A method for analyzing the correlation between the stomatal behavior of wheat and drought resistance

Abstract

The present application belongs to the field of agriculture, and particularly relates to a method for correlational analysis of wheat stomatal behavior and drought resistance, which comprises the following steps: constructing a multi-dimensional time series data set comprising a time series of stomatal conductance, a time series of transpiration rate and a time series of environmental parameters; extracting six core characteristic indexes including baseline stomatal conductance, fluctuation coefficient, response delay time and the like based on the data set; dividing wheat varieties into three types of stomatal strategy including preventive type, conservative type and balanced type by using a decision tree classification algorithm; respectively establishing correlational analysis models of stomatal characteristics and drought resistance indexes including survival rate, relative yield, water use efficiency and the like for different strategy types; and constructing a coupling relationship network among the stomatal characteristics by coupling feature analysis to deduce a regulation mechanism dominated by ion channels or signal transduction, which realizes panoramic characterization of the dynamic behavior of stomata, and provides technical support for evaluation of wheat drought resistance and variety breeding.

Description

Technical Field

[0001] This invention belongs to the field of agriculture, specifically a method for analyzing the correlation between wheat stomatal behavior and drought resistance. Background Technology

[0002] Wheat is an important global food crop, and drought stress is a core adverse factor affecting wheat yield and quality. As a key channel for plant gas exchange and water regulation, the behavior of stomata directly determines wheat water use efficiency and drought resistance. Based on stomatal characteristics, wheat drought resistance assessment is an important technical means for wheat variety selection and refined cultivation management.

[0003] Existing technologies for analyzing the correlation between wheat stomata and drought resistance mainly employ manual microscopic observation to extract single static morphological parameters such as stomatal density and stomatal aperture. These parameters are then combined with correlation analysis or machine learning classification models to complete the statistical correlation assessment between the parameters and drought resistance indicators. This enables basic analysis of wheat drought resistance performance and preliminary variety screening.

[0004] Existing technologies only focus on the extraction and analysis of static stomatal morphological features, which cannot characterize the core physiological characteristics such as stomatal dynamic response and multidimensional coupling regulation under drought stress. They are also difficult to effectively capture the subtle stomatal response differences among different wheat varieties, which greatly reduces the accuracy and reliability of drought resistance correlation analysis. This has become a core technical bottleneck restricting the screening of drought-resistant wheat germplasm and the upgrading of cultivation optimization technologies.

[0005] Therefore, this invention provides a method for analyzing the correlation between wheat stomatal behavior and drought resistance. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: a method for correlation analysis of wheat stomatal behavior and drought resistance, comprising the following steps:

[0008] Step 1: Dynamic monitoring of stomatal behavior under drought stress. Several robust wheat plants of the variety to be analyzed were selected and subjected to drought stress treatment under natural light conditions. The starting point of drought stress was defined as the soil moisture content dropping to 40% to 50% of field capacity. This starting point was determined in real time by a soil moisture sensor. Before the drought stress treatment, the leaf surface of the second leaf from the top of each wheat plant was selected as the observation area. Each observation area contained at least three different measurement sites. A portable stomatometer was used to simultaneously measure the stomatal conductance and evapotranspiration at each measurement site before the drought stress treatment and at four, eight, twelve, twenty-four, forty-eight, seventy-two, ninety-six, one hundred and twenty, one hundred and forty-four, and one hundred and sixty-eight hours after the start of the drought stress treatment. The transpiration rate was measured by taking readings three minutes after the leaf was stabilized in the stomatal chamber. Three consecutive readings were taken at each location, and the arithmetic mean was used as the measured value for that location. Simultaneously, a small weather station was used to collect atmospheric temperature, relative humidity, and light intensity data at various time points. This resulted in a multidimensional time-series dataset containing stomatal conductance, transpiration rate, and environmental parameters. The stomatal conductance time series was expressed in micromoles per square meter per second, the transpiration rate time series in millimoles per square meter per second, and the environmental parameters in the time series in degrees Celsius, relative humidity as a percentage, and photosynthetically active radiation in micromoles per square meter per second. This multidimensional time-series dataset provided a complete data foundation for subsequent feature extraction.

[0009] Step 2: Perform multidimensional quantitative characterization of stomatal dynamics. Based on the multidimensional time-series dataset constructed in Step 1, feature extraction and calculation are performed on the stomatal conductance time series for each measurement site. The specific calculation process for feature extraction is as follows: The baseline stomatal conductance value is calculated by taking the arithmetic mean of the stomatal conductance values ​​at three time points before drought stress treatment. The three time points are the measurements taken at the same time for three consecutive days before stress, which serve as the stable stomatal aperture level of the measurement site under normal moisture conditions. The stomatal conductance fluctuation coefficient is calculated by taking the ratio of the standard deviation of the stomatal conductance values ​​at all time points during drought stress treatment to the baseline stomatal conductance value, which is used to quantify the degree of stomatal aperture oscillation during drought stress. The maximum response amplitude is calculated by dividing the difference between the minimum stomatal conductance value during drought stress treatment and the baseline stomatal conductance value by the baseline stomatal conductance value and multiplying by 100% to obtain the stomatal aperture amplitude from the baseline level. The maximum percentage decrease; the response delay time is calculated by finding the time point when stomatal conductance first falls below 80% of the baseline stomatal conductance, subtracting it from the start time of drought stress treatment, and expressing the time difference in hours. This parameter characterizes the sensitivity of stomata to drought stress; the cumulative response index is calculated by approximating the difference between the stomatal conductance value at each time point during drought stress treatment and the baseline stomatal conductance value by definite integral, and normalizing it by the product of the baseline stomatal conductance value and the duration of stress. This index characterizes the cumulative response intensity of stomata during drought stress; the stomatal recovery index is calculated by the proportion of plants whose stomatal conductance recovers to more than 80% of the baseline stomatal conductance within 24 to 72 hours after rehydration following the end of drought stress. This index characterizes the recoverability of stomatal function. Based on the above calculation methods, the values ​​of six core characteristic indicators for each measurement site are obtained.

