Quantitative Analysis Methods for Vegetation Physiological and Structural Responses under Extreme Dry and Hot Combined Stress

Abstract

This application provides a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry and hot combined stress, relating to the field of vegetation remote sensing. The method includes: collecting hydrological and meteorological data, vegetation structure data, vegetation function data, and soil moisture data; calculating the soil drought index of soil moisture data to determine the drought peak; extracting outlier forms using a local weighted regression method; calculating the average change trajectory of hydrological and meteorological data; establishing a random forest model; training the random forest model based on a time window centered on the drought peak to obtain prediction results; obtaining the proportion of physiological signals in vegetation function data based on the prediction results; extracting driving variables based on the drought peak and outlier forms and performing attribution analysis to obtain the relative importance ranking of the driving variables to the three types of vegetation function variables. This application can accurately separate the physiological and structural response signals of vegetation, improving the monitoring accuracy and analytical capabilities of vegetation dynamic changes.

Description

Technical Field

[0001] This application relates to the field of vegetation remote sensing, and in particular to a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry and hot combined stress. Background Technology

[0002] In recent years, global warming has exacerbated the concurrent occurrence of droughts and heat waves, forming more destructive combined dry-heat extreme events that pose a serious threat to water resources, food security, and ecosystems. These events are caused by the combined effects of atmospheric heat and drought and soil drought, producing a synergistic effect. Combined dry-heat stress affects the physiological functions and structural characteristics of vegetation, leading to reduced photosynthetic and transpiration efficiency, ultimately inhibiting vegetation productivity and potentially even causing vegetation death. Although structural changes in vegetation are significant and have been studied extensively, physiological responses are more subtle and rapid, and have not yet been fully investigated. In-depth exploration of vegetation physiological responses will help understand their adaptation mechanisms in extreme climate events and provide scientific evidence for clarifying the main driving factors of vegetation function decline.

[0003] Under combined drought and heat stress, vegetation functional responses primarily involve three mechanisms: photosynthesis, transpiration, and water regulation. First, drought and high temperatures lead to stomatal closure and reduced carbon dioxide uptake, thus affecting photosynthetic efficiency. Furthermore, high temperatures may damage the photosystem, weakening light energy utilization efficiency. Second, vegetation reduces water loss by decreasing transpiration, which is closely related to stomatal closure. Finally, water regulation mechanisms help vegetation regulate water behavior during drought periods, maintaining water potential balance in leaves and roots and reducing the risk of xylem cavitation. These three responses reflect the complex coping mechanisms of vegetation systems, including stomatal regulation, root deepening, and leaf adjustment, demonstrating vegetation's ability to cope with extreme drought and heat disasters in multiple dimensions. Focusing only on a single physiological mechanism makes it difficult to comprehensively capture the drought-affected state of vegetation and subsequent changes in productivity. Therefore, a comprehensive assessment of vegetation's photosynthesis, transpiration, and water regulation is crucial for a more accurate understanding of the impact of combined drought and heat events on vegetation.

[0004] Satellite remote sensing technology is an important tool for studying the physiological and structural responses of vegetation under combined drought and heat stress. The application of data such as sunlight-induced chlorophyll fluorescence, near-infrared reflectance index, evapotranspiration, vegetation optical thickness, and leaf area index has significantly promoted the development of related research. However, current research faces the significant challenge of the confluence of physiological and structural signals. Satellite-acquired fluorescence and evapotranspiration data simultaneously reflect both physiological and structural changes in vegetation, making it difficult to separate these two signals for independent analysis. Therefore, to gain a deeper understanding of the impact of combined drought and heat events on vegetation, it is essential to improve the accuracy of separating physiological and structural signals. Summary of the Invention

[0005] The purpose of this invention is to provide a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry and hot combined stress, in order to solve the technical problems of the difficulty in separating physiological response signals and structural response signals in large-scale vegetation monitoring using traditional remote sensing technology, as well as the limited ability of models to interpret changes in vegetation physiological functions.

[0006] The above-mentioned objective of this application is achieved through the following technical solution:

[0007] S1: Collect hydrological and meteorological data, vegetation structure data, and vegetation function data of the study area and perform preprocessing;

[0008] S2: Obtain soil moisture data for the study area; determine the drought peak by calculating the soil drought index from the soil moisture data;

[0009] S3: Extract outlier forms from hydrological and meteorological data using a local weighted regression method;

[0010] S4: Set a time window centered on the drought peak and calculate the average change trajectory of each variable in the hydrological and meteorological data during the drought period;

[0011] S5: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a random forest model is established; based on the time window, the random forest model is trained, and the prediction results for extremely arid areas are obtained.

