Evapotranspiration component prediction method, system, equipment and medium

Abstract

The invention relates to an evapotranspiration component prediction method, system and device and a medium. The method comprises the following steps: acquiring four types of meteorological elements, including net radiation, air temperature, wind speed and relative humidity, influencing evapotranspiration components in a target area, and screening core meteorological elements in combination with area vegetation coverage and soil moisture retention characteristics; constructing a core meteorological element historical simulation sequence, comparing the core meteorological element historical simulation sequence with a corresponding historical observation sequence to calculate a Pearson's correlation coefficient, determining an initial fusion prediction value based on the coefficient, applying an emergence constraint, and generating a core meteorological element future prediction sequence; inputting the prediction sequence into a hydrological cycle model for simulation and correction to obtain a future soil moisture prediction result of the target area; and jointly inputting the future prediction sequence of the core meteorological elements and a future soil moisture prediction result into the double-layer evapotranspiration model, outputting and correcting a primary evapotranspiration component sequence, and finally obtaining an adjusted evapotranspiration component sequence. The method can improve prediction precision and stability.

Description

Technical Field

[0001] This invention belongs to the field of hydrological and meteorological forecasting technology, and in particular relates to a method, system, equipment and medium for predicting evapotranspiration components. Background Technology

[0002] Evapotranspiration components (including vegetation transpiration and soil evaporation) are a core component of the hydrological cycle and energy balance. Accurate prediction of these components plays a crucial supporting role in fields such as agricultural irrigation optimization, water resource allocation, and ecological environment assessment. Existing technologies mostly employ traditional evapotranspiration models (such as the Penman-Monteith model and single / double-layer empirical models) combined with meteorological observation data for calculation, but they have significant technical limitations: On the one hand, the prediction of core meteorological elements relies on the output of a single climate model and does not fully consider the fusion of multi-source data and regional characteristic constraints, resulting in high uncertainty of future meteorological inputs; on the other hand, soil moisture, as a key limiting factor for evapotranspiration, often ignores the coupling effect of underlying surface properties and topographic features in its prediction process, leading to a large deviation from the actual surface moisture distribution; at the same time, existing evapotranspiration models often simplify the energy interaction mechanism between upper and lower layers, lacking dynamic feedback and iterative convergence optimization design for soil moisture, resulting in insufficient accuracy and poor stability in component prediction, making it difficult to meet the needs of refined applications under complex underlying surface conditions. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, system, device, and medium for predicting evaporative emission components that can improve the accuracy and stability of evaporative emission component prediction, in order to address the above-mentioned technical problems.

[0004] In a first aspect, this application provides a method for predicting evaporation components, including:

[0005] Four types of meteorological elements affecting evapotranspiration components in the target area were obtained: net radiation, air temperature, wind speed, and relative humidity. Core meteorological elements were selected from these four types of meteorological elements based on regional vegetation cover and soil moisture retention characteristics.

[0006] The Pearson correlation coefficient between the historical simulation sequence of core meteorological elements and the corresponding historical observation sequence of the target area is calculated to determine the initial fusion prediction value. Preset emergence constraints are applied to the initial fusion prediction value to generate the future prediction sequence of core meteorological elements.

[0007] Based on the future forecast sequence of core meteorological elements, a preset hydrological cycle model is input, and the future soil moisture forecast results of the target area are obtained through simulation and correction of the hydrological cycle model.

[0008] The future forecast sequence of core meteorological elements and the future soil moisture forecast result are input into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence. The preliminary evapotranspiration component sequence is then corrected to obtain the adjusted evapotranspiration component sequence.

[0009] In one embodiment, the Pearson correlation coefficient between the historical simulation sequence of core meteorological elements and the corresponding historical observation sequence of the target area is calculated to determine the initial fusion prediction value. A preset emergence constraint is then applied to the initial fusion prediction value to generate the future prediction sequence of the core meteorological elements, including:

[0010] Multiple climate models were selected from the CMIP6 climate model set, and historical simulation sequences and historical observation sequences of core meteorological elements in the target area were obtained from the output of each selected climate model.

[0011] Calculate the Pearson correlation coefficient between the historical simulation sequence and the historical observation sequence for each selected climate model.