[0010] Step 3: Automatic classification and labeling of stomatal strategy types. Based on the core characteristic indicators of each measurement site calculated in Step 2, the data of all measurement sites for the same wheat variety are summarized and statistically analyzed. The arithmetic mean of each characteristic indicator of all measurement sites for the variety is calculated as the variety characteristic value. At the same time, the standard deviation of each characteristic indicator of all measurement sites for the variety is calculated as the measure of intravariety. Based on the variety characteristic value, the wheat varieties to be analyzed are divided into three stomatal strategy types according to the preset classification rules: preventive strategy, conservative strategy, and balanced strategy. The criteria for the preventive strategy are a response delay time of less than or equal to 72 hours and a maximum response amplitude of greater than or equal to 60%. Varieties that meet this condition quickly initiate stomatal strategies at the beginning of drought stress. The stomatal closure mechanism prioritizes water retention but may sacrifice some photosynthetic efficiency. The criteria for a conservative strategy are a response delay time greater than 72 hours and a cumulative response index less than or equal to 30% of the baseline cumulative value. Varieties meeting these criteria maintain low stomatal aperture throughout the drought stress period, with transpiration consumption strictly controlled but photosynthetic carbon assimilation significantly suppressed. The criteria for a balanced strategy are a response delay time between 72 and 96 hours and a fluctuation coefficient greater than or equal to 120% of the baseline fluctuation coefficient. Varieties meeting these criteria adopt differentiated stomatal regulation strategies at different drought stages, maintaining high photosynthesis in the early stages of drought and gradually tightening stomatal aperture in the severe drought stages, thereby achieving a dynamic balance between water use and carbon assimilation.

[0011] Step 4: Conduct correlation analysis between stomatal strategies and drought resistance indicators. Based on the stomatal strategy types classified in Step 3, establish correlation analysis models between stomatal characteristics and drought resistance indicators for each strategy type. The drought resistance indicators include three dimensions: survival rate, calculated by dividing the number of plants that can recover growth after drought stress by the total number of plants before stress; relative yield, calculated by dividing the grain yield of the same variety under drought stress by the grain yield of the same variety under normal water conditions; and water use efficiency, calculated by the grain yield produced per unit of water consumption under drought stress. The specific process of the correlation analysis is as follows: For each stomatal strategy type, calculate the five core stomatal characteristics of all varieties under that strategy type. The Pearson correlation coefficients between the indicators and three drought resistance indicators were used to construct a correlation matrix between stomatal features and drought resistance indicators under this strategy type. A strong correlation was defined as a correlation coefficient with an absolute value greater than or equal to 0.7, a moderate correlation as a correlation coefficient with an absolute value between 0.4 and 0.7, and a weak correlation as a correlation coefficient with an absolute value less than 0.4. Based on the correlation matrix, the key stomatal features that contribute the most to drought resistance indicators in each strategy type were identified. Among them, the key features that contribute the most to survival rate, the key features that contribute the most to relative yield, and the key features that contribute the most to water use efficiency together constitute the core evaluation indicator combination of this strategy type. A drought resistance prediction model for this strategy type was established based on the core evaluation indicator combination.

[0012] Step 5: Perform coupling characteristic analysis and inference of the regulation mechanism of stomatal behavior. Based on the core characteristic index data set calculated in Step 2, calculate the Pearson correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index, the Pearson correlation coefficient between the response delay time and the maximum response amplitude, and the Pearson correlation coefficient between the baseline stomatal conductance value and the stomatal recovery index. This constructs a coupling relationship network among the stomatal characteristic indicators. The coupling relationship network uses the stomatal characteristic indicators as network nodes and the correlation coefficients between the indicators as the weights of the network edges. A strong coupling relationship is determined when the absolute value of the correlation coefficient is greater than or equal to 0.6; a moderate coupling relationship is determined when the absolute value of the correlation coefficient is between 0.4 and 0.6; and a weak coupling relationship is determined when the absolute value of the correlation coefficient is less than 0.4. Based on the analysis of this coupling relationship network... As a result, stomatal regulation mechanisms were classified into two types: ion channel-dominated and signal transduction-dominated. The criteria for ion channel-dominated regulation mechanisms were that the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index was greater than or equal to 0.6, and the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the response delay time was also greater than or equal to 0.6. Varieties meeting these criteria exhibited fast stomatal response speed, large fluctuation amplitude, and significantly enhanced ion channel activity in guard cells. The criteria for signal transduction-dominated regulation mechanisms were those not meeting the ion channel-dominated criteria. These varieties exhibited relatively stable stomatal response, significant cumulative effects, and high participation of signal transduction substances such as abscisic acid and reactive oxygen species. Based on the coupling characteristic analysis and regulation mechanism inference results, a theoretical basis was provided for optimizing the stomatal regulation mechanisms of different varieties.

[0013] Preferably, the drought stress treatment in step one is carried out using a gradient water control method. Specifically, wheat potted plants are planted in water-controllable containers with bottom trays, and soil moisture content is controlled by precise weighing. The starting point of drought stress is set when the soil moisture content drops to 45% of field capacity. During the drought stress period, weighing and watering are carried out at noon every day, and the amount of water added is the daily evapotranspiration to maintain the set stress level. The end point of drought stress is set when the soil moisture content drops to 20% of field capacity or when the leaves show obvious wilting symptoms, and then the stress treatment is stopped. This gradient water control method can ensure that stomatal behavior can be compared between different varieties under the same stress process.

[0014] Preferably, the multidimensional time-series data acquisition frequency in step one is configured to acquire data once every six hours, with the acquisition time points fixed at midnight, 6:00 AM, 12:00 PM, and 6:00 PM. The entire monitoring cycle lasts for fourteen days. This combination of acquisition frequency and fixed time points is based on the following technical considerations: the fixed time points cover the typical characteristic periods of wheat stomatal diurnal variation. Midnight reflects the lowest stomatal activity level at night, 6:00 AM reflects the stomatal opening process, 12:00 PM reflects the typical stomatal response under high temperature and strong light, and 6:00 PM reflects the gradual closing process of stomatals. The time interval of once every six hours can capture the key turning points of stomatal response while avoiding data redundancy and computational burden caused by excessively high sampling frequency.

[0015] Preferably, the multidimensional quantitative characterization in step two further includes two extended features: a response rate parameter and a recovery rate parameter. The response rate parameter is defined as the time required for stomatal conductance to decrease from the baseline stomatal conductance value to 50% of the baseline stomatal conductance. It is calculated by finding the time point when stomatal conductance first falls below 50% of the baseline stomatal conductance, subtracting it from the time point when drought stress treatment begins, and expressing the time difference in hours. This parameter reflects the speed at which guard cells respond to drought signals. The recovery rate parameter is defined as the time required for stomatal conductance to recover from its lowest value to 50% of the baseline stomatal conductance after rehydration. It is calculated by finding the time point when stomatal conductance first exceeds 50% of the baseline stomatal conductance after rehydration, subtracting it from the time point when rehydration begins, and expressing the time difference in hours. This parameter reflects the speed at which guard cell function recovers.

