencing factors, calculate the contribution rate of each factor to the change of vegetation evapotranspiration, and identify the main driving factors of vegetation evapotranspiration change.

[0051] The evapotranspiration analysis based on the evapotranspiration data and the influencing factors, calculating the contribution rate of each factor to the change of vegetation evapotranspiration, and identifying the main driving factors of the change of vegetation evapotranspiration include the following steps: First, each set of data needs to be normalized so that different units of different variables can be calculated together:

[0052] Among them, X m It is the normalized data of environmental factors such as NDVI, SSR, Ta, PRE and SM.

[0053] Then, evapotranspiration and environmental factors are input into the ridge regression model:

[0054] Among them, Y m is the standardized ET; is the standardized environmental factor; is the regression coefficient.

[0055] Finally, the relative and absolute contributions of each factor to evapotranspiration were calculated using the ridge regression coefficients and the standardized trends of environmental factors:

[0056] in, is the relative contribution of the first environmental factor, is the coefficient, The detrended value of the first environmental factor, The relative contribution ratio of the first environmental factor, is the relative contribution of the second environmental factor, is the relative contribution of the third environmental factor, is the relative contribution of the nth environmental factor, is the absolute contribution, The non-trend component of vegetation evapotranspiration, Trend component of vegetation evapotranspiration.

[0057] The embodiment of the present invention takes the Hetao Irrigation District as a case area and the period from 2000 to 2022 as a research period, and provides a method for dynamic monitoring of vegetation changes and identification of driving forces based on kNDVI, which specifically includes the following steps: S1. Data collection: Download Landsat and MODIS data to invert evapotranspiration. The following five environmental factors are mainly considered: Leaf Area Index (LAI): a parameter that quantifies the canopy structure of vegetation, indicating the total area of ​​green leaves per unit surface area; the MODIS MOD15A2H v6.1 dataset is used, which is an 8-day synthetic composite data with a resolution of 500 meters. Precipitation (PRE): describes the process of atmospheric moisture falling to the surface; the CHIRPS dataset is used to integrate satellite and ground observations to provide global rainfall data with a resolution of 0.05° since 1981. Soil Moisture (SM): indicates the soil moisture content, which affects the soil and surface water cycle; the GLEAM dataset is used, which is based on microwave observations and provides soil moisture information on the surface and root zone. Surface Net Radiation (SSR): refers to the difference between the solar shortwave radiation received by the surface and the longwave radiation emitted; the data comes from the China Solar Net Radiation Dataset from 1982 to 2020 provided by the National Ecological Data Center, with a resolution of 0.05°, and is generated based on the interpolation of the ERA5 dataset. 2-meter air temperature (Ta): refers to the temperature at an altitude of 2 meters above the ground. It is a key variable in climate and ecological models. The data comes from the ERA5 reanalysis dataset, which combines the model with field observations to provide high-precision temperature data.

[0058] S2. Data preprocessing: The GEE platform is applicable and superior in the remote sensing data processing process. When processing satellite data, the GEE platform is first used to perform screening, cloud removal, cropping and interpolation steps, and then calculations are performed.

[0059] S3. Evapotranspiration inversion: The SEBAL model is a method for calculating surface evapotranspiration through the energy balance method and is a remote sensing evapotranspiration model. The surface energy balance formula is as follows:

[0060] in, ET is latent heat flux (w / m 2 ); is the latent heat evaporation coefficient, taken as 2.49×10 6 W / (m 2 .mm); is the net radiation flux of the surface (w / m 2 ); is the soil heat flux (w / m 2 ); is the sensible heat flux (w / m 2 ).

[0061] Where: Rn The calculation formula of the net radiation flux on the surface is:

[0062] in, Solar shortwave radiation; is the atmospheric long-wave radiation; It is the upward longwave radiation from the surface.

[0063] In the SEBAL model, soil heat flux Mainly related to the normalized vegetation index, surface temperature, surface albedo and net radiation:

[0064] In the formula, is the surface temperature; is the surface albedo; is the Normalized Difference Vegetation Index.