[0012] S6: Based on the prediction results, obtain the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data;

[0013] The three categories of vegetation function variables include: photosynthetic function variables, transpiration function variables, and water regulation function variables;

[0014] S7: Based on the drought peak and outlier forms, extract the driving variables for the drought development and recovery periods; the driving variables include: hydrological and meteorological data, vegetation characteristics, and drought duration;

[0015] S8: Using three types of vegetation functional variables as the object, and combining the proportion of physiological signals, attribution analysis was performed on the driving variables to obtain the relative importance ranking of the driving variables to the three types of vegetation functional variables, and to complete the quantitative analysis of vegetation physiological and structural responses under extreme dry and hot combined stress.

[0016] Optionally, step S1 includes:

[0017] The hydrological and meteorological data include: air temperature, rainfall, soil moisture, and solar radiation data;

[0018] The vegetation structure data includes: leaf area index and near-infrared vegetation reflectance;

[0019] The vegetation functional data include: solar-induced chlorophyll fluorescence, vegetation optical thickness, and evapotranspiration.

[0020] Solar-induced chlorophyll fluorescence data included: instantaneous data under clear skies and daily average data;

[0021] Vegetation optical thickness includes daytime vegetation optical depth data and nighttime vegetation optical depth data.

[0022] Optionally, the specific steps of the preprocessing include:

[0023] Quality screening and coordinate transformation were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data.

[0024] Mean synthesis and resampling were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data to unify temporal and spatial resolution.

[0025] Optionally, step S2 includes:

[0026] S21: Using historical soil moisture data for the study area for the predetermined years, calculate the soil drought index as follows:

[0027]

[0028] in Indicates the soil drought index; Indicates the first Month and the Annual soil moisture; and They are Mean and standard deviation over all years;

[0029] Filter out The pixels, as in extremely arid regions;

[0030] S22: The time when the minimum soil moisture in an extremely arid region occurs is taken as the drought peak, and the time when the drought peak occurs is obtained.

[0031] Optionally, step S3 includes:

[0032] S31: Extract the long-term trend of hydrological and meteorological data using the LOWESS-based local weighted regression method;

[0033] S32: Calculate the monthly average value of hydrological and meteorological data to obtain the average seasonal cycle;

[0034] S33: Based on hydrological and meteorological data, remove long-term trends and average seasonal cycles to obtain the outlier form of hydrological and meteorological data.

[0035] Optionally, step S5 includes:

[0036] S51: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a first random forest model is established.

[0037] S52: Using hydrological and meteorological data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a second random forest model is established.

[0038] S53: Train the first random forest model and the second random forest model using hydrological and meteorological data, vegetation structure data and vegetation function data outside the time window;

[0039] S54: Using the trained first and second random forest models, predictions are made on hydrological and meteorological data and vegetation structure data outside the time window to obtain prediction results.

[0040] Optionally, step S6 includes:

[0041] Calculate the residuals between the prediction results of the first random forest model and the prediction results of the second random forest model, and separate the physiological signals of vegetation function data.

[0042] Based on different stages of drought development and humidity levels, and combined with physiological signals, the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data were statistically analyzed.

[0043] Optionally, step S7 includes:

[0044] The time window before the drought peak is defined as the drought development period, and the time window after the drought peak is defined as the drought recovery period.

[0045] Hydrometeorological data include: air temperature, rainfall, solar radiation, and soil moisture;

[0046] Vegetation characteristics include: vegetation cover fraction and moisture content;

[0047] The duration of drought is determined by identifying the time it takes for soil moisture to recover from outliers to normal levels before and after the drought peak.

[0048] Optionally, step S8 includes:

[0049] Using the SHAP method, attribution analysis was performed on the driving variables to quantify the characteristic contributions of hydrological and meteorological data, vegetation characteristics, and drought duration to the three types of vegetation function variables, and to obtain the relative importance ranking of the driving variables to the three types of vegetation function variables.

[0050] An electronic device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to enable the electronic device to perform a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry and hot combined stress.

[0051] A computer-readable storage medium storing instructions that, when executed, perform a method for quantitative analysis of the physiological and structural responses of vegetation under extreme dry-heat combined stress.