[0012] Based on the Pearson correlation coefficient, the weight coefficients corresponding to each selected climate model are determined. After normalizing the weight coefficients, they are weighted and summed with each core meteorological element to obtain the initial fusion prediction value of the core meteorological element.

[0013] Pre-defined emergence constraints are applied to the initial fusion prediction values ​​to obtain the core meteorological element fusion prediction values ​​after constraint adjustment; the emergence constraints are pre-set based on the physical laws of the climate system in the target area and the relevant constraint requirements for evapotranspiration component prediction.

[0014] The future forecast sequence of core meteorological elements is generated based on the fusion forecast values ​​of core meteorological elements.

[0015] The system verifies whether the future forecast sequence meets the preset physical consistency threshold. If it does not, the weight coefficients of each selected climate model are readjusted based on the verification results to generate a future forecast sequence of core meteorological elements that meets the physical consistency threshold. The physical consistency threshold is preset based on the physical attributes of the meteorological elements themselves and the regional climate characteristics.

[0016] In one embodiment, a preset hydrological cycle model is input based on the future forecast sequence of core meteorological elements. The future soil moisture forecast result for the target area is obtained through simulation and correction adjustments using the hydrological cycle model, including:

[0017] The future prediction sequence of core meteorological elements is input into a hydrological cycle model coupled with the characteristic parameters of the underlying surface of the target area for simulation, and the preliminary prediction results of soil moisture in the target area are obtained. The characteristic parameters of the underlying surface include surface cover parameters, soil physicochemical parameters and hydrological correlation parameters.

[0018] Acquire key terrain feature data, including slope, aspect, and elevation gradient of the target area, and perform standardized preprocessing on the key terrain feature data to obtain terrain input data.

[0019] Based on the preliminary prediction results and topographic input data, the soil moisture distribution correction value is obtained by quantitative calculation through the topographic-soil moisture response relationship model.

[0020] The soil moisture distribution correction value is checked to see if it exceeds the preset reasonable threshold. If it does, the preliminary prediction result is dynamically corrected and adjusted according to the degree of deviation between the soil moisture distribution correction value and the reasonable threshold to obtain the future soil moisture prediction result.

[0021] In one embodiment, the soil moisture distribution correction value is calculated using the following formula:

[0022]

[0023] in, This indicates the correction value for soil moisture distribution. This indicates the preliminary prediction results of soil moisture. Indicates the terrain-underlying surface adaptation coefficient. , , , This represents the weighting coefficient, which is set based on the dominant type of the underlying surface in the region. , Indicates the average vegetation cover of the region. , Indicates soil porosity. , This indicates the percentage of impermeable area. The topographic sensitivity coefficient is calculated based on nearly 10 years of topographic-soil moisture observation data for the target area. Indicates the standardized terrain composite index, , , , This represents the weighting coefficient, which is set based on the dominant factors of regional topography. , Indicates slope, Indicates the maximum slope of the area. , Indicates slope direction. , Indicates altitude, This indicates the maximum and minimum elevation values ​​of a region.

[0024] In one embodiment, the future forecast sequence of core meteorological elements and the future soil moisture forecast result are input into a two-layer evapotranspiration model to output a preliminary evapotranspiration component sequence. The preliminary evapotranspiration component sequence is then corrected to obtain an adjusted evapotranspiration component sequence, including:

[0025] The future forecast sequences of core meteorological elements and the future soil moisture forecast results are input into the two-layer evapotranspiration model, and the initial energy balance parameters of the upper vegetation canopy and the lower soil surface are output. The energy balance parameters include the radiation balance coefficient, the sensible heat flux coefficient, and the initial threshold of the latent heat flux. The two-layer evapotranspiration model is an improved Shuttleworth-Wallace model.

[0026] Based on the energy balance parameters, the Newton-Raphson iterative solution method is used to iteratively calculate the energy balance equations of the upper and lower layers, and obtain a preliminary evapotranspiration component sequence including time series data of vegetation transpiration, soil evaporation and total evapotranspiration; the energy balance equations include the balance relationships of radiation balance, sensible heat flux and latent heat flux.

[0027] Historical evapotranspiration composition data of the target area were selected, and the preliminary evapotranspiration composition sequence was compared with the measured data time by time to calculate the prediction accuracy index, including the coefficient of determination and root mean square error.