[0065] The calculation formula of sensible heat flux is as follows:

[0066] In the formula, is the air density, which is taken as 1.29 kg / m3 in this study; is the specific heat capacity of air at constant pressure, which is 1004 J / (kg·K); is the evaporation surface temperature; is the air temperature at the reference altitude; is the aerodynamic impedance.

[0067] Latent heat flux is the water vapor and heat flux lost due to evapotranspiration. Therefore, estimating daily evaporation should start with latent heat flux calculation. According to the surface energy balance equation, the latent heat flux calculation formula is as follows:

[0068] Instantaneous evaporation The calculation formula is as follows:

[0069] The SEBAL model assumes that the evaporation ratio remains constant throughout the day, so the evaporation ratio can be calculated as follows:

[0070] In the formula, is the instantaneous evaporation ratio; is the 24-hour evaporation ratio; is the 24-hour net radiation, and the calculation formula is as follows; is the 24-hour soil heat flux.

[0071]

[0072]

[0073]

[0074] In the formula, is the distance between the Earth and the Sun; is the geographical latitude; is the solar declination.

[0075] The 24-hour evapotranspiration is calculated as follows:

[0076] S4, Local Weighted Scatter Smoothing (LOWESS) analysis: First, select a local neighborhood, for each prediction point , select the K data points closest to it from the data set as the local neighborhood, where K is usually an odd number. Then calculate the weights and calculate these neighborhood data With prediction point The distance between And convert the distances to weights:

[0077] Where: It is the bandwidth parameter that controls the width of the weight distribution. The weight is determined by methods such as Gaussian kernel function.

[0078] Finally, weighted fitting is performed according to these weights , for data points in the neighborhood Perform a weighted fit, usually using the least squares method:

[0079] Where: It is the prediction point The estimated value of .

[0080] S5. Hurst index analysis: In this study, a method based on re-standardized range analysis was used to calculate the Hurst index. This method has significant advantages in terms of the reliability of the results. The calculation formula is as follows:

[0081] Where: is the Hurst exponent; Indicates the length of the time series or the number of data points; Indicates that when the time series length is The rescaled range at , that is, the range of the cumulative sum of the deviation sequence; Indicates that when the time series length is The standard deviation at time , used to standardize the rescaled range .

[0082] S6. Ridge regression analysis: First, each set of data needs to be normalized so that different units of different variables can be calculated together:

[0083] Among them, Xm is the normalized data of environmental factors such as NDVI, SSR, Ta, PRE and SM. Then, ET and environmental factors are input into the ridge regression model:

[0084] where Y m is the standardized ET; is the standardized environmental factor; is the regression coefficient. Finally, the relative and absolute contributions of each factor to evapotranspiration are calculated using the ridge regression coefficient and the standardized trend of the environmental factor:

[0085]

[0086]

[0087] in, is the relative contribution of the first environmental factor, is the coefficient, The detrended value of the first environmental factor, The relative contribution ratio of the first environmental factor, is the relative contribution of the second environmental factor, is the relative contribution of the third environmental factor, is the relative contribution of the nth environmental factor, is the absolute contribution, The non-trend component of vegetation evapotranspiration, Trend component of vegetation evapotranspiration.

[0088] In this embodiment, the present invention is based on the initial design as follows Figure 1 The process shown in Figure 3-5 The results shown. From the above results, we can know that: ① Based on the SEBAL model, evapotranspiration inversion is performed on the Loess Plateau (evapotranspiration inversion is as follows Figure 2 As shown in Figure 2, the annual evapotranspiration in the agricultural area of ​​the Loess Plateau showed a trend of gradually increasing from northwest to southeast, and the change trend showed obvious spatial heterogeneity, with 91.46% of the areas showing a significant upward trend (p<0.05) (e.g. Figure 3 ). Through the Hurst index (such as Figure 4 )’s sustainability assessment shows that this trend has strong sustainability, so the future trend is likely to continue the previous upward trend.

