Method for estimating canopy transpiration of shallow-rooted plants

Abstract

This invention relates to the field of plant canopy transpiration estimation, specifically a method for estimating canopy transpiration in shallow-rooted plants. The invention collects readily available surface environmental factors such as total radiation, vapor pressure deficit, soil moisture content at 20cm depth, canopy leaf area index, and wind speed. It then uses the envelope method to extract the corresponding standardized canopy transpiration rate values ​​and fits control functions one by one. Finally, it constructs a canopy transpiration estimation model by multiplying the output values ​​of each control function, the leaf area index, and the canopy transpiration rate per unit leaf area. This effectively avoids the shortcomings of traditional Penman-Monteith-type models, which require a large number of difficult-to-obtain parameters such as crop coefficients and canopy impedance, significantly lowering the barrier to model use. Furthermore, considering the root distribution characteristics of shallow-rooted plants, it uses soil moisture content at a depth of 20cm to improve the accuracy of canopy transpiration estimation.

Description

Technical Field

[0001] This invention relates to the field of plant canopy transpiration estimation, specifically a method for estimating canopy transpiration in shallow-rooted plants. Background Technology

[0002] Direct measurement of vegetation canopy transpiration typically requires expensive instruments, such as thermal runoff meters and eddy covariance systems, and necessitates regular maintenance by professionals, resulting in high monitoring costs. Therefore, estimation through mathematical simulation has become the primary approach for estimating vegetation transpiration. Currently, classic surface evapotranspiration or potential evapotranspiration models, such as the Penman-Monteith model or its improved versions, such as the potential evapotranspiration model and the HS model, are widely used to estimate potential surface evapotranspiration. However, these models often overemphasize atmospheric physical factors and lack sufficient consideration of biological factors affecting transpiration, such as the plant response to soil drought through stomatal and root water absorption. Directly using these models to estimate vegetation transpiration can lead to significant errors.

[0003] To address these shortcomings, existing technologies have further revised or improved the Penman-Monteith model. Key measures include incorporating soil moisture influence coefficients, crop coefficients, and canopy impedance simulations to enhance the model's predictive capabilities. Representative improvements include the FAO-56 recommended crop evaporation model, the Jarvis model, and the Berry model. However, these vegetation evaporation models, based on improvements to the Penman-Monteith model, suffer from numerous variables and parameters, placing high demands on users. For instance, when using crop evaporation models, data related to crop coefficients in vegetation or trees are scarce. The Jarvis model involves complex environmental factors and their control relationships, requiring a substantial understanding of relevant literature and application experience to obtain realistic results. Furthermore, in the Berry model, net photosynthetic rate and leaf surface CO2 concentration typically exhibit diurnal and seasonal variations, with significant differences across different parts of the canopy, making accurate numerical values ​​difficult to obtain. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for estimating canopy transpiration in shallow-rooted plants. This method solves the problem that current vegetation evaporation models based on the Penman-Monteith model have too many variables or parameters when estimating canopy transpiration, and thus require a high level of expertise from users.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for estimating canopy transpiration in shallow-rooted plants, comprising the following steps: S1. Collect surface environmental factors and canopy transpiration rate per unit leaf area in the sample area over a period of time, and obtain the maximum canopy transpiration rate per unit leaf area during the same period. Use the ratio of the canopy transpiration rate per unit leaf area to the maximum canopy transpiration rate per unit leaf area as the standardized canopy transpiration rate value, and combine it with surface environmental factors to obtain sample data. The surface environmental factors include at least total radiation, water vapor pressure deficit, soil moisture content at 20 cm depth, canopy leaf area index, and wind speed. S2. For different variation ranges of multiple surface environmental factors in the sample data, the corresponding standardized canopy evaporation rate values ​​are extracted using the envelope method, which are used as the envelope data points of each surface environmental factor. S3. By fitting multiple preset control functions to characterize the influence of surface environmental factors on canopy transpiration one by one through the outer envelope data points, the parameters in each control function and the control function after the parameters are determined are obtained. S4. Define the canopy transpiration estimation model as the product of the output values ​​of each control function, the canopy leaf area index (LAI), and the maximum canopy transpiration rate per unit leaf area. S5. Obtain multiple surface environmental factors of the area to be evaluated, and substitute them into the corresponding control function and canopy evaporation estimation model to calculate the estimated value of canopy evaporation in the area.