[0028] If the prediction accuracy index is lower than the preset accuracy threshold, an uncertainty adjustment factor is obtained based on the error characteristics of the future prediction sequence of core meteorological elements and the future soil moisture prediction results; the accuracy threshold is preset based on the regional evapotranspiration prediction application requirements.

[0029] An uncertainty adjustment factor was incorporated into the time-by-time data of the preliminary evapotranspiration component sequence using a weighted correction method to dynamically correct the sequence, resulting in the adjusted evapotranspiration component sequence.

[0030] In one embodiment, the preliminary evaporation component sequence is represented by the following formula:

[0031]

[0032] in, Indicates the first Upper-layer transpiration in the next iteration Indicates the first Evaporation rate in the next iteration of the lower layer Indicates the first Total evapotranspiration in each iteration , Indicates the radiation balance coefficient between the upper and lower layers. , , , Indicates the sensible heat flux coefficients of the upper and lower layers. , , , This indicates the net radiation between the upper and lower levels. , Indicates soil moisture limiting factor, , Indicates field holding capacity. Indicates the moisture content during wilting. This represents the predicted future soil moisture value for the target area. This represents the interlayer energy coupling coefficient. , This represents the reference evapotranspiration, calculated based on the core meteorological forecast sequence. Represents the iterative correction coefficient. , Indicates the number of iterations. The iteration termination condition is .

[0033] Secondly, this application also provides a system for predicting evapotranspiration components, the system comprising:

[0034] The meteorological element screening module is used to obtain four types of meteorological elements that affect the evapotranspiration components of the target area: net radiation, air temperature, wind speed, and relative humidity. Core meteorological elements are selected from these four types of meteorological elements based on the regional vegetation coverage and soil moisture retention characteristics.

[0035] The future weather forecast module is used to calculate the Pearson correlation coefficient between the historical simulation sequence of core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion forecast value, and to apply preset emergence constraints to the initial fusion forecast value to generate the future forecast sequence of core meteorological elements.

[0036] The soil moisture prediction module is used to input a preset hydrological cycle model based on the future prediction sequence of core meteorological elements. The model is then used to simulate and correct the soil moisture prediction results for the target area.

[0037] The evapotranspiration adjustment module is used to input the future forecast sequence of core meteorological elements and the future soil moisture forecast results into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence, and to correct the preliminary evapotranspiration component sequence to obtain the adjusted evapotranspiration component sequence.

[0038] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0039] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method.

[0040] The aforementioned method, system, computer equipment, and storage medium for predicting evapotranspiration components acquire four meteorological elements affecting evapotranspiration components in a target area: net radiation, air temperature, wind speed, and relative humidity. Combining the vegetation cover and soil moisture retention characteristics of the target area, core meteorological elements are selected from these four elements. Based on the selected core meteorological elements, a historical simulation sequence is constructed. This historical simulation sequence is compared with the historical observation sequences of the corresponding core meteorological elements in the target area, and the Pearson correlation coefficient between the two is calculated. Based on this correlation coefficient, an initial fusion prediction value for the core meteorological elements is determined. Subsequently, a preset emergence constraint is applied to the initial fusion prediction value to generate a future prediction sequence for the core meteorological elements. The future prediction sequence of the core meteorological elements is input into a preset hydrological cycle model. The hydrological cycle model simulates and obtains a preliminary prediction result for soil moisture in the target area. After correcting and adjusting the preliminary prediction result, the future soil moisture prediction result for the target area is obtained. The future prediction sequence of the core meteorological elements and the aforementioned future soil moisture prediction result are used as joint driving inputs and imported into a two-layer evapotranspiration model. The model calculates and outputs a preliminary evapotranspiration component sequence. The preliminary evapotranspiration component sequence is then corrected to obtain the final adjusted evapotranspiration component sequence. This method achieves accurate prediction of evapotranspiration components through multi-stage data coupling and layer-by-layer optimization design. First, core meteorological elements are selected based on regional vegetation cover and soil moisture retention characteristics to ensure the correlation between input data and evapotranspiration components, eliminate redundant information, and improve the efficiency and relevance of subsequent prediction stages. Second, Pearson correlation coefficients are used to construct initial fusion prediction values ​​and apply emergence constraints to reduce the uncertainty of future predictions of core meteorological elements, improving the spatiotemporal matching and reliability of the prediction sequence. Third, the future prediction sequence of core meteorological elements drives the hydrological cycle model and completes correction adjustments, ensuring that future soil moisture prediction results align with the actual surface conditions of the target area, providing accurate water-limiting factor inputs for evapotranspiration component prediction. Fourth, the core meteorological and soil moisture prediction results jointly drive a two-layer evapotranspiration model, combined with preliminary sequence correction, fully coupling the comprehensive impact of meteorology and soil moisture on evapotranspiration components, compensating for the shortcomings of traditional models that neglect inter-layer interaction and dynamic feedback, ultimately improving the accuracy and stability of evapotranspiration component prediction, and providing reliable data support for scenarios such as water resource allocation, agricultural irrigation optimization, and ecological environment assessment. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1A flowchart of a method for predicting evapotranspiration components provided in an embodiment of the present invention;