[0089] ② Through ridge regression analysis, it was found that LAI is the dominant factor in the annual evapotranspiration change (contribution is as follows Figure 5 As shown). Evapotranspiration in the agricultural area of ​​the Loess Plateau is mainly affected by LAI, and the relative contribution of LAI to vegetation evapotranspiration reaches 55.92%, which is particularly prominent in the northwest. The relative contribution of SM is 16.76%. This shows that the role of soil moisture in maintaining vegetation growth, promoting transpiration and stabilizing vegetation evapotranspiration cannot be ignored. Soil moisture provides plants with the necessary water supply and is the basis for the balance of vegetation growth and vegetation evapotranspiration. The contribution of SSR to evapotranspiration is 13.07%. The relative contribution of SSR is particularly prominent in the southern region, and the south is mainly negative, which is related to the continuous decrease of SSR in recent years. The relative contribution of Ta is only 5.15%. The contribution of PRE to vegetation evapotranspiration is 9.10%.

[0090] In summary, the embodiment of the present invention provides a method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces, which collects multi-source information including MODIS, Landsat images, and environmental data such as precipitation and solar radiation; and performs refined preprocessing on the collected data to ensure the quality and uniformity of the data; accurately calculates the evapotranspiration of vegetation based on the evapotranspiration inversion module; uses local weighted scatter point smoothing technology to smooth the evapotranspiration time series data to reveal its long-term change trend; evaluates the persistence of vegetation evapotranspiration changes and its future trend through Hurst index analysis; uses ridge regression analysis method to quantify the contribution of environmental factors to evapotranspiration changes and identify key driving factors; and intuitively displays the analysis results in the form of charts and reports, which is convenient for users to make visual interpretations and scientific decisions. This method can accurately capture the dynamic change characteristics of vegetation evapotranspiration and effectively distinguish the influence of different environmental factors, providing important technical support for regional ecosystem monitoring, environmental governance and scientific decision-making.

[0091] According to a second aspect of the present invention, a system for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces is provided. Figure 6 Schematic diagram of a system for dynamic monitoring of vegetation evapotranspiration and identification of driving force according to an embodiment of the present application. Figure 6 As shown, the system includes: Data acquisition module, used to obtain multi-time series remote sensing images, including: remote sensing image data and environmental influencing factor data, including MODIS, Landsat images, and precipitation and solar radiation meteorological data; Data processing module, used for preprocessing the collected data, including denoising, resampling, projection correction and normalization; Evapotranspiration inversion module, used to calculate vegetation evapotranspiration based on remote sensing evapotranspiration model; Trend analysis module, which is used to smooth the evapotranspiration time series using the local weighted scatter point smoothing method to reveal the long-term trend; The persistence assessment module is used to analyze the evapotranspiration time series through the Hurst index and assess the persistence of its changes and future trends; The driving force identification module is used to analyze the relative and absolute contributions of environmental factors to evapotranspiration changes through ridge regression and identify the main driving factors; The display and output module is used to output the analysis results in the form of graphics and reports to facilitate visual interpretation and scientific decision-making.

[0092] It can be understood that the system for dynamically monitoring changes in vegetation evapotranspiration and identifying driving force provided by the present invention corresponds to the method for dynamically monitoring changes in vegetation evapotranspiration and identifying driving force provided by the aforementioned embodiments. The relevant technical features of the system for dynamically monitoring changes in vegetation evapotranspiration and identifying driving force can refer to the relevant technical features of the method for dynamically monitoring changes in vegetation evapotranspiration and identifying driving force, which will not be repeated here.