[0006] Preferably, the control function includes control functions corresponding to total radiation, water vapor pressure deficit, soil moisture content at 20cm depth, and wind speed. The control function corresponding to the total radiation is characterized by the quotient of the total radiation value and the sum of the total radiation value and a constant. The control function corresponding to the water vapor pressure deficit is represented by the product of a constant, the water vapor pressure deficit, and an exponential function. The base of the exponential function is a natural constant, and its exponent is the product of another constant and the water vapor pressure deficit. The control function corresponding to the soil moisture content at 20cm is characterized by a quadratic function, with the independent variable being the relative effective soil moisture content. The relative effective soil moisture content is calculated based on the soil moisture content at 20cm, the soil wilting moisture content, and the field capacity. The control function corresponding to the wind speed is represented by a cubic function, with the wind speed as its independent variable.

[0007] Preferably, the relative effective water content of the soil is determined in the following manner: When the soil moisture content at 20cm depth is less than or equal to the soil wilting moisture content, the relative effective soil moisture content is 0. When the soil moisture content at 20cm depth is greater than or equal to the field capacity, the relative effective soil moisture content is 1. When the soil moisture content at 20cm is between the field capacity and the wilting moisture content, the relative effective soil moisture content is the ratio of the difference between the soil moisture content at 20cm and the wilting moisture content to the difference between the field capacity and the wilting moisture content.

[0008] Preferably, the method for determining the maximum canopy transpiration rate per unit leaf area can be either method one or method two. Method 1 is to take the maximum value of the canopy transpiration rate per unit leaf area in the sampled area that is concurrent with the surface environmental factors as the maximum canopy transpiration rate per unit leaf area. Method 2 involves obtaining surface environmental factors in the validation sample area over a period of time as validation sample data, and acquiring corresponding canopy transpiration observation values. Then, multiple original maximum leaf area canopy transpiration rates to be screened are set. Different canopy transpiration estimation models are determined sequentially based on different original maximum leaf area canopy transpiration rates and validation sample data, and canopy transpiration prediction values ​​are calculated. For each canopy transpiration estimation model, its output canopy transpiration prediction values ​​are linearly fitted with the corresponding canopy transpiration observation values. The original maximum leaf area canopy transpiration rate corresponding to the canopy transpiration estimation model with a slope of 1 after fitting is taken as the final maximum leaf area canopy transpiration rate.

[0009] Preferably, the control function corresponding to the soil moisture content at 20cm depth also includes a dynamic correction step based on the soil moisture consumption rate during the hot and dry season, to characterize the adaptive response of shallow-rooted plants to the rapid decline in surface soil moisture. The specific steps are as follows: For sample data covering the change of soil moisture content from wet to dry during the hot and dry season, a time series of relative effective moisture content was established based on the soil moisture content at 20 cm in the sample data. The decrease value of relative effective moisture content at 20 cm in the soil during the active period of canopy transpiration during the day in the hot and dry season was extracted, and the average decrease value during this period was calculated. Compare the decrease in relative available water content at 20cm depth with the average decrease during the same period, and generate a water stress correction coefficient based on the comparison results. If the decrease in the relative effective water content of soil at 20cm depth is greater than or equal to the average decrease during that period, the water stress correction factor is 1. If the decrease is less than the average decrease, the water stress correction factor is determined based on the average decrease and the decrease on that day, and its calculation formula is as follows: ,in, This is the correction factor for water stress. This refers to the proportion of plant root biomass within the 0-20cm soil layer to the total plant root biomass. and The values ​​represent the average decrease and the decrease in relative available water content at a depth of 20 cm in the soil, respectively. The control function corresponding to the current soil moisture content at 20cm depth is calculated by summing the control function and the water stress correction coefficient to obtain the final control function corresponding to the soil moisture content at 20cm depth.

[0010] Preferably, step S5 further includes a step of correcting the output value of the canopy transpiration estimation model based on the canopy transpiration estimation model. The specific steps are as follows: Substitute the surface environmental factors from the sample data into the canopy evaporation estimation model to calculate a set of initial canopy evaporation prediction values, and subtract them from the corresponding measured canopy evaporation values ​​to obtain a set of prediction residuals; At least two environmental factors with known synergistic effects on canopy transpiration are selected, and partitioned according to their value ranges. The predicted residuals are classified by partition, the average residual value in each partition is calculated, and a residual correction table indexed by partition is constructed. When calculating canopy evaporation in the area to be evaluated, the zone to which it belongs is determined based on the input environmental factor values. The corresponding average residual value is extracted from the residual correction table as the correction amount. The correction amount is added to the canopy evaporation estimate calculated by the canopy evaporation estimation model to obtain the final canopy evaporation estimate and output it.