[0016] Preferably, the stomatal strategy classification algorithm in step three adopts an automatic classification method based on decision trees. The decision tree classifier uses response delay time, cumulative response index, and volatility coefficient as splitting feature variables, and preventive, conservative, and balanced types as classification targets. The decision tree classifier is trained using sample specimens of known strategy types from pre-experiments. The training sample size is no less than sixty varieties, with twenty varieties for each strategy type. The depth of the decision tree is limited to four layers to prevent overfitting. This automatic classification method can avoid the subjective bias of human judgment and improve the repeatability and objectivity of the classification results.

[0017] Preferably, the method for calculating the comprehensive drought resistance score in step four is to perform a weighted summation of three drought resistance indicators, wherein the weight coefficient of the survival rate indicator is configured as 0.4, the weight coefficient of the relative yield indicator is configured as 0.4, and the weight coefficient of the water use efficiency indicator is configured as 0.2. This weight configuration reflects the following technical considerations: survival rate and relative yield are the two most important target traits in breeding practice, while water use efficiency reflects the water use economy of the variety. The weight configuration of the three is determined by optimization using variety verification data in the preliminary experiment. The objective function of optimization is to maximize the Spearman rank correlation coefficient between the predicted score and the actual drought resistance ranking.

[0018] Preferably, the coupling feature analysis in step five also includes principal component verification analysis. The five core stomatal feature indicators calculated in step two are subjected to dimensionality reduction using the principal component analysis method. The five core stomatal feature indicators include baseline stomatal conductance value, stomatal conductance fluctuation coefficient, maximum response amplitude, response delay time, and cumulative response index. The principal component analysis outputs the cumulative contribution rate of the first two principal components. When the cumulative contribution rate is greater than or equal to 70%, it is determined that the stomatal strategy classification is statistically significant. At this time, the two-dimensional scatter plot of the first two principal components can intuitively present the clustering distribution characteristics of varieties with different strategy types, thereby verifying the rationality and discriminativeness of the stomatal strategy classification.

[0019] Preferably, the method for analyzing the correlation between wheat stomatal behavior and drought resistance of the present invention further includes a stomatal spatial heterogeneity correction step. The specific method for spatial heterogeneity correction is as follows: in the data acquisition stage of step one, each observation area includes three sub-regions: leaf tip area, leaf middle area, and leaf base area. Three measurement sites are selected in each sub-region, for a total of nine sites. In the feature calculation of step two, the mean value of the feature index of each sub-region is calculated respectively. Then, the variance analysis is performed on the mean value of the feature index of the three sub-regions. When the mean value of the feature index of a certain sub-region is significantly different from the overall mean value, the data of that sub-region is downweighted when summarizing the variety characteristics. The purpose of spatial heterogeneity correction is to eliminate the influence of sampling bias caused by uneven stomatal distribution on the overall analysis results.

[0020] The beneficial effects of this invention are as follows:

[0021] 1. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention first constructs a multidimensional time-series dataset of stomatal behavior through dynamic monitoring at multiple time points. Then, it extracts six core feature indicators through multidimensional quantitative characterization to comprehensively depict the dynamic response characteristics of stomatal activity. Subsequently, it classifies wheat varieties into three stomatal strategy types—preventive, conservative, and balanced—through an automatic classification algorithm. Then, it establishes correlation analysis models between stomatal characteristics and drought resistance indicators for different strategy types. Finally, it infers the type of stomatal regulation mechanism through coupled feature analysis. This invention overcomes the shortcomings of existing technologies, such as confusion between passive response and active strategy, neglect of the conflict between survival and yield goals, inversion of response and regulation, and insufficient distinction between instantaneous state and cumulative effect.

[0022] 2. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention achieves a panoramic characterization of stomatal dynamic behavior by constructing a multidimensional time-series dataset that includes stomatal conductance time series, transpiration rate time series, and environmental parameter time series. Compared with existing technologies that only focus on instantaneous parameters at a single moment, multidimensional time-series data can completely record the complete trajectory of stomatal movement from normal state to stress response and then to recovery process, providing a data foundation for a deeper understanding of stomatal behavior. This technical feature directly solves the problem that instantaneous state and cumulative effect are ignored in existing technologies.

[0023] 3. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention uses an automatic classification algorithm to clearly classify the stomatal strategies of wheat varieties into three types: preventive, conservative, and balanced. Each strategy type corresponds to a different drought adaptation mechanism: preventive varieties reduce water loss by closing stomata in advance, conservative varieties strictly control transpiration consumption by maintaining a low stomatal opening, and balanced varieties achieve a balance between water use and photosynthesis through dynamic adjustment. This strategy classification method directly solves the problem of confusion between passive adjustment and active strategies in the prior art, enabling the stomatal behavior of different varieties to be compared under a unified analytical framework.

[0024] 4. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention establishes correlation analysis models for different stomatal strategy types, avoiding analytical bias caused by lumping varieties with different drought resistance targets together. At the same time, it introduces the concept of comprehensive drought resistance score, comprehensively considering three dimensions: survival rate, relative yield, and water use efficiency, and uses a weighted summation with a weight configuration of 0.4, 0.4, and 0.2. This method directly solves the problem of ignoring the conflict between survival and yield targets in the prior art, and can provide differentiated guidance for variety selection with different planting objectives.

[0025] 5. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention constructs a coupling relationship network among stomatal characteristic indicators through coupling feature analysis. Based on the coupling relationship, the regulatory mechanism is divided into two types: ion channel-dominated and signal transduction-dominated. This classification method reveals that there may be completely different regulatory mechanisms behind seemingly similar response performances, providing an analytical means for a deeper understanding of the intrinsic physiological mechanism of stomatal behavior. This technical feature directly solves the problem of the inversion of response and regulation in the prior art, enabling the correlation analysis to penetrate to the mechanism level.

[0026] 6. The method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention characterizes the degree of variation among different individuals of the same variety by calculating the standard deviation within the variety and characterizes the typical characteristics of the variety by using the mean within the variety. This method can distinguish the contribution of genetic variation and environmental variation to stomatal phenotype, and provides a technical means for identifying genetically stable drought resistance evaluation indicators. This technical feature directly solves the problem that genetic stability and phenotypic plasticity are not distinguished in the prior art. Attached Figure Description

[0027] The invention will now be further described with reference to the accompanying drawings.