[0043] Figure 2 This is a structural block diagram of an evapotranspiration component prediction system provided in an embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] In one embodiment, such as Figure 1 As shown, this application provides a method for predicting evapotranspiration components, which may include the following steps:

[0046] Step S101: Obtain four types of meteorological elements affecting evapotranspiration components in the target area: net radiation, air temperature, wind speed, and relative humidity. Select core meteorological elements from the four types of meteorological elements based on the area's vegetation cover and soil moisture retention characteristics.

[0047] Specifically, four meteorological elements affecting evapotranspiration components in the target area are acquired: net radiation, air temperature, wind speed, and relative humidity. The data sources for these four meteorological elements include historical simulation data from the CMIP6 multi-climate model and concurrent measured meteorological data from the target area. By combining vegetation cover data (such as vegetation type and quantitative indicators of cover density) and soil moisture retention characteristics data (such as quantitative parameters of soil texture, porosity, and field capacity) in the target area, the correlation strength between the four meteorological elements and vegetation transpiration and soil evaporation is analyzed. Elements that have a significant impact on evapotranspiration components are selected as core meteorological elements, while redundant data with weak correlations are eliminated.

[0048] Step S102: Calculate the Pearson correlation coefficient between the historical simulation sequence of the core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion prediction value, and apply a preset emergence constraint to the initial fusion prediction value to generate the future prediction sequence of the core meteorological elements.

[0049] Based on the core meteorological elements obtained through screening, historical simulation sequences of core meteorological elements output by the CMIP6 multi-climate model are extracted, and historical observation sequences of core meteorological elements for the corresponding time period in the target area are collected. The Pearson correlation coefficient between the historical simulation sequence and the historical observation sequence of core meteorological elements under each climate model is calculated. The historical simulation sequences of the multi-model are weighted and fused using this correlation coefficient as the weight to obtain the initial fused prediction value of the core meteorological elements. Based on the inherent laws of the climate system in the target area, preset emergence constraints (such as the consistency constraints of the spatiotemporal evolution of meteorological elements, extreme value threshold constraints, etc.) are set and applied to the initial fused prediction value. Predictive data that exceed the reasonable physical range are eliminated to generate future prediction sequences of core meteorological elements with higher accuracy and lower uncertainty.

[0050] Step S103: Based on the future prediction sequence of core meteorological elements, input the preset hydrological cycle model, and obtain the future soil moisture prediction results of the target area through simulation and correction of the hydrological cycle model.

[0051] The generated future forecast sequence of core meteorological elements is used as the driving input and imported into a preset hydrological cycle model (the hydrological cycle model is coupled with the underlying surface attribute parameters of the target area, including land cover type, soil physicochemical properties, etc.). The hydrological cycle model simulates the complete process of precipitation interception, water infiltration, runoff formation and soil moisture storage in the target area in the future period, and obtains the preliminary prediction results of soil moisture in the target area. Combined with the topographic feature data of the target area (such as slope, aspect and elevation gradient) and historical soil moisture observation data, the preliminary prediction results of soil moisture are corrected and adjusted to eliminate the model simulation error and the distribution deviation caused by topographic factors, and finally obtain the future soil moisture prediction results that fit the actual surface conditions of the target area.

[0052] Step S104: Input the future prediction sequence of core meteorological elements and the future soil moisture prediction results into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence, and correct the preliminary evapotranspiration component sequence to obtain the adjusted evapotranspiration component sequence.