[0011] Compared with existing technologies, this invention provides a method for estimating canopy transpiration in shallow-rooted plants, which has the following beneficial effects: 1. This invention collects readily available surface environmental factors such as total radiation, water vapor pressure deficit, soil moisture content at 20cm depth, canopy leaf area index, and wind speed. It then uses the envelope method to extract the corresponding standardized canopy transpiration rate values ​​and fits control functions one by one. Finally, it constructs a canopy transpiration estimation model by multiplying the output values ​​of each control function, the leaf area index, and the canopy transpiration rate per unit leaf area. This effectively avoids the shortcomings of traditional Penman-Monteith-type models, which require a large number of difficult-to-obtain parameters such as crop coefficients and canopy impedance, significantly lowering the barrier to model use. Furthermore, considering the root distribution characteristics of shallow-rooted plants, soil moisture content at a depth of 20cm is incorporated, further improving the accuracy of canopy transpiration estimation.

[0012] 2. The present invention uses a rectangular hyperbola for total radiation, an exponential product function for water vapor pressure deficit, a quadratic function for soil moisture based on relative effective water content, and a cubic function for wind speed. It also introduces a dynamic correction step based on the soil moisture consumption rate. By calculating the average decrease in soil moisture content at 20 cm depth during the active transpiration period and comparing it with the decrease value, a water stress correction coefficient is generated to correct the soil moisture control function. This effectively characterizes the adaptive response of shallow-rooted plants to surface soil drought and further improves the estimation accuracy.

[0013] 3. This invention adds a correction step for the estimated canopy transpiration based on residual partitioning. It partitions environmental factors with synergistic effects, calculates the average predicted residual within each partition to construct a correction table, and uses this correction table to correct the model estimates. This effectively captures the interaction effects between environmental factors without increasing the complexity of the envelope method or relying on multidimensional envelope surfaces. Attached Figure Description

[0014] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of the method for estimating canopy transpiration in shallow-rooted plants according to the present invention. Detailed Implementation

[0015] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. This will allow for a full understanding of how the present application uses technical means to solve technical problems and achieve technical effects, and to facilitate its implementation.

[0016] Those skilled in the art will understand that all or part of the steps in the methods of the following embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0017] To address the issue that current vegetation evaporation models based on the Penman-Monteith model have numerous variables or parameters when estimating plant canopy transpiration, placing high demands on users, this embodiment provides a method for estimating canopy transpiration in shallow-rooted plants. By employing several easily measurable indicators and fitting the model, the response relationship between plant canopy transpiration and environmental factors is identified, and a model is constructed to estimate plant canopy transpiration. The specific steps include: S1. First, surface environmental factors and canopy transpiration rate per unit leaf area were collected in the sample area over a period of time. The maximum canopy transpiration rate per unit leaf area during the same period was obtained. The ratio of the canopy transpiration rate per unit leaf area to the maximum canopy transpiration rate per unit leaf area was used as the standardized canopy transpiration rate value. This value was combined with surface environmental factors to obtain sample data. The surface environmental factors included at least total radiation, vapor pressure deficit, soil moisture content at 20 cm depth, canopy leaf area index, and wind speed. In this embodiment, these five parameters were selected based on a systematic analysis of the canopy transpiration driving mechanism. These five factors represent three major driving forces affecting plant canopy transpiration: energy supply conditions, atmospheric evaporation demand, soil water supply capacity, and the transpiration area of ​​the vegetation itself. Total radiation was used to characterize energy supply conditions, vapor pressure deficit and wind speed were used to characterize atmospheric evaporation demand, soil moisture content at 20 cm depth was used to characterize soil water supply capacity, and canopy leaf area index was used to characterize the transpiration area of ​​the vegetation itself. Specifically, total radiation provides the energy source for the phase change of water in the stomata of plant leaves, and is the core driving force of transpiration; vapor pressure deficit characterizes the atmosphere's ability to absorb water vapor, directly affecting the water vapor concentration gradient between leaves and the atmosphere; soil moisture content at a depth of 20 cm directly reflects the moisture status of the main water absorption layer of shallow-rooted plants, and is a key variable for simulating the water response of shallow-rooted plants; wind speed affects the thickness of the leaf boundary layer and the rate of water vapor diffusion; and the canopy leaf area index directly determines the total transpiration area of ​​the entire canopy. For shallow-rooted plants, choosing a soil moisture content at a depth of 20 cm is of particular significance. The roots of shallow-rooted plants are mainly distributed in the shallow soil layer (0-30 cm below the surface), with a depth of 20 cm typically representing the area of ​​highest root density, thus sensitively reflecting changes in the actual available water for the plant. In addition, when selecting a period, it is best to select a period in which various environmental factors change significantly. The length of this period should cover the entire growing season of the study area or include at least several wet-dry cycles to ensure that the sample data can fully reflect the canopy transpiration response under various combinations of environmental conditions. It is generally recommended that this period be no less than 3 months, and the ideal period is the entire growing season with wet-dry cycles (such as May to October).