[0052] The beneficial effects of the technical solution provided in this application are:

[0053] This study utilizes locally weighted regression to extract anomalous changes in hydro-meteorological, vegetation structure, and vegetation function data. By setting time windows, the trajectories of each index during drought periods are analyzed. Random forest modeling is employed for the hydro-meteorological driving variables, vegetation structure variables, and vegetation function variables. Based on the model residual concept, physiological and structural signals of the three functional responses of vegetation under extreme heat and dryness events are separated. Model-independent machine learning interpretation methods are used to quantify the characteristic contributions of each driving variable to vegetation photosynthetic, transpiration, and water regulation functions. This clarifies the physiological and structural signals of the three functional responses of vegetation under heat and dryness stress, overcoming the difficulty of mixed physiological response signals in large-scale vegetation remote sensing monitoring. It provides methodological support for improving the accurate monitoring and analysis capabilities of vegetation dynamic changes and is of great significance for accurately assessing changes in vegetation productivity and ecosystem carbon sequestration functions. Attached Figure Description

[0054] The present application will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0055] Figure 1 This is a step diagram of an embodiment of this application;

[0056] Figure 2 This is a map showing the locations of drought events in the study area as described in this application embodiment;

[0057] Figure 3 This is a time-series change diagram of various hydrological and meteorological data and vegetation index data before and after extreme drought in the embodiments of this application;

[0058] Figure 4 This is a diagram showing the separation results of physiological components of vegetation's functional response during drought in an embodiment of this application;

[0059] Figure 5 This is a zonal statistical result diagram of the overall functional response and physiological component of vegetation during drought in the embodiments of this application;

[0060] Figure 6 This is a graph showing the attribution analysis results of the physiological response driving factors of vegetation during drought in the embodiments of this application;

[0061] Figure 7 This is a schematic diagram of the electronic device structure in the embodiments of this application. Detailed Implementation

[0062] To provide a clearer understanding of the technical features, objectives, and effects of this application, the specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0063] The embodiments of this application provide a method for quantitative analysis of the physiological and structural responses of vegetation under extreme dry and hot combined stress.

[0064] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry-heat combined stress, as described in an embodiment of this application.

[0065] S1: Collect hydrological and meteorological data, vegetation structure data, and vegetation function data of the study area and perform preprocessing;

[0066] S2: Obtain soil moisture data for the study area; determine the drought peak by calculating the soil drought index from the soil moisture data;

[0067] S3: Extract outlier forms from hydrological and meteorological data using a local weighted regression method;

[0068] S4: Set a time window centered on the drought peak and calculate the average change trajectory of each variable in the hydrological and meteorological data during the drought period;

[0069] Specifically, the time window is set to be three months before and after the drought peak, which is the time when the minimum soil moisture value of the pixel occurs over four days. By extracting extreme drought events pixel by pixel and statistically analyzing the average change trends of various hydrological and meteorological data and vegetation index data during the drought, the average change trajectory of each variable during the drought is finally obtained; the vegetation index data includes vegetation structure data and vegetation function data, which are preprocessed.

[0070] S5: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a random forest model is established; based on the time window, the random forest model is trained, and the prediction results for extremely arid areas are obtained.

[0071] S6: Based on the prediction results, obtain the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data;

[0072] The three categories of vegetation function variables include: photosynthetic function variables, transpiration function variables, and water regulation function variables;

[0073] S7: Based on the drought peak and outlier forms, extract the driving variables for the drought development and recovery periods; the driving variables include: hydrological and meteorological data, vegetation characteristics, and drought duration;

[0074] S8: Using three types of vegetation functional variables as the object, and combining the proportion of physiological signals, attribution analysis was performed on the driving variables to obtain the relative importance ranking of the driving variables to the three types of vegetation functional variables, and to complete the quantitative analysis of vegetation physiological and structural responses under extreme dry and hot combined stress.

[0075] Specifically, Figure 1 This is a flowchart illustrating the implementation of a quantitative analysis method for the physiological and structural responses of vegetation under extreme dry and hot combined stress. Figure 2 This is a map showing the locations and timing of drought events in the study area of ​​the Amazon region from 2016 to 2020. Figure 3 It is a time-series change graph of hydrological and meteorological data such as soil moisture, vegetation index data such as LAI, and vegetation function index such as SIF before and after extreme drought. Figure 4 It is the result of the separation of physiological components in response to drought by three types of vegetation functional variables: photosynthesis, transpiration and water regulation. Figure 5 This is a zonal statistical diagram showing the overall functional response and physiological components of vegetation during drought; Figure 6 This is the result of attribution analysis of the physiological response drivers of three types of functional variables during the drought development and recovery periods.

[0076] Step S1 includes:

[0077] The hydrological and meteorological data include: air temperature, rainfall, soil moisture, and solar radiation data;

[0078] Specifically, the hydrological and meteorological data were derived from ERA5-Land reanalysis data, with a daily temporal resolution and a spatial resolution of 0.1°. The ERA5-Land reanalysis data was calculated and exported using the Google Earth Engine (GEE) platform. Air temperature was calculated using 2m surface air temperature data, and soil moisture was calculated using a weighted average of soil moisture content data at 7cm, 21cm, and 72cm depths. The meteorological data used were astronomical synthetic reanalysis data, exported after 4 days of synthesis and cropping within the GEE platform.