[0028] Figure 1 This is a flowchart of a method for analyzing the correlation between wheat stomatal behavior and drought resistance in this invention;

[0029] Figure 2 This is a logical flowchart of the inference of the stomatal feature coupling relationship network and the type of regulation mechanism in this invention. Detailed Implementation

[0030] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0031] Example 1:

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0033] like Figure 1 and Figure 2 As shown, the method for analyzing the correlation between wheat stomatal behavior and drought resistance provided by this invention includes the following steps in its overall technical architecture:

[0034] In step one, the dynamic monitoring of stomatal behavior under drought stress conditions specifically includes the following procedures: Selecting robust wheat plants of the variety to be analyzed as experimental materials, with the selection criteria being mature plants of uniform height, normal leaf color, and no diseases or pests, and selecting no fewer than fifteen plants for each variety to ensure the statistical reliability of the data; planting the selected wheat plants in pots in a controllable water container with a bottom tray, wherein the controllable water container is a plastic pot with a water outlet at the bottom, the pot volume is five liters, and the bottom tray is used to collect seepage water to prevent water loss from affecting the weighing accuracy;

[0035] The drought stress treatment was carried out using a gradient water control method, which controlled soil moisture content through precise weighing. The specific operation procedure is as follows: The pot containing wheat plants was placed on a professional electronic balance with an accuracy of ±0.1 gram. Weighing was carried out at noon every day and the data was recorded. The evapotranspiration of the day was calculated based on the change in the weight of the pot, and the corresponding weight of water was added to maintain the set stress level. The starting point of drought stress was set at when the soil moisture content dropped to 45% of field capacity, and the ending point of drought stress was set at when the soil moisture content dropped to 20% of field capacity or when the leaves showed obvious wilting symptoms.

[0036] Regarding data acquisition, the multidimensional time-series data acquisition frequency is configured to acquire data once every six hours, with the acquisition time points fixed at 0:00 AM, 6:00 AM, 12:00 PM, and 6:00 PM, and the entire monitoring cycle lasts for fourteen days; before the drought stress treatment, the leaf surface of the second leaf from the top of each wheat plant is selected as the observation area, and each observation area contains at least three different measurement sites.

[0037] The specific method for selecting measurement sites is as follows: each observation area includes three sub-regions: the leaf tip region, the leaf middle region, and the leaf base region. Three measurement sites are selected in each sub-region, for a total of nine sites. The three sub-regions are located in three equally divided areas from the leaf base to the leaf tip. The three measurement sites in each sub-region are arranged linearly with a spacing of three millimeters between the sites. The leaf tip region is defined as the first third of the distance from the leaf base to the leaf tip, the leaf middle region is defined as the middle third of the distance from the leaf base to the leaf tip, and the leaf base region is defined as the last third of the distance from the leaf base to the leaf tip.

[0038] A portable stomatalometer was used to simultaneously measure stomatal conductance and transpiration rate. During measurement, the leaf was allowed to stabilize in the stomatalometer chamber for three minutes before the measurement was taken. Data was taken three times consecutively at each location, and the arithmetic mean was used as the measured value for that location. The portable stomatalometer was selected to be capable of measuring stomatal conductance in the range of 0 to 1,000 micromoles per square meter per second with a resolution of not less than 0.1 micromoles per square meter per second. During measurement, the stomatalometer chamber was kept in close contact with the leaf surface to ensure a stable measurement environment. At the same time, a small weather station was used to collect atmospheric temperature, relative humidity, and light intensity data at various time points. The small weather station was installed in the center of the experimental site, 1.5 meters above the plant canopy.

[0039] The multidimensional time-series dataset includes stomatal conductance time series, transpiration rate time series, and environmental parameter time series. The stomatal conductance time series is in micromoles per square meter per second, the transpiration rate time series is in millimoles per square meter per second, and the atmospheric temperature, relative humidity, and photosynthetically active radiation in the environmental parameter time series are in degrees Celsius, percentage, and micromoles per square meter per second.

[0040] Specific measurement time points included before drought stress treatment, and at four, eight, twelve, twenty-four, forty-eight, seventy-two, ninety-six, one hundred and twenty, one hundred and forty-four hours, and one hundred and sixty-eight hours after the start of drought stress treatment. Measurements were taken continuously for three days before the stress treatment to obtain stable stomatal baseline data. During the stress treatment, measurements were taken at fixed time points every day, for a total of eleven measurement time points. Five plants of each variety were randomly selected for detailed observation of stomatal behavior, and the remaining ten plants were used as backup and supplementary data sources.

[0041] In step two, the multidimensional quantitative characterization of stomatal dynamics is specifically calculated based on the multidimensional time-series dataset constructed in step one. Feature extraction and calculation are performed on the stomatal conductance time series of each measurement site to obtain the values ​​of six core feature indicators.

[0042] The first characteristic indicator is the baseline stomatal conductance value, which is calculated by taking the arithmetic mean of the stomatal conductance values ​​at three time points before the drought stress treatment. The three time points are the measurements taken at the same time on three consecutive days before the stress, and are used as the stable stomatal aperture level of that measurement site under normal moisture conditions. In the specific calculation, the stomatal conductance data of three measurement time points on three consecutive days before the drought stress treatment are extracted, and the calculation formula is as follows:

[0043] ,in, Indicates the baseline stomatal conductance value. Indicates the first Porous conductance measurements at various time points.

[0044] The second characteristic indicator is the stomatal conductance fluctuation coefficient, which is calculated by taking the ratio of the standard deviation of stomatal conductance values ​​at all time points during the drought stress treatment to the baseline stomatal conductance value. This is used to quantify the degree of stomatal opening oscillation during drought stress. The calculation of the fluctuation coefficient enhances the comparability between different varieties because the fluctuation coefficient has one dimension, eliminating the influence of baseline level differences. In specific calculation, stomatal conductance data at all measurement time points during the drought stress treatment are extracted, and the standard deviation is calculated and then divided by the baseline stomatal conductance value.

[0045] The third characteristic indicator is the maximum response amplitude, which is calculated by dividing the difference between the minimum stomatal conductance during drought stress treatment and the baseline stomatal conductance value by the baseline stomatal conductance value and multiplying by 100% to obtain the maximum percentage decrease in stomatal closure from the baseline level. This indicator represents the upper limit of stomatal closure capacity and reflects the water retention potential of the variety under extreme drought conditions.

[0046] The fourth characteristic indicator is the response delay time. It is calculated by finding the time point when the stomatal conductance first falls below 80% of the baseline stomatal conductance, subtracting it from the time point when the drought stress treatment begins, and expressing the time difference in hours. This parameter characterizes the sensitivity of stomata to drought stress. The shorter the response delay time, the faster the variety perceives and responds to drought signals.