[0018] In this embodiment, the specific methods for obtaining sample data include, but are not limited to, plot monitoring or acquisition from public meteorological platforms. Total radiation, vapor pressure deficit, wind speed, and soil moisture content at 20cm depth can be obtained through ground-based meteorological stations or automatic environmental monitoring sensors. The canopy leaf area index (TDP) can be measured using a plant canopy analyzer. The average transpiration rate per unit leaf area can be obtained by measuring the trunk sap flow rate using a thermal diffusion runoff meter and then converting it to the average transpiration rate per tree. The average transpiration rate per tree is then divided by the average leaf area per tree to obtain the canopy transpiration rate per unit leaf area. The average leaf area per tree equals the average canopy area multiplied by the leaf area index. Specifically, a TDP probe is installed on the tree trunk. The TDP probe detects the thermal signal and transmits it to a data acquisition unit, which stores the signal and calculates the sap flow based on the TDP-measured data.

[0019] S2. After obtaining these sample data, it is necessary to further establish control functions for the influence of various environmental factors on canopy transpiration. In actual field environments, there are often complex covariation relationships among environmental factors. For example, high radiation is usually accompanied by high temperature and high vapor pressure deficit, while changes in soil moisture are relatively lagging. This multi-factor covariation means that when directly using all data points for multivariate statistical analysis, the independent effects of each environmental factor are often masked or confused, resulting in the fitted response relationship not accurately reflecting the true limiting effect of a single factor on transpiration. Therefore, in this embodiment, for different variation ranges of multiple surface environmental factors in the sample data, the envelope method is used to extract the corresponding standardized canopy transpiration rate values ​​as the envelope data points for each surface environmental factor. The core idea of ​​the envelope method is that under any given environmental factor condition, the standardized canopy transpiration rate value represents the maximum transpiration capacity that plants can achieve under that environmental factor condition without significant limitation by other factors. By extracting the standardized canopy transpiration rate values ​​of each environmental factor within its variation range, the synergistic limiting effects of other environmental factors can be removed, thereby obtaining the independent maximum control response of each environmental factor to transpiration.

[0020] S3. After obtaining the envelope data points of each environmental factor, the relationship between each environmental factor and the canopy transpiration rate per unit leaf area has been presented in the form of discrete data points. However, discrete envelope data points are difficult to directly apply to prediction scenarios where independent variables change continuously. In order to calculate the corresponding control coefficient for any given environmental factor value, it is necessary to establish a continuous mathematical function, i.e., a control function, for each environmental factor, and make its curve shape fit the distribution trend of the envelope data points as closely as possible. Specifically, in this embodiment, multiple preset control functions that characterize the degree of influence of surface environmental factors on canopy transpiration are fitted one by one by the envelope data points to obtain the parameters in each control function and the control function after the parameters are determined. Thus, the control functions corresponding to the four surface environmental factors of total radiation, water vapor pressure deficit, soil moisture content at 20 cm, and wind speed are finally obtained. When fitting each control function, the nonlinear least squares method (Levenberg-Marquardt algorithm) is used to perform curve fitting on the envelope data points. During the fitting process, the objective function is to minimize the sum of squared residuals. The upper limit of the number of iterations can be set to 500, and the convergence tolerance can be set to 1×10⁻⁶. -6 After fitting, the coefficient of determination (R²) and root mean square error (RMSE) are calculated as evaluation indicators of the model's fit. It is permissible that the standard error of each parameter estimate is less than 20% of the parameter estimate; otherwise, the rationality of the outer envelope data points needs to be re-examined or the sample size increased. The control function corresponding to the total radiation is characterized by the quotient of the total radiation value and the sum of the total radiation value and a constant. Its expression is:

[0021] The control function corresponding to the vapor pressure deficit is characterized by the product of a constant, the vapor pressure deficit, and an exponential function. The base of this exponential function is a natural constant, and its exponent is the product of another constant and the vapor pressure deficit. Its expression is as follows:

[0022] The control function corresponding to the soil moisture content at 20cm depth is characterized by a quadratic function, with the independent variable being the relative available soil moisture content. The relative available soil moisture content is calculated based on the soil moisture content at 20cm depth, the soil wilting moisture content, and the field capacity. The calculation formula is as follows:

[0023] The relative effective water content of the soil is determined as follows: When the soil moisture content at 20cm depth is less than or equal to the wilting moisture content, the relative effective soil moisture content is 0; when the soil moisture content at 20cm depth is greater than or equal to the field capacity, the relative effective soil moisture content is 1; when the soil moisture content at 20cm depth is between the field capacity and the wilting moisture content, the relative effective soil moisture content is the ratio of the difference between the soil moisture content at 20cm depth and the wilting moisture content to the difference between the field capacity and the wilting moisture content, expressed as:

[0024] The control function corresponding to wind speed is represented by a cubic function, with wind speed as the independent variable, and its expression is:

[0025] In the above formula, , , , These are the control functions corresponding to total radiation, vapor pressure deficit, soil moisture content at 20cm depth, and wind speed, respectively. For total radiation, Due to water vapor pressure deficit This refers to the soil moisture content at a depth of 20cm. This refers to the relative available water content of the soil. Wind speed; , , , , , , , , , is the parameter to be determined; e is the natural constant. and These are soil wilting moisture content and field holding capacity, respectively. This refers to the lower limit of soil moisture content at which plants permanently wilt. When soil moisture content drops to or below the wilting point, the soil matrix potential is too low, and the water in the soil is firmly adsorbed by soil particles. Plant roots cannot overcome this potential to absorb water from the soil. In this situation, from the perspective of plant water absorption, there is no effective water in the soil. Therefore, [the following is a continuation of the previous sentence, but the context is unclear]. A value of 0 indicates that there is no available water in the soil for plant absorption, and the soil moisture's limitation on transpiration is at its maximum. Field holding capacity. This refers to the maximum amount of water that soil can retain after gravity drains the water. When the soil moisture content reaches or exceeds field capacity, the soil water supply is sufficient, and plant roots are not under water stress when absorbing water. At this time, A value of 1 indicates that soil moisture has no limitation on transpiration. When the soil moisture content at 20cm depth is between field capacity and wilting water content, the ratio of the current soil moisture content to wilting water content (i.e., the current available water content) is divided by the maximum possible available water content (i.e., the difference between field capacity and wilting water content) to obtain the ratio of the soil's relative available water content. This ratio continuously ranges from 0 to 1, representing the proportion of plant-available water to the maximum available water, thus quantifying the influence of soil moisture conditions on plant transpiration. Soil wilting water content and field capacity can be obtained in two ways: First, for areas where soil physical property surveys have been conducted, relevant soil databases or literature can be consulted directly; second, for areas lacking relevant data, undisturbed soil samples can be collected and measured using indoor testing methods such as pressure membrane testing or centrifugation. Specific measurement procedures can refer to relevant technical regulations such as "Methods for Determination of Soil Moisture Parameters" (LY / T 1217-1999).

[0026] The control function corresponding to the soil moisture content at 20cm depth is determined based on the relative effective moisture content at 20cm depth, neglecting the adaptive response of plants to soil moisture changes. This deficiency is particularly pronounced for shallow-rooted plants, whose roots are mainly distributed in the top 0-30cm of soil. While their root density is high, their shallow distribution makes them extremely sensitive to topsoil moisture deficit. When topsoil moisture gradually decreases, even if the absolute soil moisture content has not yet reached the wilting point, plants may reduce the water absorption weight of their surface roots. Therefore, this embodiment also includes a dynamic correction step based on the soil moisture consumption rate during hot, dry seasons to characterize the adaptive response of shallow-rooted plants to the decrease in topsoil moisture. The specific steps are as follows: This embodiment uses sample data covering the change in soil moisture content from wet to dry during the hot and dry season. Based on the soil moisture content at 20cm depth in the sample data, a time series of relative available water content is established. Then, the decrease in relative available water content at 20cm depth is extracted during the daily active canopy transpiration periods in the hot and dry season. Active canopy transpiration periods are typically the peak transpiration period during the daytime, for example, between 9:00 AM and 5:00 PM local time on sunny days. During this period, solar radiation is strong, temperatures are high, plant stomata are open, and transpiration is vigorous, making it the main period when soil moisture changes significantly due to plant water absorption. The method for extracting the decrease in relative available water content at 20cm depth is as follows: subtract the relative available water content at 20cm depth at the end of the period from the relative available water content at the beginning of the period. If the result is positive, it is recorded as the decrease for that day; if the result is negative or zero (e.g., soil moisture content increases instead of decreasing due to rainfall or irrigation), it is not included or recorded as zero. After obtaining the daily decrease values ​​for this period, calculate the arithmetic mean of the daily decrease values. The average decrease value during this period is obtained. This value reflects the rate of soil moisture loss and is more closely related to the difficulty of water absorption by plant roots in shallow soil. For shallow-rooted plants, the slow decrease in surface soil moisture on sunny days means that the soil moisture is low, the evaporation of the soil surface is limited by the soil water supply, the effective water near the shallow roots is gradually depleting, and the plant may have begun to feel water stress.