[0079] The vegetation structure data includes: leaf area index and near-infrared vegetation reflectance;

[0080] Specifically, the vegetation structure data includes leaf area index (LAI) and near-infrared vegetation reflectance (NIRv). The vegetation structure data is derived from the MODIS MCD15A3H and MCD43A4 datasets, with a temporal resolution of 4 days and a spatial resolution of 500 meters. The MODIS MCD15A3H dataset is a 4-day temporal scale, while MCD43A4 is a daily-scale dataset. Both datasets were calculated, synthesized, and exported using the GEE platform.

[0081] The vegetation functional data include: solar-induced chlorophyll fluorescence, vegetation optical thickness, and evapotranspiration.

[0082] Specifically, the vegetation function data includes solar-induced chlorophyll fluorescence (SIF), vegetation optical thickness (VOD), and evapotranspiration (ET). The vegetation function data are derived from the global spatial continuous solar-induced fluorescence (CSIF) dataset and the LPDR v3 vegetation optical thickness dataset, with temporal resolutions of 4 days and daily, and spatial resolutions of 0.05° and 25 km, respectively.

[0083] Solar-induced chlorophyll fluorescence data included: instantaneous data under clear skies and daily average data;

[0084] Vegetation optical thickness includes daytime vegetation optical depth data and nighttime vegetation optical depth data.

[0085] In one embodiment of this application, daily average data is used in the screening calculations for the growth period because it reflects the long-term changing trend of photosynthetic function; instantaneous clear-sky data is used in subsequent physiological component separation calculations; vegetation optical thickness includes daytime and nighttime vegetation optical depth data, the ratio of which is used here to represent the vegetation's water regulation function, hereinafter referred to as VOD ratio. Additionally, vegetation cover score and irrigation score data are included, sourced from the ESACCI 2020 global vegetation cover score dataset and the FAO 2005 global irrigation score dataset.

[0086] The specific steps of the preprocessing include:

[0087] Quality screening and coordinate transformation were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data.

[0088] Mean synthesis and resampling were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data to unify temporal and spatial resolution.

[0089] In one embodiment of this application, the preprocessing process includes quality screening, coordinate transformation, resampling, and synthesis of the above data. All data have a temporal resolution of 4 days and a spatial resolution of 0.1 degrees. Subsequently, uniform cropping and masking are performed to obtain data within the specific study area and study time period. After processing, all data are calculated in the Python environment in NumPy data format.

[0090] Step S2 includes:

[0091] S21: Using historical soil moisture data for the study area for the predetermined years, calculate the soil drought index as follows:

[0092]

[0093] in Indicates the soil drought index; Indicates the first Month and the Annual soil moisture; and They are Mean and standard deviation over all years;

[0094] Filter out The pixels, as in extremely arid regions;

[0095] S22: The time when the minimum soil moisture in an extremely arid region occurs is taken as the drought peak, and the time when the drought peak occurs is obtained.

[0096] Specifically, by using the time of minimum soil moisture occurrence in areas that have experienced extreme drought as a mask, a time step and pixel location map of extreme drought events in the study area is obtained, such as... Figure 2 As shown.

[0097] Step S3 includes:

[0098] S31: Extract the long-term trend of hydrological and meteorological data using the LOWESS-based local weighted regression method;

[0099] S32: Calculate the monthly average value of hydrological and meteorological data to obtain the average seasonal cycle;

[0100] S33: Based on hydrological and meteorological data, remove long-term trends and average seasonal cycles to obtain the outlier form of hydrological and meteorological data.

[0101] Specifically, the long-term trend is extracted from the data using the LOWESS package in a Python environment. This method extracts the long-term seasonal trend of the time series through locally weighted regression, and then subtracts the long-term seasonal trend from the original time series to obtain the outlier form of the variable after removing the seasonal trend. Hydrological and meteorological data from 2016 to 2020 are selected. First, the monthly average of all data is calculated, and the average seasonal cycle is extracted. Then, the average seasonal cycle is subtracted from the original data to further eliminate the influence of seasonal fluctuations, thus obtaining the outlier form of all data.

[0102] Step S5 includes:

[0103] S51: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a first random forest model is established.

[0104] S52: Using hydrological and meteorological data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a second random forest model is established.

[0105] S53: Train the first random forest model and the second random forest model using hydrological and meteorological data, vegetation structure data and vegetation function data outside the time window;

[0106] S54: Using the trained first and second random forest models, predictions are made on hydrological and meteorological data and vegetation structure data outside the time window to obtain prediction results.