[0047] The fifth characteristic index is the cumulative response index, which is calculated by approximating the difference between the stomatal conductance value at each time point during the drought stress treatment and the baseline stomatal conductance value by definite integral, and normalizing it by the product of the baseline stomatal conductance value and the duration of stress. This index characterizes the cumulative response intensity of stomata during drought stress. The definite integral approximation uses the trapezoidal rule, which discretizes the stomatal conductance time series during the stress treatment and then sums it.

[0048] The sixth characteristic indicator is the stomatal recovery index, which is calculated as the proportion of plants whose stomatal conductance recovers to more than 80% of the baseline stomatal conductance within 24 to 72 hours after the end of drought stress and rehydration. This index represents the degree of recovery of stomatal function. This index is a population indicator rather than an individual indicator, reflecting the overall ability of the variety to recover physiological function after the stress is relieved.

[0049] In step two, in addition to the six core characteristic indicators mentioned above, the multidimensional quantitative characterization also includes two extended features: the response rate parameter is defined as the time required for stomatal conductance to decrease from the baseline stomatal conductance value to 50% of the baseline stomatal conductance. It is calculated by finding the point in time when stomatal conductance first falls below 50% of the baseline stomatal conductance, subtracting this point from the start time of drought stress treatment, and expressing the time difference in hours. This parameter reflects the speed of guard cell response to drought signals. The recovery rate parameter is defined as the time required for stomatal conductance to recover from its lowest value to 50% of the baseline stomatal conductance after rehydration. It is calculated by finding the point in time when stomatal conductance first exceeds 50% of the baseline stomatal conductance after rehydration, subtracting this point from the start time of rehydration, and expressing the time difference in hours. This parameter reflects the speed of guard cell function recovery.

[0050] In step three, the automatic classification and labeling of stomatal strategy types is carried out as follows: Based on the core feature indicators of each measurement site calculated in step two, the data of all measurement sites for the same wheat variety are summarized and statistically analyzed; firstly, the arithmetic mean of each feature indicator of all measurement sites for this variety is calculated as the variety feature value, and the calculation formula is as follows:

[0051] ,in, This represents a specific characteristic index value of a variety. Indicates the first The value of this characteristic index at each measurement site, This represents the total number of measurement sites for this variety; simultaneously, the standard deviation of each characteristic index for all measurement sites of this variety is calculated as a measure of intra-variety variation. The formula for calculating the standard deviation is as follows:

[0052] The calculation of intravariety variation measures is used to assess the stability of stomatal behavior and provide data support for subsequent genetic stability analysis.

[0053] Based on varietal characteristics, the wheat varieties to be analyzed are classified into three types of stomatal strategies using preset classification rules: preventive strategy, conservative strategy, and balanced strategy.

[0054] The classification rules adopt an automatic classification method based on decision trees. The decision tree classifier uses response delay time, cumulative response index, and volatility coefficient as splitting feature variables, and preventive, conservative, and balanced types as classification targets. The training sample size of the decision tree classifier is no less than sixty varieties, of which twenty varieties are for each strategy type. The depth of the decision tree is limited to four layers to prevent overfitting. The specific classification judgment conditions are as follows.

[0055] The criteria for determining a preventative strategy are a response delay time of less than or equal to 72 hours and a maximum response amplitude of greater than or equal to 60%. Varieties that meet these criteria can quickly activate stomatal closure mechanisms at the beginning of drought stress to prioritize water retention, but may sacrifice some photosynthetic efficiency.

[0056] The physiological mechanism of the preventive strategy is that after the variety senses the early signal of soil moisture decline, it rapidly synthesizes and accumulates signaling molecules such as abscisic acid, triggering the rapid activation of ion channels in guard cells, promoting the outflow of potassium and chloride ions, resulting in a rapid decrease in guard cell turgor pressure and rapid stomata closure.

[0057] The criteria for a conservative strategy are a response delay time greater than 72 hours and a cumulative response index less than or equal to 30% of the baseline cumulative value. Varieties that meet these criteria maintain a low stomatal aperture throughout the drought stress period, with transpiration consumption strictly controlled but photosynthetic carbon assimilation significantly inhibited. The physiological mechanism of the conservative strategy is that the variety maintains the stomatal in a semi-closed state by continuously regulating the activity of ion channels in guard cells, thereby reducing water transpiration while retaining basic photosynthetic capacity to survive the long-term drought environment.

[0058] The criteria for a balanced strategy are a response delay time between 72 and 96 hours and a fluctuation coefficient greater than or equal to 120 percent of the baseline fluctuation coefficient. Varieties that meet these criteria adopt differentiated stomatal regulation strategies at different stages of drought, maintaining high photosynthesis in the early stages of drought and gradually tightening stomatal aperture in the severe drought stage, thereby achieving a dynamic balance between water use and carbon assimilation. The physiological mechanism of the balanced strategy is that the variety dynamically adjusts stomatal aperture according to the degree of drought stress, maintaining high stomatal conductance in the mild stress stage to ensure photosynthetic carbon assimilation, and gradually tightening stomatal aperture in the severe stress stage to reduce water loss.

[0059] In step four, the correlation analysis between stomatal strategies and drought resistance indicators is performed, and the specific process is as follows: Based on the stomatal strategy types obtained in step three, correlation analysis models between stomatal characteristics and drought resistance indicators are established for each strategy type; the drought resistance indicators include three dimensions:

[0060] The first dimension is the survival rate index, which is calculated by dividing the number of plants that can recover growth after the drought stress ends by the total number of plants before the stress.

[0061] The second dimension is the relative yield index, which is calculated by dividing the grain yield of the same variety under drought stress conditions by the grain yield of the same variety under normal moisture conditions.

[0062] The third dimension is the water use efficiency index, which is calculated based on the grain yield produced per unit of water consumption under drought stress.

[0063] For each stomatal strategy type, the Pearson correlation coefficients between the five core stomatal characteristic indicators and the three drought resistance indicators for all varieties under that strategy type were calculated, and a correlation matrix between stomatal characteristics and drought resistance indicators under that strategy type was constructed. The formula for calculating the Pearson correlation coefficient is as follows:

[0064] ,in, and They represent the first The values ​​of a certain stomatal characteristic index and a certain drought resistance index for a certain variety. and These represent the mean values ​​of the stomatal characteristic index and the drought resistance index, respectively. This indicates the total number of varieties under this strategy type; when the absolute value of the correlation coefficient is greater than or equal to 0.7, it is considered a strong correlation; when the absolute value of the correlation coefficient is between 0.4 and 0.7, it is considered a moderate correlation; and when the absolute value of the correlation coefficient is less than 0.4, it is considered a weak correlation.