[0027] The decrease in relative available water content at 20cm depth is compared with the average decrease over the same period, and a water stress correction coefficient is generated based on the comparison results. If the decrease is greater than or equal to the average decrease, the water stress correction coefficient is 1; if the decrease is less than the average decrease, the water stress correction coefficient is determined based on the difference between the average decrease and the decrease on that day, and the calculation formula is as follows: , in, This is the correction factor for water stress. The proportion of plant root biomass in the 0-20cm soil layer to the total plant root biomass can be determined through preliminary experiments or literature review. , These represent the average decrease and the decrease in relative available water content at 20cm soil depth, respectively. Finally, the control function corresponding to the current soil water content at 20cm depth is calculated, along with the water stress correction coefficient, to obtain the final control function corresponding to the soil water content at 20cm depth.

[0028] S4. The canopy transpiration estimation model is defined as the product of the output values ​​of each control function, the canopy leaf area index (LAI), and the maximum canopy transpiration rate per unit leaf area. The maximum canopy transpiration rate per unit leaf area represents the upper limit of the transpiration rate per unit leaf area that plants can achieve under optimal environmental conditions, i.e., sufficient light, abundant water, moderate atmospheric evaporation demand, and suitable wind speed. It is the scale benchmark value in the model and determines the magnitude of the entire model output value. By multiplying several parameters, it reflects that transpiration is the result of the synergistic effect of energy drive, atmospheric demand, and soil water supply. The deficiency of any one factor will limit transpiration. The canopy leaf area index (LAI) is also multiplied to realize the conversion from the "unit leaf area" scale to the "unit canopy area" scale, so as to obtain the total transpiration of the entire canopy. This processing method allows the model output value to be directly compared and verified with the unit canopy transpiration calculated based on plot observations (such as the trunk sap flow method). For vegetation types where LAI changes are not significant or for application scenarios with short estimation periods, such as when the canopy structure is relatively stable during the estimation period, LAI can be considered a constant and its changes can be disregarded.

[0029] The canopy transpiration estimation model in this embodiment involves several parameters in four control functions. Compared with the large number of physical parameters and canopy resistance parameters required by Penman-Monteith-type models, the structure of the model of this invention is significantly simplified. At the same time, each parameter has a clear physiological and ecological significance, which makes it easy for those skilled in the art to adjust the parameters and interpret the model according to the actual situation.

[0030] S5. Obtain multiple surface environmental factors of the area to be evaluated, and substitute them into the corresponding control function and canopy evaporation estimation model to calculate the estimated value of canopy evaporation in the area.

[0031] In addition, in step S1 of this embodiment, the method for determining the maximum canopy transpiration rate per unit leaf area can be either method one or method two. Method 1 is to take the maximum value of the canopy transpiration rate per unit leaf area in the sample area that is concurrent with the surface environmental factors as the maximum canopy transpiration rate per unit leaf area.

[0032] Method Two involves obtaining surface environmental factors over a period of time in the validation sample area as validation sample data, and acquiring corresponding canopy transpiration observations. Then, multiple raw maximum leaf area (MLA) canopy transpiration rates are set for selection. Different canopy transpiration estimation models are determined sequentially based on different raw MLA canopy transpiration rates and validation sample data, and canopy transpiration prediction values ​​are calculated. For each canopy transpiration estimation model, its output canopy transpiration prediction values ​​are linearly fitted with the corresponding canopy transpiration observation values. The raw MLA canopy transpiration rate corresponding to the canopy transpiration estimation model with a slope of 1 after fitting is taken as the final MLA canopy transpiration rate. Since the maximum canopy transpiration rate only occurs when multiple influencing factors are simultaneously optimal throughout the year, the sample data needs to cover a long monitoring period to detect this situation. The core idea of ​​Method Two is to infer the optimal parameters from the data. This method treats the MLA canopy transpiration rate as a global parameter to be calibrated, and through iterative adjustments, the model's predicted values ​​achieve the best statistical match with the measured values.

[0033] The canopy transpiration estimation model in this embodiment is a product of multiple parameters. While this structure reflects the law of limiting factors, it essentially assumes that the effects of each environmental factor are independent. In natural environments, significant interaction effects may exist between certain environmental factors. For example, under high radiation conditions, plants may be more sensitive to soil moisture deficit because high radiation accelerates leaf transpiration demand, leading to greater pressure on root water supply. Conversely, under low radiation conditions, even with low soil moisture content, plant transpiration may not decrease as significantly as the model predicts. Such interaction effects are difficult to capture directly in the single-factor envelope method because the design goal of the envelope method is to isolate the independent maximum effect of a single factor from multi-factor covariance. Therefore, this embodiment further proposes a canopy transpiration estimation correction method based on residual partitioning. This method corrects the interaction effects between environmental factors without increasing the complexity of the envelope method or relying on multi-dimensional envelope surfaces. The specific steps are as follows: By substituting the surface environmental factors from the sample data into the canopy evapotranspiration estimation model, a set of initial canopy evapotranspiration prediction values ​​are calculated. These prediction residuals are then subtracted from the corresponding measured canopy evapotranspiration values ​​to obtain a set of prediction residuals. The prediction residuals contain all the information not explained by the canopy evapotranspiration estimation model, including the interaction effect signals between factors. Through subsequent systematic analysis of the residuals, these interaction effect signals can be extracted and used to correct the output of the canopy evapotranspiration estimation model.