[0107] Specifically, data within a time window is selected for model prediction, while data outside the time window is used for model training. Random forest modeling and prediction are performed pixel by pixel, ultimately yielding the modeling results within pixels in extremely arid regions.

[0108] Step S6 includes:

[0109] Calculate the residuals between the prediction results of the first random forest model and the prediction results of the second random forest model, and separate the physiological signals of vegetation function data.

[0110] Based on different stages of drought development and humidity levels, and combined with physiological signals, the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data were statistically analyzed.

[0111] Specifically, both random forest models use vegetation function data as output variables and model with different input variables. In the final random forest prediction results, the model difference is the outlier change caused by vegetation physiological components. The vegetation physiological response is thus separated by the model residual method.

[0112] Step S7 includes:

[0113] The time window before the drought peak is defined as the drought development period, and the time window after the drought peak is defined as the drought recovery period.

[0114] Hydrometeorological data include: air temperature, rainfall, solar radiation, and soil moisture;

[0115] Vegetation characteristics include: vegetation cover fraction and moisture content;

[0116] The duration of drought is determined by identifying the time it takes for soil moisture to recover from outliers to normal levels before and after the drought peak.

[0117] Specifically, within the set time window, the time window before the peak is defined as the development period, and the window after the peak is defined as the recovery period; the hydrological and meteorological data include air temperature, rainfall, radiation, and soil moisture, and the vegetation characteristics include vegetation cover fraction and humidity; while the drought duration is calculated by identifying the time it takes for soil moisture to recover from outliers to normal levels before and after the drought peak.

[0118] Step S8 includes:

[0119] Using the SHAP method, attribution analysis was performed on the driving variables to quantify the characteristic contributions of hydrological and meteorological data, vegetation characteristics, and drought duration to the three types of vegetation function variables, and to obtain the relative importance ranking of the driving variables to the three types of vegetation function variables.

[0120] Specifically, the SHAP (SHapley Additive exPlanations) method is used. Long-term variables are reduced in dimensionality by calculating the average value of that period. All driving variables are then quantified by calculating their spatial correlation with the three types of vegetation function variables to obtain the relative importance ranking of all driving variables to the different function variables.

[0121] The present invention provides the following embodiments:

[0122] Example (1): Collection and processing of basic data;

[0123] In this embodiment, four types of hydrometeorological data were collected: air temperature, rainfall, soil moisture, and solar radiation. The hydrometeorological data were processed using a 4-day composite method on the GEE platform to obtain uniform temporal and spatial resolution. The weighted average of soil moisture data at different depths effectively reflects the overall soil moisture content.

[0124] Example (2): Extracting the extremely arid regions of the study area and the time of drought peak occurrence;

[0125] In this embodiment, the SSMI index is calculated to extract regions globally that have experienced drought and their corresponding drought event times. First, the monthly SSMI index (soil drought index) is calculated using soil moisture data from the past 40 years. This standardizes the time series of soil moisture for each month to reflect the severity of drought. To screen for extremely drought-stricken areas, pixels with SSMI values ​​less than -2 are selected and marked as regions that have experienced extreme drought. Next, the timing of drought events is further determined by examining soil moisture records. Specifically, four days of soil moisture records are used, and the minimum value among these records is identified as the drought peak, with its occurrence time recorded. This allows the determination of the specific time step for each drought event. Finally, by masking the minimum soil moisture values ​​with the selected extremely drought-stricken regions, the occurrence times of extreme drought events within the study area and their corresponding pixel location distributions (e.g., [missing information]) are obtained. Figure 2 (As shown). The generated time-step and location data can be used to further analyze the spatial and temporal characteristics of drought, helping to identify and respond to extreme drought events.

[0126] Example (3): Remove the average seasonal cycle and long-term seasonal trend of each variable in hydrological and meteorological data to extract outliers;

[0127] In this embodiment, hydrological and meteorological data from 2016 to 2020 were selected. First, the monthly average of all data was calculated to extract the average seasonal cycle for each variable. This step involves averaging data from multiple years for the same month to obtain seasonal fluctuation characteristics. The LOWESS method was then used to extract the long-term seasonal trend of each variable. In the Python environment, this was done by calling the LOWESS package, and the extracted long-term seasonal trend was subtracted from the original time series to eliminate its influence on the data. Finally, the average seasonal cycle was subtracted from the original data to further eliminate the impact of seasonal fluctuations, thereby obtaining the outlier forms of the hydrological and meteorological data and vegetation index data.