[0093] It should be noted that the core advantage of the vegetation evapotranspiration dynamic monitoring and driving force identification system proposed in accordance with the embodiment of the present invention lies in the integration and coordination of system modules. First, the device is equipped with a remote sensing image and environmental factor data acquisition module, which is responsible for collecting multi-source information including MODIS, Landsat images, and environmental data such as precipitation and solar radiation; secondly, the data processing module performs refined preprocessing on the collected data, including denoising, resampling, projection correction and normalization to ensure the quality and uniformity of the data; then, the evapotranspiration inversion module accurately calculates the evapotranspiration of vegetation based on the SEBAL model; in addition, the trend analysis module uses the local weighted scatter point smoothing (LOWESS) technology to smooth the evapotranspiration time series data to reveal its long-term change trend; the persistence assessment module evaluates the persistence of evapotranspiration changes and its future trends through Hurst index analysis; the driving force identification module uses ridge regression analysis to quantify the contribution of environmental factors to evapotranspiration changes and identify key driving factors; finally, the display and output module intuitively displays the analysis results in the form of charts and reports, which is convenient for users to make visual interpretations and scientific decisions. Through the efficient coordination of this series of modules, the accuracy and practicality of vegetation evapotranspiration monitoring can be greatly improved, providing strong technical support for water resources management and sustainable agricultural development.

[0094] According to a third aspect of the present invention, there is provided an electronic device, Figure 7 The electronic device structure involved in the embodiment of the present invention is shown in detail. Its design highly integrates multiple key components to ensure the accurate implementation of the dynamic change monitoring of vegetation evapotranspiration and its driving force identification technology. The following is a detailed description of the various components of the device and their functional analysis.

[0095] The electronic device includes a core component memory module 201, a processor module 203, a communication interface 202, and a display module 204. The memory module 201 is used to store computer programs, and the processor module 203 is responsible for loading and running the program, thereby realizing the method flow of dynamic change monitoring of vegetation evapotranspiration and identification of driving factors described in the above embodiments. The communication interface 202 is used to realize data transmission and instruction interaction between modules. The display module 204 provides a visual output function for real-time display of data processing results or system status information, further improving the interactive performance and user experience of the device.

[0096] It should be noted that the memory module 201 can support multiple storage types, including high-speed random access memory RAM and non-volatile storage media such as solid-state hard disks or disk storage. This diversified storage configuration ensures that the device has efficient data reading and persistent storage capabilities, thereby meeting the needs of complex computing tasks.

[0097] In terms of specific implementation, if the memory module 201, the processor module 203, the communication interface 202 and the display module 204 exist as independent hardware components, they are usually interconnected and coordinated through a system bus. The bus can adopt a variety of industry standards, such as the Industrial Standard Architecture (ISA) bus, the Peripheral Component Interconnect (PCI) bus or the Extended Industrial Standard Architecture (EISA) bus, and is further divided into an address bus, a data bus and a control bus to achieve efficient transmission. Although Figure 7 A single thick line represents a bus, but the actual implementation may involve multiple buses of different types, which are flexibly configured according to the functional requirements of the device.

[0098] In another implementation, if the memory 201, the processor 202, the communication interface 203 and the display module 204 are integrated into a single chip, the communication and resource sharing between the modules are completed through the on-chip interconnect. This integrated design not only significantly improves the integration of the device, but also reduces signal delay and power consumption.

[0099] The processor module 202 may be a computing core in various forms, such as a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or a system-on-chip (SoC) suitable for a specific application scenario. The display module 204 may use a liquid crystal display (LCD), an organic light-emitting diode display (OLED) or other display technologies to ensure that the visualization of data meets actual needs.

[0100] Through the highly coordinated design of the above modules, the electronic device achieves a deep integration of hardware and software, providing a solid technical guarantee for accurate and efficient dynamic monitoring of vegetation evapotranspiration and driving force analysis.

[0101] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces, characterized in that: include: Acquire multi-time series remote sensing images, including: remote sensing image data and environmental impact factor data; Preprocess the acquired remote sensing image data and environmental impact factor data; Invert evapotranspiration from remote sensing image data based on remote sensing evapotranspiration model; The local weighted scatter point smoothing algorithm is used to smooth the time series data of evapotranspiration to remove noise and reveal long-term trends; Calculate the Hurst index of evapotranspiration time series to assess the persistence of vegetation evapotranspiration changes and determine future trends; Evapotranspiration analysis was conducted based on evapotranspiration data and influencing factors, the contribution rate of each factor to the change of vegetation evapotranspiration was calculated, and the main driving factors of the change of vegetation evapotranspiration were identified.

2. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The remote sensing image data include MODIS and Landsat images, and the environmental influencing factor data include precipitation and solar radiation meteorological data.

3. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The preprocessing includes denoising, resampling, projection correction and normalization.

4. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The remote sensing evapotranspiration model calculates surface evapotranspiration through energy balance, wherein the surface evapotranspiration is calculated as follows: In the formula, ET 24 is the 24-hour latent heat flux, w / m 2 , is the 24-hour evaporation ratio; is the 24-hour net radiation, is the latent heat of evaporation coefficient.

5. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The method of smoothing the time series data of evapotranspiration by using a local weighted scatter point smoothing algorithm includes: Select a local neighborhood, for each prediction point , select the K data points closest to it from the data set as the local neighborhood, where K is an odd number; Calculate neighborhood data With prediction point The distance between And convert the distance into weight ; Perform weighted fitting according to the weights , for data points in the neighborhood Weighted fitting was performed using the least squares method.

6. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The Hurst exponent is calculated using the following formula: Where: Indicates the length of the time series or the number of data points; Indicates that when the time series length is The rescaled range at , that is, the range of the cumulative sum of the deviation sequence; Indicates that when the time series length is The standard deviation at time , used to standardize the rescaled range ; is the Hurst exponent, when When , it indicates that the KNDVI time series shows persistence, that is, the future change trend is likely to continue the past trend; when When , the kernel normalized difference vegetation index time series shows randomness, indicating that the future trend has nothing to do with the past; when When , it indicates that the KNDVI time series has anti-persistence, that is, the future trend may be opposite to the past; when The closer it is to 1, the stronger the persistence of the time series. Conversely, the closer it is to 0, the more significant the anti-persistence is.

7. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 1, characterized in that: The evapotranspiration analysis based on evapotranspiration data and influencing factors, and calculation of the contribution rate of each factor to the change of vegetation evapotranspiration include: Normalize each set of data and calculate different units of different variables together; Vegetation evapotranspiration and environmental factors were input into the ridge regression model; The relative and absolute contributions of each factor to evapotranspiration were calculated using the ridge regression coefficients and the standardized trends of environmental factors.

8. The method for dynamic monitoring of vegetation evapotranspiration changes and identification of driving forces according to claim 7, characterized in that: The following formula is used to calculate the relative and absolute contribution of each factor to evapotranspiration: in, is the relative contribution of the first environmental factor, is the coefficient, The detrended value of the first environmental factor, The relative contribution ratio of the first environmental factor, is the relative contribution of the second environmental factor, is the relative contribution of the third environmental factor, is the relative contribution of the nth environmental factor, is the absolute contribution, The non-trend component of vegetation evapotranspiration, Trend component of vegetation evapotranspiration.

9. A system for dynamic monitoring of vegetation evapotranspiration and identification of driving force, characterized in that: include: Data acquisition module, used to obtain multi-time series remote sensing images, including: remote sensing image data and environmental impact factor data; Data processing module, used for preprocessing the collected data, including denoising, resampling, projection correction and normalization; Evapotranspiration inversion module, used to calculate vegetation evapotranspiration based on remote sensing evapotranspiration model; Trend analysis module, which is used to smooth the evapotranspiration time series using the local weighted scatter point smoothing method to reveal the long-term trend; The persistence assessment module is used to analyze the evapotranspiration time series through the Hurst index and assess the persistence of its changes and future trends; The driving force identification module is used to analyze the relative and absolute contributions of environmental factors to evapotranspiration changes through ridge regression and identify the main driving factors; The display and output module is used to output the analysis results in the form of graphics and reports to facilitate visual interpretation and scientific decision-making.

10. An electronic device, characterized in that: include: A memory, a processor, a display, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for dynamically monitoring changes in vegetation evapotranspiration and identifying driving forces as described in any one of claims 1 to 8.

Images