[0034] Next, at least two environmental factors with known synergistic effects on canopy transpiration were selected, partitioned according to their value ranges, and the predicted residuals were categorized by partition. The average residual within each partition was calculated, and a residual correction table indexed by partition was constructed. Specifically, in this embodiment, based on existing research in the field of plant physiological ecology, combinations of factors with significant synergistic effects on canopy transpiration include, but are not limited to: total radiation and soil moisture content at 20 cm depth, and vapor pressure deficit and soil moisture content at 20 cm depth. Considering that under high radiation conditions, plant transpiration demand is vigorous, and dependence on soil moisture supply is enhanced, the limiting effect of soil moisture deficit on transpiration is more significant under high radiation background; under low radiation conditions, transpiration demand is inherently lower, and even if soil moisture is insufficient, the magnitude of transpiration attenuation is relatively limited. This embodiment selects total radiation and soil moisture content at 20cm depth as a pair of environmental factors for partitioning. The numerical range of total radiation is divided into M intervals, and the numerical range of soil moisture content at 20cm depth is divided into N intervals. The combination of these two interval divisions forms M×N partitions. The intervals can be divided using either equal spacing or equal frequency. Equal spacing divides the data evenly across the numerical range, suitable for cases where the data is relatively evenly distributed. Equal frequency divides the data evenly based on the number of data points, ensuring each interval contains approximately the same number of data points, suitable for cases where the data distribution is uneven. In this embodiment, equal frequency division is used, with M and N taking values ​​between 5 and 8, ensuring that the number of data points in each partition meets the requirements for statistical calculation. Interaction effects vary under different combinations of environmental factors, and partitioning can localize these effects in the environmental factor space. For example, in a high-radiation × low-soil-moisture partition, the interaction effect may manifest as a systematic overestimation of the model (negative residuals); in a low-radiation × high-soil-moisture partition, the interaction effect may be negligible (residuals close to zero). By analyzing data separately by region, we can precisely depict the differentiated performance of interaction effects under different environmental conditions.

[0035] Finally, when calculating canopy evapotranspiration in the area to be evaluated, the zone to which the area belongs is determined based on the input environmental factor values. The corresponding average residual value is extracted from the residual correction table as the correction amount. This correction amount is added to the canopy evapotranspiration estimate calculated by the canopy evapotranspiration estimation model to obtain the final canopy evapotranspiration estimate, which is then output. The envelope method excels at extracting the single-factor independent maximum control effect, but it cannot capture multi-factor interaction effects. If the envelope method is extended from a single dimension to multiple dimensions to capture interaction effects, it will face substantial difficulties such as exponentially increasing data requirements and sparse data in the multi-dimensional envelope surface. Therefore, this embodiment first uses the envelope method to construct a single-factor model to obtain the main simulation, and then corrects the interaction effect bias by performing systematic zoning statistics on the prediction residuals. Without increasing data requirements or changing the core algorithm structure of the envelope method, it effectively improves the prediction accuracy of the model under conditions of multiple environmental factor coupling.

[0036] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for estimating canopy transpiration of shallow-rooted plants, characterized by, Includes the following steps: S1. Collect surface environmental factors and canopy transpiration rate per unit leaf area in the sample area over a period of time, and obtain the maximum canopy transpiration rate per unit leaf area during the same period. Use the ratio of the canopy transpiration rate per unit leaf area to the maximum canopy transpiration rate per unit leaf area as the standardized canopy transpiration rate value, and combine it with surface environmental factors to obtain sample data. The surface environmental factors include at least total radiation, water vapor pressure deficit, soil moisture content at 20 cm depth, canopy leaf area index, and wind speed. S2. For different variation ranges of multiple surface environmental factors in the sample data, the corresponding standardized canopy evaporation rate values ​​are extracted using the envelope method, which are used as the envelope data points of each surface environmental factor. S3. By fitting multiple preset control functions to characterize the influence of surface environmental factors on canopy transpiration one by one through the outer envelope data points, the parameters in each control function and the control function after the parameters are determined are obtained. S4. Define the canopy transpiration estimation model as the product of the output values ​​of each control function, the canopy leaf area index (LAI), and the maximum canopy transpiration rate per unit leaf area. S5. Obtain multiple surface environmental factors of the area to be evaluated, and substitute them into the corresponding control function and canopy evaporation estimation model to calculate the estimated value of canopy evaporation in the area.