[0128] Example (4): Set a time window to extract the average change trajectory of each variable;

[0129] In this embodiment, the average change trajectory of each variable during the drought period is calculated by setting a three-month time window centered on the drought peak. Specifically, the drought peak is first determined, which is the point in time when the minimum soil moisture value occurs for each pixel. Then, within the three-month period before and after this point, hydrological and meteorological data and vegetation index data (such as LAI, NIRv, SIF, etc.) are extracted pixel by pixel. By statistically analyzing the average change trend of these data, the average change trajectory of each variable during the drought period is finally obtained, clearly demonstrating the dynamic change characteristics from the drought development stage, the drought peak, to the drought recovery stage.

[0130] During the drought development period, vegetation indices such as LAI, NIRv, and SIF all showed a trend of first rising and then falling, indicating the vegetation's response in the early stages of drought and its decline process as the drought intensifies (e.g., Figure 3 (As shown in the figure). Similar trends were also observed in the trajectories of VOD ratio, ET, and soil moisture. Soil moisture decreased significantly before the drought peak and then gradually recovered during the recovery period, reflecting the direct impact of drought on soil moisture storage. These trajectories reveal the dynamic response mechanism between meteorological conditions and vegetation function during drought.

[0131] Example (5): Establishing a random forest model of vegetation function variables to separate physiological response signals

[0132] In this embodiment, two random forest models were constructed based on hydrometeorological data and vegetation structure variables to predict changes in vegetation function variables such as photosynthetic function (SIF), transpiration function (ET), and water regulation function (VOD ratio). During model construction, pixel-by-pixel random forest modeling and prediction were performed using a combination of hydrometeorological data and vegetation structure variables, as well as individual hydrometeorological data, respectively. Through pixel-by-pixel random forest model analysis, the model accurately captured the changing trends of various vegetation function variables during drought, ultimately obtaining vegetation function modeling results within pixels that had experienced drought, thus revealing the different impacts of drought on vegetation function (e.g., ...). Figure 4 (As shown).

[0133] Example (6): Based on the model residuals, the physiological response signals of vegetation are separated and their distribution patterns are statistically analyzed;

[0134] In this embodiment, two random forest models are used to separate physiological and structural signals of three types of vegetation functional variables, aiming to deeply study the impact of drought events on vegetation function. Two random forest models are established, one using hydro-meteorological data plus vegetation structural variables and the other using only hydro-meteorological data as independent variables. The dependent variables are photosynthetic function (SIF), transpiration function (ET), and water regulation function (VOD ratio). Finally, changes in vegetation physiological components are separated by calculating the model residuals. The model residuals reflect abnormal changes in vegetation function caused by physiological factors, and this method provides a more intuitive path to distinguish between vegetation structure and physiological signals. The time window covers the entire development stage of the drought event, centered on the drought peak. The results of the random forest models show that vegetation functional performance is highly correlated with hydro-meteorological data during drought. By analyzing the residuals, abnormalities caused by physiological responses can be accurately separated. Separate modeling of SIF, ET, and VOD ratio shows that the responses of different functional variables differ significantly during drought.

[0135] Photosynthetic function was predominantly negative throughout the drought period, particularly at the peak of the drought, indicating that the drought event had a strong inhibitory effect on vegetation photosynthetic function (e.g., Figure 5 (As shown). ET mainly showed positive anomalies during drought, indicating that plants may alleviate drought stress by increasing water transpiration. VOD ratio, as a representative of water regulation function, showed abnormalities that varied with different drought levels. In some areas, VOD ratio showed significant fluctuations during drought, reflecting differences in the response of vegetation water status to drought.

[0136] In specific drought groups, the proportions of physiological anomalies in SIF, ET, and VOD ratio within pixels with different drought intensities were calculated. The results showed that physiological anomalies accounted for a significantly higher proportion than structural anomalies. For example, in the physiological response of SIF, the proportions under different drought levels were 0.97, 0.91, 0.97, and 0.92, respectively, indicating that the photosynthetic function of vegetation under different drought intensities was mainly dominated by physiological factors. For ET and VOD ratio, the proportion of physiological anomalies was also concentrated between 0.92 and 0.97, indicating that physiological anomalies played a dominant role in the abnormal changes in overall vegetation function, especially during extreme droughts. This high proportion of physiological anomalies suggests that the impact of drought events on vegetation function is mainly reflected through physiological functions, while structural anomalies account for a smaller proportion. This finding reveals the dynamic response mechanism of vegetation under drought stress: when soil moisture decreases, vegetation responds to environmental pressure through a series of physiological adjustments (such as photosynthesis and transpiration), while structural changes are relatively slow.