[0065] Based on the correlation matrix, the key stomatal features that contribute most to drought resistance indicators for each strategy type were identified: the key features that contribute most to survival rate, the key features that contribute most to relative yield, and the key features that contribute most to water use efficiency together constitute the core evaluation index combination for this strategy type. A drought resistance prediction model for this strategy type was established based on this core evaluation index combination. The prediction model uses a multiple linear regression method, and the model form is as follows:

[0066] ,in, This indicates the overall score for drought resistance. , , These represent three key stomatal characteristic indicators in the core evaluation indicator combination. , , , This represents the regression coefficient.

[0067] The comprehensive drought resistance score is calculated by weighted summation of three drought resistance indicators, where the weight coefficient for the survival rate indicator is set to 0.4, the weight coefficient for the relative yield indicator is set to 0.4, and the weight coefficient for the water use efficiency indicator is set to 0.2. This weighting configuration is optimized and determined using variety validation data from preliminary experiments. The objective function of the optimization is to maximize the Spearman rank correlation coefficient between the predicted score and the actual drought resistance ranking.

[0068] In step five, the specific process of performing coupled feature analysis and inference of the regulation mechanism of stomatal behavior is as follows:

[0069] Based on the core feature index data set obtained in step two, the Pearson correlation coefficient between stomatal feature indices is calculated, and a coupling relationship network between stomatal feature indices is constructed.

[0070] The specific correlation coefficients calculated include: the Pearson correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index, the Pearson correlation coefficient between the response delay time and the maximum response amplitude, and the Pearson correlation coefficient between the baseline stomatal conductance value and the stomatal recovery index.

[0071] The coupling relationship network uses stomatal feature indicators as network nodes and correlation coefficients between indicators as weights of network edges. When the absolute value of the correlation coefficient is greater than or equal to 0.6, it is determined to be a strong coupling relationship; when the absolute value of the correlation coefficient is between 0.4 and 0.6, it is determined to be a medium coupling relationship; and when the absolute value of the correlation coefficient is less than 0.4, it is determined to be a weak coupling relationship.

[0072] Based on the analysis results of the coupling relationship network, the stomatal regulation mechanism is divided into two types: ion channel-dominated and signal transduction-dominated.

[0073] The criteria for determining the ion channel-dominated regulatory mechanism are that the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index is greater than or equal to 0.6 and the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the response delay time is greater than or equal to 0.6.

[0074] Varieties that meet this condition exhibit rapid stomatal response, large fluctuation range, and significantly enhanced ion channel activity in guard cells. The physiological mechanism is that guard cells rapidly activate ion channels on the vacuolar membrane and plasma membrane, promoting the transmembrane transport of potassium, chloride, and malate ions, leading to a rapid decrease in the osmotic potential of guard cells, a reduction in cell sap water potential, cell dehydration and shrinkage, and rapid stomata closure.

[0075] The criteria for determining a signal transduction-dominant regulatory mechanism are varieties that do not meet the criteria for an ion channel-dominant mechanism. These varieties exhibit relatively stable stomatal responses, significant cumulative effects, and high levels of involvement of signal transduction substances such as abscisic acid and reactive oxygen species.

[0076] The physiological mechanism of varieties that meet this condition is that guard cells synthesize and accumulate signaling molecules such as abscisic acid and jasmonic acid by sensing drought signals from the soil and atmosphere. These signaling molecules activate downstream ion channels and enzymes through signal cascade amplification pathways, indirectly regulating stomatal opening.

[0077] In step five, the coupled feature analysis also includes principal component verification analysis. Principal component analysis is used to reduce the dimensionality of the six core stomatal feature indicators calculated in step two. The five core stomatal feature indicators include baseline stomatal conductance value, stomatal conductance fluctuation coefficient, maximum response amplitude, response delay time, cumulative response index, and stomatal recovery index. The results of principal component analysis output the cumulative contribution rate of the first two principal components. When the cumulative contribution rate is greater than or equal to 70%, it is determined that the stomatal strategy classification is statistically significant. At this time, the two-dimensional scatter plot of the first two principal components can intuitively present the clustering distribution characteristics of varieties with different strategy types, thereby verifying the rationality and discriminative power of the stomatal strategy classification.

[0078] The method for analyzing the correlation between wheat stomatal behavior and drought resistance in this invention also includes a stomatal spatial heterogeneity correction step. The specific method for spatial heterogeneity correction is as follows: In the data acquisition stage of step one, each observation region includes three sub-regions: the leaf tip region, the leaf middle region, and the leaf base region. Three measurement sites are selected in each sub-region, for a total of nine sites. In the feature calculation stage of step two, the mean values ​​of the feature indicators for each sub-region are calculated. Then, variance analysis is performed on the mean values ​​of the feature indicators for the three sub-regions. When the mean value of the feature indicator for a certain sub-region differs significantly from the overall mean, the data for that sub-region is weighted less during the variety feature aggregation. The purpose of spatial heterogeneity correction is to eliminate the influence of sampling bias caused by uneven stomatal distribution on the overall analysis results.

[0079] The technical solution of the present invention will be further illustrated below with specific application examples; three wheat varieties, A, B and C, were selected for experimental verification. Fifteen robust plants of each variety were selected and subjected to drought stress treatment and dynamic monitoring of stomatal behavior according to the above method; during the monitoring period, stomatal conductance data and meteorological data were collected at fixed time points to construct a multidimensional time-series dataset; six core feature indicators and two extended features were calculated for each measurement site to obtain the feature indicator set for each variety;

[0080] Based on the varietal characteristics, a decision tree classifier was used to classify the stomatal strategy types of the three varieties. The results showed that: Variety A had a response delay of 56 hours and a maximum response amplitude of 72%, and was classified as a preventative strategy; Variety B had a response delay of 96 hours and a cumulative response index of 25% of the baseline cumulative value, and was classified as a conservative strategy; Variety C had a response delay of 84 hours and a volatility coefficient of 135% of the baseline volatility coefficient, and was classified as a balanced strategy.

[0081] For preventative strategies, a correlation analysis model between stomatal characteristics and drought resistance indicators was established. Pearson correlation coefficients were calculated between five core stomatal characteristic indicators and three drought resistance indicators. The results showed that the correlation coefficient between response delay time and survival rate was -0.75, and the correlation coefficient between stomatal recovery index and water use efficiency was +0.68. Response delay time and stomatal recovery index were identified as key stomatal characteristics. Based on these key stomatal characteristics, a drought resistance prediction model was established, with the following expression: ,in To provide a standardized value for the response delay time, This is the standardized value of the stomatal recovery index.