[0053] The obtained future forecast sequences of core meteorological elements and future soil moisture prediction results are used as joint driving inputs and imported into a two-layer evapotranspiration model (i.e., an improved Shuttleworth-Wallace model, including an upper vegetation canopy evapotranspiration module and a lower soil surface evaporation module). The model calculates the transpiration process of the upper vegetation canopy and the evaporation process of the lower soil surface, and outputs a preliminary evapotranspiration component sequence containing vegetation transpiration, soil evaporation, and total evapotranspiration. Historical evapotranspiration component data of the target area are selected for the same period, and the deviation statistical index between the preliminary evapotranspiration component sequence and the measured data is calculated. Based on this index, a correction factor is determined, and the preliminary evapotranspiration component sequence is corrected time-by-time to obtain an adjusted evapotranspiration component sequence that meets the preset accuracy requirements and conforms to the physical meaning of the actual evapotranspiration process.

[0054] The aforementioned method for predicting evapotranspiration components acquires four meteorological elements affecting evapotranspiration components in the target area: net radiation, air temperature, wind speed, and relative humidity. It then selects core meteorological elements by combining regional vegetation cover and soil moisture retention characteristics. A historical simulation sequence of the core meteorological elements is constructed, and the Pearson correlation coefficient is calculated by comparing it with the corresponding historical observation sequence. Based on this coefficient, an initial fusion prediction value is determined, and emergence constraints are applied to generate a future prediction sequence of the core meteorological elements. This prediction sequence is input into a hydrological cycle model for simulation and correction to obtain the future soil moisture prediction result for the target area. Finally, the future prediction sequence of the core meteorological elements and the future soil moisture prediction result are jointly input into a two-layer evapotranspiration model to output a preliminary evapotranspiration component sequence, which is then corrected to obtain the final adjusted evapotranspiration component sequence. This method achieves accurate prediction of evapotranspiration components through multi-stage coupling and layer-by-layer optimization: First, it screens core meteorological elements, eliminates redundant information, and improves the targeting of subsequent stages; second, it uses Pearson correlation coefficient and emergence constraints to reduce the uncertainty of meteorological forecasts and enhance the reliability of sequences; third, it uses hydrological cycle model simulation and correction to make soil moisture predictions fit the actual regional conditions and provide accurate water limiting factors; fourth, it jointly drives a two-layer evapotranspiration model and corrects sequences, coupling comprehensive influences, making up for the shortcomings of traditional models, improving prediction accuracy and stability, and providing reliable support for scenarios such as water resource allocation, agricultural irrigation optimization, and ecological environment assessment.

[0055] In one embodiment, calculating the Pearson correlation coefficient between the historical simulation sequence of core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion prediction value, and applying a preset emergence constraint to the initial fusion prediction value to generate the future prediction sequence of core meteorological elements, may include the following steps:

[0056] Step S201: Select multiple climate models from the CMIP6 climate model set, and obtain the historical simulation sequences output by each selected climate model and the historical observation sequences of the core meteorological elements of the target area.

[0057] Preferably, the CMIP6 climate model set, namely the Coupled Model Intercomparison Project Phase 6 model set, is an international collaborative climate simulation dataset led by the World Climate Research Programme (WCRP). Its core function is to provide data support for climate change simulation and future projections, and it directly serves the IPCC Sixth Assessment Report (AR6).

[0058] From the CMIP6 climate model set, several climate models with climate simulation capabilities for the target area and verified data reliability were selected (selection criteria included the model's simulation accuracy of the target area's climate system, the completeness of the data time series, and the generally accepted reliability evaluation results in the field). Historical simulation sequences corresponding to the core meteorological elements (net radiation, temperature, wind speed, and relative humidity) of the target area were extracted from the output of each selected climate model (the time range must be consistent with the observation data of the target area). At the same time, historical observation sequences of the core meteorological elements of the target area during the same period were collected (data sources include measured data from regional meteorological stations, remote sensing inversion correction data, etc.) to ensure that the time scale (e.g., daily scale, monthly scale) and spatial coverage of the two types of sequences are completely matched.

[0059] Step S202: Calculate the Pearson correlation coefficient between the historical simulation sequence and the historical observation sequence for each selected climate model.