[0137] Furthermore, the abnormal patterns of vegetation functional variables exhibited similar trends at different stages of drought development. Particularly around the peak of the drought, the physiological anomalies in SIF, ET, and VOD ratio followed the same trend as the overall anomalies, indicating that plants demonstrated a strong physiological adjustment capacity during drought response. In certain extreme drought phases, the positive anomalies in ET and the fluctuations in VOD ratio were particularly pronounced, further reflecting the importance of vegetation in water regulation. These dynamic characteristics provide important evidence for assessing the comprehensive impact of drought on ecosystems.

[0138] Example (7): Extracting driving variables such as hydrological and meteorological conditions during different drought development periods;

[0139] In this embodiment, a time window method is used to extract hydrological and meteorological data, vegetation characteristics, and drought duration during the drought development and recovery periods, centered on the drought peak. Specifically, according to a set time window, the time window before the drought peak is defined as the drought development period, and the time window after the peak is the drought recovery period. During these two stages, the main hydrological and meteorological data focus includes key meteorological factors such as air temperature, precipitation, radiation, and soil moisture; while vegetation characteristics mainly cover vegetation cover fraction and humidity. The drought duration is calculated based on the changes in soil moisture anomalies. By identifying soil moisture anomalies before and after the drought peak, the time period from extreme values ​​to normal levels is used to define the drought duration.

[0140] Example (8): Attribution analysis of driving variables based on SHAP method;

[0141] In this embodiment, the Shapley Additive exPlanations (SHAP) method was used to conduct attribution analysis on hydrological and meteorological data, vegetation, and drought characteristics for three types of vegetation function variables: photosynthetic function, transpiration function, and water regulation function. This method aims to quantify the contribution of each driving variable to vegetation function characteristics, helping to gain a deeper understanding of the impact of these variables on vegetation function during drought development and recovery periods.

[0142] During the drought development stage, studies have shown that the main driving variables affecting vegetation function are hydrometeorological factors such as temperature and soil moisture (e.g., Figure 6 (As shown in the figure). These variables showed significant correlations during the drought process, especially during peak temperatures, when vegetation function was significantly affected. Meanwhile, geographical features such as vegetation cover also had some influence, but their importance was relatively smaller compared to meteorological driving variables. Although the duration of drought had a limited impact on vegetation function, changes in soil moisture still affected vegetation water supply in some cases. During the drought recovery period, the influence of each driving variable was roughly similar to that during the drought development period, but the importance of the driving variables was generally higher. This indicates that as vegetation gradually recovers, the influence of hydrometeorological data such as temperature and radiation reduction gradually increases.

[0143] Analysis of the driving variables revealed that ambient temperature had the greatest impact on transpiration (ET), while drought duration and ambient humidity also played significant roles. Particularly during the drought recovery phase, increased ambient humidity helped improve vegetation transpiration and enhance water use efficiency. Furthermore, soil moisture and ambient aridity also significantly affected the VOD ratio, reflecting the vegetation's water regulation capacity in response to drought. Although drought duration had a relatively small impact on the VOD ratio, it still demonstrated its importance to vegetation water status under specific conditions. This multi-dimensional impact assessment provides strong support for a deeper understanding of the functional responses of vegetation under drought conditions.

[0144] Analysis based on the SHAP method indicates that during drought development, temperature and soil moisture, as key hydrometeorological data, significantly influenced vegetation photosynthesis and transpiration. During the drought recovery phase, these variables remained significant and their importance surpassed that of the drought development phase. This suggests that meteorological drivers play a crucial role in the recovery of vegetation function. The study also found different response patterns among different vegetation function variables. The anomalous responses of SIF, ET, and VOD ratio during drought events revealed a complex relationship between vegetation's physiological regulatory capacity and structural changes. This dynamic response mechanism further emphasizes the combined impact of soil moisture and meteorological conditions on vegetation function.

[0145] This application also discloses an electronic device. (See reference...) Figure 7 , Figure 7 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. The electronic device 500 may include: at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

[0146] The communication bus 502 is used to enable communication between these components.

[0147] The user interface 503 may include a display screen, and optionally, the user interface 503 may also include a standard wired interface or a wireless interface.

[0148] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0149] This application also discloses a computer-readable storage medium storing multiple instructions adapted for loading by a processor to execute the above-described method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress.

[0150] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will readily conceive of those skilled in the art upon consideration of the specification and the disclosure of practical truths.