[0082] For conservative and balanced strategy types, correlation analysis and prediction models were established using the same method. Coupled characteristic analysis was performed on all varieties to calculate the coupling relationship between stomatal characteristic indicators. The coupling analysis results for variety A showed that the correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index was -0.68, and the correlation coefficient between the stomatal conductance fluctuation coefficient and the response delay time was -0.63, indicating that variety A is characterized by an ion channel-dominated regulatory mechanism. The coupling analysis results for varieties B and C showed that both correlation coefficients were less than 0.6, indicating that they are characterized by a signal transduction-dominated regulatory mechanism.

[0083] Principal component analysis was used to reduce the dimensionality of the characteristic indicators of three types of varieties: preventive strategy, conservative strategy, and balanced strategy. The cumulative contribution rate of the first two principal components was 78%, which meets the criterion of greater than or equal to 70%. The two-dimensional scatter plot of the first two principal components showed that the three types of varieties exhibited obvious clustering distribution characteristics, proving that the stomatal strategy classification has statistical rationality and good discriminative power.

[0084] Example 2:

[0085] The present invention also provides an alternative implementation method, which differs from Embodiment 1 in that the data acquisition strategy and feature extraction method are optimized. In terms of the data acquisition strategy, a continuous monitoring method with higher temporal resolution is adopted. Specifically, during the peak period of drought stress treatment, that is, from 24 hours to 96 hours after the start of stress, the data acquisition frequency is increased to once every 15 minutes, and during other periods, it is kept to be once every 6 hours. This strategy can obtain more dense data points during the critical transition period of stomatal response and improve the accuracy of feature extraction.

[0086] In terms of feature extraction methods, time series analysis is introduced to calculate the autocorrelation function and cross-correlation function of stomatal conductance time series. The autocorrelation function is used to characterize the time dependence of stomatal conductance time series, and the magnitude of the autocorrelation function value with a lag time of one measurement interval reflects the inertia of stomatal behavior. The cross-correlation function is used to characterize the time lag relationship between stomatal conductance and environmental parameters, and the peak lag time of the cross-correlation function reflects the lag degree of stomatal response relative to environmental changes. The autocorrelation function features and cross-correlation function features are incorporated into the multi-dimensional quantitative characterization system of stomatal dynamic features, expanding the feature indicators from seven core features to more dimensions including autocorrelation features and cross-correlation features.

[0087] In stomatal strategy classification, a support vector machine (SVM) classifier is used instead of a decision tree classifier. The SVM classifier uses response delay time, cumulative response exponent, volatility coefficient, and autocorrelation features as input feature vectors, and uses preventive, conservative, and balanced types as classification labels. It uses radial basis functions as kernel functions, and the kernel function parameters are optimized and determined through cross-validation. The advantage of the SVM classifier is that it can handle nonlinear classification boundaries in the feature space and maintain good classification performance even when the feature dimension is high.

[0088] Comparative example:

[0089] To verify the technical effect of the present invention, typical methods in the prior art were selected as comparative examples for comparative analysis. Comparative example 1 used the static stomatal parameter method to evaluate drought resistance. Specifically, the second leaf from the top of each variety was collected during the wheat grain-filling stage. After preparation, the stomatal morphology was observed under a microscope, and four static parameters, namely stomatal density, stomatal long axis length, stomatal short axis length, and stomatal area, were measured. The differences in static parameters among varieties were statistically analyzed, and parameters with significant differences were used as drought resistance evaluation indicators. Comparative example 2 used the instantaneous stomatal response method to evaluate drought resistance. Specifically, the stomatal conductance and transpiration rate of each variety were measured forty-eight hours after drought stress treatment, and the response rate parameter was calculated. The response rate parameter was used as a drought resistance prediction indicator. Comparative example 3 used the traditional regression analysis method for correlation analysis. Specifically, without distinguishing stomatal strategy types, the stomatal parameters of all varieties were uniformly regressed with drought resistance indicators to establish a drought resistance prediction model.

[0090] The drought resistance of the same batch of wheat varieties was evaluated using the method of this invention and the comparative method. There were twenty varieties, covering three strategy types: preventive, conservative, and balanced. The Spearman rank correlation coefficient between the drought resistance evaluation results and the actual drought resistance ranking was used as the evaluation index. The larger the correlation coefficient, the higher the prediction accuracy.

[0091] Table 1: Comparison of Embodiments of the Invention with Comparative Examples

[0092] Comparison Projects Embodiment 1 of the present invention Embodiment 2 of the present invention Comparative Example 1 Comparison ratio 2 Comparative Example 3 Number of evaluation indicators 6 core features + 2 extended features 8 features + autocorrelation features 4 static parameters 2 instantaneous parameters 5 unified parameters Strategy type distinction Three types of classification Three types of classification Undistinguished Undistinguished Undistinguished Time dimension coverage Full-process time-series monitoring High-resolution time series monitoring Single point sampling Single point sampling Single point sampling Spearman correlation coefficient 0.89 0.92 0.52 0.61 0.68 Identification of regulatory mechanisms support support Not supported Not supported Not supported Spatial Heterogeneity Correction support support Not supported Not supported Not supported

[0093] The above comparative data fully demonstrates that the intelligent evaluation technology solution of this invention, which combines multi-dimensional feature fusion with time-series full-process monitoring, is significantly superior to the comparative schemes that rely only on a few static or instantaneous parameters and lack the ability to distinguish and correct, in terms of evaluation index dimensions, strategy type differentiation ability, time dimension coverage, Spearman correlation coefficient, control mechanism identification, and spatial heterogeneity correction. Embodiment 1 of this invention is based on 7 core features plus 2 extended features, while Embodiment 2 expands to 9 features and introduces autocorrelation features. Both achieve clear classification of the three strategy types and full-process (…). The three comparative examples used only 2-5 static or instantaneous parameters, without strategy differentiation and only single-point sampling. In terms of correlation evaluation, the Spearman correlation coefficients of Example 1 and Example 2 reached 0.89 and 0.92 respectively, which are much higher than the 0.52-0.68 of the comparative examples. In addition, all embodiments of the present invention support the identification of control mechanisms and spatial heterogeneity correction, while none of the comparative examples have the above functions. The synergistic effect of multi-dimensional feature time series acquisition, strategy classification and spatial heterogeneity correction enables the reliability of pattern recognition, dynamic monitoring and evaluation in complex scenarios to reach the optimal level.