[0060] For each core meteorological element, calculate the Pearson correlation coefficient between the historical simulation sequence corresponding to each selected climate model and the historical observation sequence of that element in the target area. During the calculation, ensure that the sample size of the two types of sequences is consistent and there are no missing values ​​(missing data are supplemented by linear interpolation, and the supplemented data accounts for no more than 5% of the total sample size). The Pearson correlation coefficient ranges from [-1, 1]. This coefficient is used to quantify the degree of linear correlation between the simulation sequence and the observation sequence. The closer the coefficient is to 1, the better the historical simulation effect of the corresponding climate model on the core meteorological element.

[0061] Step S203: Based on the Pearson correlation coefficient, determine the weight coefficients corresponding to each selected climate model. After normalizing the weight coefficients, perform a weighted sum with each core meteorological element to obtain the initial fusion prediction value of the core meteorological elements.

[0062] Step S204: Apply preset emergence constraints to the initial fusion prediction values ​​to obtain the core meteorological element fusion prediction values ​​after constraint adjustment; the emergence constraints are preset based on the physical laws of the climate system in the target area and the relevant constraint requirements for evapotranspiration component prediction.

[0063] Step S205: Generate future forecast sequences of core meteorological elements based on the fused forecast values ​​of core meteorological elements.

[0064] Step S206: Verify whether the future prediction sequence meets the preset physical consistency threshold. If it does not meet the threshold, readjust the weight coefficients of each selected climate model based on the verification results to generate a future prediction sequence of core meteorological elements that meets the physical consistency threshold. The physical consistency threshold is preset based on the physical attributes of the meteorological elements themselves and the regional climate characteristics.

[0065] Based on the physical properties of meteorological elements (such as non-negative net radiation and non-negative wind speed) and the climate characteristics of the target area (such as the seasonal variation patterns of precipitation and temperature), a physical consistency threshold is pre-set (such as the daily temperature range not exceeding 30℃ and the wind speed not exceeding 1.2 times the historical maximum wind speed of the area). A time-by-time verification method is used to verify whether the future prediction sequence of core meteorological elements generated in step 5 meets the physical consistency threshold. If there are time periods in the sequence that do not meet the threshold, the climate model corresponding to the abnormal data is located based on the verification results, and the weight coefficient of the model is readjusted (the weight of the abnormal contribution model is reduced and the weight of the stable model is increased). The fusion, constraint and sequence generation process of steps 203-205 is executed again until the future prediction sequence of core meteorological elements that meets the physical consistency threshold is obtained, so as to ensure the reliability and applicability of the sequence.

[0066] Specifically, multiple climate models with climate simulation capabilities for the target region are selected from the CMIP6 climate model set. Historical simulation sequences of core meteorological elements output by each selected climate model are extracted, and historical observation sequences of core meteorological elements for the corresponding time periods in the target region are collected. For each selected climate model, the Pearson correlation coefficient between its historical simulation sequence and historical observation sequence is calculated. Based on this coefficient, the weighting coefficients corresponding to each model are determined. After normalizing all weighting coefficients, a weighted summation operation is performed with the corresponding data of each core meteorological element to obtain the initial fusion prediction values ​​of the core meteorological elements. Based on the physical laws of the climate system in the target region and the evapotranspiration composition, the predictions are then calculated. The relevant constraints require the pre-set emergence constraints, which are then applied to the initial fusion prediction values ​​to obtain the constraint-adjusted fusion prediction values ​​of the core meteorological elements. Based on the constraint-adjusted fusion prediction values, combined with the pre-set future climate scenarios and time steps, a future prediction sequence of the core meteorological elements is generated. A physical consistency threshold is pre-set based on the physical properties of the meteorological elements themselves and the regional climate characteristics. The future prediction sequence is then verified time-by-time. If the sequence does not meet the threshold, the climate model corresponding to the abnormal data is located based on the verification results, its weight coefficients are readjusted, and the fusion, constraint, and sequence generation process is executed again until a future prediction sequence of the core meteorological elements that meets the physical consistency threshold is generated.