[0151] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A quantitative analysis method for the physiological and structural responses of vegetation under extreme dry-heat combined stress, characterized in that, The method includes the following steps: S1: Collect hydrological and meteorological data, vegetation structure data, and vegetation function data of the study area and perform preprocessing; S2: Obtain soil moisture data for the study area; determine the drought peak by calculating the soil drought index from the soil moisture data; S3: Extract outlier forms from hydrological and meteorological data using a local weighted regression method; S4: Set a time window centered on the drought peak and calculate the average change trajectory of each variable in the hydrological and meteorological data during the drought period; S5: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a random forest model is established; based on the time window, the random forest model is trained, and the prediction results for extremely arid areas are obtained. S6: Based on the prediction results, obtain the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data; The three categories of vegetation function variables include: photosynthetic function variables, transpiration function variables, and water regulation function variables; S7: Based on the drought peak and outlier forms, extract the driving variables for the drought development and recovery periods; the driving variables include: hydrological and meteorological data, vegetation characteristics, and drought duration; S8: Using three types of vegetation functional variables as the object, and combining the proportion of physiological signals, attribution analysis was performed on the driving variables to obtain the relative importance ranking of the driving variables to the three types of vegetation functional variables, and to complete the quantitative analysis of vegetation physiological and structural responses under extreme dry and hot combined stress.

2. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S1 includes: The hydrological and meteorological data include: air temperature, rainfall, soil moisture, and solar radiation data; The vegetation structure data includes: leaf area index and near-infrared vegetation reflectance; The vegetation functional data include: solar-induced chlorophyll fluorescence, vegetation optical thickness, and evapotranspiration. Solar-induced chlorophyll fluorescence data included: instantaneous data under clear skies and daily average data; Vegetation optical thickness includes: daytime vegetation optical depth data and nighttime vegetation optical depth data; The specific steps of the preprocessing include: Quality screening and coordinate transformation were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data. Mean synthesis and resampling were performed on hydrological and meteorological data, vegetation structure data, and vegetation function data to unify temporal and spatial resolution.

3. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S2 includes: S21: Using historical soil moisture data for the study area for the predetermined years, calculate the soil drought index as follows: in Indicates the soil drought index; Indicates the first Month and the Annual soil moisture; and They are Mean and standard deviation over all years; Filter out The pixels, as in extremely arid regions; S22: The time when the minimum soil moisture in an extremely arid region occurs is taken as the drought peak, and the time when the drought peak occurs is obtained.

4. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S3 includes: S31: Extract the long-term trend of hydrological and meteorological data using the LOWESS-based local weighted regression method; S32: Calculate the monthly average value of hydrological and meteorological data to obtain the average seasonal cycle; S33: Based on hydrological and meteorological data, remove long-term trends and average seasonal cycles to obtain the outlier form of hydrological and meteorological data.

5. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S5 includes: S51: Using hydrological and meteorological data and vegetation structure data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a first random forest model is established. S52: Using hydrological and meteorological data as independent variables and vegetation function data as dependent variables, and combining the average change trajectory, a second random forest model is established. S53: Train the first random forest model and the second random forest model using hydrological and meteorological data, vegetation structure data and vegetation function data outside the time window; S54: Using the trained first and second random forest models, predictions are made on hydrological and meteorological data and vegetation structure data outside the time window to obtain prediction results.

6. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 5, characterized in that, Step S6 includes: Calculate the residuals between the prediction results of the first random forest model and the prediction results of the second random forest model, and separate the physiological signals of vegetation function data. Based on different stages of drought development and humidity levels, and combined with physiological signals, the proportions of physiological signals of the three types of vegetation function variables in the vegetation function data were statistically analyzed.

7. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S7 includes: The time window before the drought peak is defined as the drought development period, and the time window after the drought peak is defined as the drought recovery period. Hydrometeorological data include: air temperature, rainfall, solar radiation, and soil moisture; Vegetation characteristics include: vegetation cover fraction and moisture content; The duration of drought is determined by identifying the time it takes for soil moisture to recover from outliers to normal levels before and after the drought peak.

8. The method for quantitative analysis of vegetation physiological and structural responses under extreme dry-heat combined stress as described in claim 1, characterized in that, Step S8 includes: Using the SHAP method, attribution analysis was performed on the driving variables to quantify the characteristic contributions of hydrological and meteorological data, vegetation characteristics, and drought duration to the three types of vegetation function variables, and to obtain the relative importance ranking of the driving variables to the three types of vegetation function variables.

9. An electronic device, characterized in that, The device includes a processor (501), a memory (505), a user interface (503), and a network interface (504). The memory (505) is used to store instructions. The user interface (503) and the network interface (504) are used to communicate with other devices. The processor (501) is used to execute the instructions stored in the memory (505) to cause the electronic device to perform the method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed by a computer, perform the method as described in any one of claims 1-8.

Images