2. The shallow-rooted plant canopy transpiration estimation method according to claim 1, characterized in that, The control functions include control functions corresponding to total radiation, water vapor pressure deficit, soil moisture content at 20cm depth, and wind speed. The control function corresponding to the total radiation is characterized by the quotient of the total radiation value and the sum of the total radiation value and a constant. The control function corresponding to the water vapor pressure deficit is represented by the product of a constant, the water vapor pressure deficit, and an exponential function. The base of the exponential function is a natural constant, and its exponent is the product of another constant and the water vapor pressure deficit. The control function corresponding to the soil moisture content at 20cm is characterized by a quadratic function, with the independent variable being the relative effective soil moisture content. The relative effective soil moisture content is calculated based on the soil moisture content at 20cm, the soil wilting moisture content, and the field capacity. The control function corresponding to the wind speed is represented by a cubic function, with the wind speed as its independent variable.

3. The method of claim 1, wherein the method further comprises: The relative effective water content of the soil is determined as follows: When the soil moisture content at 20cm depth is less than or equal to the soil wilting moisture content, the relative effective soil moisture content is 0. When the soil moisture content at 20cm depth is greater than or equal to the field capacity, the relative effective soil moisture content is 1. When the soil moisture content at 20cm is between the field capacity and the wilting moisture content, the relative effective soil moisture content is the ratio of the difference between the soil moisture content at 20cm and the wilting moisture content to the difference between the field capacity and the wilting moisture content.

4. The method of claim 1, wherein the method further comprises: The method for determining the maximum canopy transpiration rate per unit leaf area can be either method one or method two. Method 1 is to take the maximum value of the canopy transpiration rate per unit leaf area in the sampled area that is concurrent with the surface environmental factors as the maximum canopy transpiration rate per unit leaf area. Method 2 involves obtaining surface environmental factors in the validation sample area over a period of time as validation sample data, and acquiring corresponding canopy transpiration observation values. Then, multiple original maximum leaf area canopy transpiration rates to be screened are set. Different canopy transpiration estimation models are determined sequentially based on different original maximum leaf area canopy transpiration rates and validation sample data, and canopy transpiration prediction values ​​are calculated. For each canopy transpiration estimation model, its output canopy transpiration prediction values ​​are linearly fitted with the corresponding canopy transpiration observation values. The original maximum leaf area canopy transpiration rate corresponding to the canopy transpiration estimation model with a slope of 1 after fitting is taken as the maximum leaf area canopy transpiration rate.

5. The method of claim 1, wherein the method further comprises: determining a canopy temperature of the shallow-rooted plant; and determining a canopy resistance of the shallow-rooted plant. The control function corresponding to the soil moisture content at 20cm depth also includes a dynamic correction step based on the soil moisture consumption rate during the high-temperature and dry season, to characterize the adaptive response of shallow-rooted plants to the rapid decline in surface soil moisture. The specific steps are as follows: For sample data covering the change of soil moisture content from wet to dry during the hot and dry season, a time series of relative effective moisture content was established based on the soil moisture content at 20 cm in the sample data. The decrease value of relative effective moisture content at 20 cm in the soil during the active period of canopy transpiration during the day in the hot and dry season was extracted, and the average decrease value during this period was calculated. Compare the decrease in relative available water content at 20cm depth with the average decrease during the same period, and generate a water stress correction coefficient based on the comparison results. If the decrease in the relative effective water content of soil at 20cm depth is greater than or equal to the average decrease during that period, the water stress correction factor is 1. If the decrease is less than the average decrease, the water stress correction factor is determined based on the average decrease and the decrease on that day. The control function corresponding to the current soil moisture content at 20cm depth is calculated by summing the control function and the water stress correction coefficient to obtain the final control function corresponding to the soil moisture content at 20cm depth.

6. The shallow-rooted plant canopy transpiration estimation method according to claim 1, characterized by, In step S5, based on the canopy transpiration estimation model, the step further includes correcting the output value of the canopy transpiration estimation model. The specific steps are as follows: Substitute the surface environmental factors from the sample data into the canopy evaporation estimation model to calculate a set of initial canopy evaporation prediction values, and subtract them from the corresponding measured canopy evaporation values ​​to obtain a set of prediction residuals; At least two environmental factors with known synergistic effects on canopy transpiration are selected, and partitioned according to their value ranges. The predicted residuals are classified by partition, the average residual value in each partition is calculated, and a residual correction table indexed by partition is constructed. When calculating canopy evaporation in the area to be evaluated, the zone to which it belongs is determined based on the input environmental factor values. The corresponding average residual value is extracted from the residual correction table as the correction amount. The correction amount is added to the canopy evaporation estimate calculated by the canopy evaporation estimation model to obtain the final canopy evaporation estimate and output it.

Images