[0094] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing the correlation between wheat stomatal behavior and drought resistance, comprising the following steps: Step 1: Conduct dynamic monitoring of stomatal behavior under drought stress conditions, and construct a multidimensional time-series dataset based on the monitoring data; Step 2: Perform multi-dimensional quantitative characterization of stomatal dynamics and calculate six core characteristic indicators; Step 3: Perform automatic classification and labeling of stomatal strategy types, and classify wheat varieties into three strategy types: preventive, conservative, and balanced, according to the classification rules; Step 4: Conduct correlation analysis between stomatal strategies and drought resistance indicators, and establish correlation analysis models between stomatal characteristics and drought resistance indicators under each strategy type; Step 5: Perform coupling feature analysis and inference of regulation mechanism of stomatal behavior, and classify the regulation mechanism type based on the coupling relationship network.

2. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the step of dynamic monitoring of stomatal behavior under drought stress, the drought stress treatment adopts a gradient water control method. Soil moisture content is controlled by precise weighing. The starting point of drought stress is set when the soil moisture content drops to 45% of field capacity. During the drought stress period, weighing and water replenishment are carried out at noon every day. The amount of water replenishment is calculated based on the daily evapotranspiration. The endpoint of drought stress is when the soil moisture content drops to 20% of field capacity or the leaves show obvious wilting symptoms.

3. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, The step of dynamic monitoring of stomatal behavior under drought stress also includes a data acquisition process. The data acquisition is carried out at a fixed frequency within the monitoring period, with the acquisition time points fixed at 0:00 AM, 6:00 AM, 12:00 PM, and 6:00 PM. At each time point, stomatal conductance, transpiration rate, atmospheric temperature, relative humidity, and light intensity data are collected synchronously. During measurement, the data at each measurement site are read three times consecutively and the arithmetic mean is taken. This constructs a multidimensional time-series dataset containing stomatal conductance time series, transpiration rate time series, and environmental parameter time series.

4. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the multidimensional quantitative characterization step of stomatal dynamics, feature extraction and calculation are performed on the stomatal conductance time series of each measurement site based on the multidimensional time series dataset. The baseline stomatal conductance value is calculated by taking the arithmetic mean of the stomatal conductance values ​​at the three time points before drought stress treatment. The stomatal conductance fluctuation coefficient is calculated by taking the ratio of the standard deviation of the stomatal conductance values ​​at all time points during drought stress treatment to the baseline stomatal conductance value. The maximum response amplitude is calculated by dividing the difference between the minimum stomatal conductance value during drought stress treatment and the baseline stomatal conductance value by the baseline stomatal conductance value and multiplying by 100%. The method for calculating the delay time is to find the time point when the stomatal conductance first falls below 80% of the baseline stomatal conductance, and subtract it from the time point at which the drought stress treatment begins. The method for calculating the cumulative response index is to approximate the difference between the stomatal conductance value at each time point during the drought stress treatment and the baseline stomatal conductance value by performing a definite integral, and then normalize it with the baseline cumulative value. The method for calculating the stomatal recovery index is to calculate the proportion of plants whose stomatal conductance recovers to more than 80% of the baseline stomatal conductance during the 24 to 72 hours after the end of the drought stress and after rehydration. The results are based on six core characteristic indicators for multidimensional quantitative characterization.

5. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, The multidimensional quantitative characterization step for stomatal dynamics also includes two extended features: response rate parameter and recovery rate parameter. The response rate parameter is defined as the time required for stomatal conductance to decrease from the baseline stomatal conductance value to 50% of the baseline stomatal conductance. The recovery rate parameter is defined as the time required for stomatal conductance to recover from the lowest value to 50% of the baseline stomatal conductance after rehydration.

6. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the step of automatically classifying and labeling stomatal strategy types, the data of all measurement sites for the same wheat variety are summarized and statistically analyzed. The arithmetic mean of each characteristic index is calculated as the variety characteristic value, and the standard deviation of each characteristic index is calculated as the measure of intravariety. Based on the variety characteristic value, the variety is divided into three stomatal strategy types: preventive, conservative, and balanced, according to preset classification rules. The criteria for the preventive strategy are a response delay time of less than or equal to 72 hours and a maximum response amplitude of greater than or equal to 60%. The criteria for the conservative strategy are a response delay time of greater than 72 hours and a cumulative response index of less than or equal to 30% of the baseline cumulative value. The criteria for the balanced strategy are a response delay time between 72 and 96 hours and a fluctuation coefficient of greater than or equal to 120% of the baseline fluctuation coefficient.

7. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the step of automatically classifying and labeling the porosity strategy types, the porosity strategy classification adopts an automatic classification method based on decision trees, using response delay time, cumulative response index, and volatility coefficient as splitting feature variables, and preventive, conservative, and balanced types as classification targets.

8. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the step of performing correlation analysis between stomatal strategies and drought resistance indicators, a correlation analysis model between stomatal features and drought resistance indicators is established for each strategy type according to the stomatal strategy types obtained by classification. The drought resistance indicators include three dimensions: survival rate, relative yield, and water use efficiency. The Pearson correlation coefficient between the five core stomatal feature indicators and the three drought resistance indicators is calculated, and a correlation matrix between stomatal features and drought resistance indicators is constructed. Based on the correlation matrix, the key stomatal feature combinations that contribute the most to each drought resistance indicator are identified, and drought resistance prediction models for each strategy type are established based on the key stomatal feature combinations.

9. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 8, characterized in that, The comprehensive drought resistance score is calculated by weighting and summing three drought resistance indicators. The weighting coefficient for the survival rate indicator is 0.4, the weighting coefficient for the relative yield indicator is 0.4, and the weighting coefficient for the water use efficiency indicator is 0.

2.

10. The method for analyzing the correlation between wheat stomatal behavior and drought resistance according to claim 1, characterized in that, In the step of analyzing the coupling characteristics of stomatal behavior and inferring the regulatory mechanism, Pearson correlation coefficients between stomatal characteristic indicators are calculated based on the core characteristic indicator data set of the variety. The core characteristic indicator data set includes baseline stomatal conductance value, stomatal conductance fluctuation coefficient, maximum response amplitude, response delay time, and cumulative response index. Pearson correlation coefficients are calculated between the stomatal conductance fluctuation coefficient and the cumulative response index, between the response delay time and the maximum response amplitude, and between the baseline stomatal conductance value and the stomatal recovery index. Based on this, a coupling relationship network between stomatal characteristic indicators is constructed. When the absolute value of the correlation coefficient is greater than or equal to 0.6, it is determined to be a strong coupling relationship. The criteria for determining the ion channel-dominated regulatory mechanism are that the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the cumulative response index is greater than or equal to 0.6 and the absolute value of the correlation coefficient between the stomatal conductance fluctuation coefficient and the response delay time is greater than or equal to 0.

6. The criteria for determining the signal transduction-dominated regulatory mechanism are that the ion channel-dominated criteria are not met. Based on this, the regulatory mechanism types are classified.

Images