[0067] This embodiment achieves low uncertainty and high reliability in the future prediction of core meteorological elements through multi-model fusion, quantitative weighting, and multi-level constraint verification. Firstly, it constructs a prediction foundation based on the CMIP6 multi-climate model, combining it with Pearson correlation coefficient weighted fusion to fully leverage the simulation advantages of different models and avoid biases caused by single models or simple averaging, thus initially reducing prediction uncertainty. Secondly, the application of emergent constraints ensures that the fused prediction values ​​conform to the physical laws of the regional climate system and the prediction requirements of evapotranspiration components, eliminating unreasonable data. Thirdly, physical consistency threshold verification and weight iterative adjustment further eliminate abnormal information in the sequence, ensuring the spatiotemporal continuity and physical rationality of the future prediction sequence. The final output of the core meteorological element future prediction sequence has higher accuracy and stronger reliability, providing high-quality input data for subsequent hydrological cycle model driving, soil moisture prediction, and evapotranspiration component simulation, laying a core foundation for improving the credibility of the overall evapotranspiration component prediction results.

[0068] In one embodiment, the future soil moisture prediction result for the target area is obtained by inputting a preset hydrological cycle model based on the future prediction sequence of core meteorological elements, and simulating and correcting the hydrological cycle model. This may include the following steps:

[0069] Step S301: Input the future prediction sequence of core meteorological elements into the hydrological cycle model coupled with the surface feature parameters of the target area for simulation to obtain the preliminary prediction results of soil moisture in the target area; the surface feature parameters include surface cover parameters, soil physicochemical parameters and hydrological correlation parameters.

[0070] Step S302: Obtain key terrain feature data including slope, aspect, and elevation gradient of the target area; perform standardized preprocessing on the key terrain feature data to obtain terrain input data.

[0071] Step S303: Based on the preliminary prediction results and terrain input data, the soil moisture distribution correction value is obtained by quantitative calculation through the terrain-soil moisture response relationship model.

[0072] Step S304: Verify whether the soil moisture distribution correction value exceeds the preset reasonable threshold. If it does, dynamically correct and adjust the preliminary prediction result according to the degree of deviation between the soil moisture distribution correction value and the reasonable threshold to obtain the future soil moisture prediction result.

[0073] First, four meteorological elements affecting evapotranspiration components in the target area were acquired: net radiation, air temperature, wind speed, and relative humidity. Combined with vegetation cover data (vegetation type, quantitative indicators of cover density) and soil moisture retention characteristics data (soil texture, porosity, etc.), core meteorological elements significantly influencing vegetation transpiration and soil evaporation were identified, and redundant data were eliminated. Then, several climate models with climate simulation capabilities and reliable data for the target area were selected from the CMIP6 climate model set. Historical simulation sequences of core meteorological elements output by each model were extracted. Simultaneously, historical sequencing data of core meteorological elements for corresponding time periods in the target area were collected. The data (including measured and remote sensing inversion correction data from meteorological stations) are used to calculate the Pearson correlation coefficient between the historical simulation sequence and the observed sequence for each model. This coefficient is then used to determine the weighting coefficients for each model, which are normalized. The initial fusion prediction value is obtained by weighted summation with the data of each core meteorological element. Based on the physical laws of the climate system in the target area (such as the spatiotemporal continuity of meteorological elements and the logical correlation between elements) and the prediction constraints of evapotranspiration components, emergent constraints (including extreme value thresholds and spatiotemporal consistency constraints) are preset and applied to the initial fusion prediction value to obtain the constraint-adjusted fusion prediction value. This is then combined with preset future climate scenarios (such as CMIP6). The system generates future prediction sequences of core meteorological elements using SSP1-2.6, SSP2-4.5, etc., and time steps (day / month / season). Based on the physical attributes of meteorological elements and regional climate characteristics, a physical consistency threshold is preset (e.g., daily temperature range not exceeding 30℃, wind speed not exceeding 1.2 times the historical maximum). The future prediction sequences are verified time-by-time. If the threshold is not met, abnormal patterns are located and weights are adjusted. The fusion, constraint, and sequence generation process is iteratively executed until a qualified future prediction sequence of core meteorological elements is obtained.

[0074] Subsequently, the future prediction sequence is input into a hydrological cycle model coupled with surface feature parameters of the target area (surface parameters include land cover, soil physicochemical, and hydrological correlation parameters) to simulate the complete process of precipitation interception, water infiltration, runoff formation, and soil moisture storage in the target area, obtaining preliminary soil moisture prediction results. Key topographic feature data of slope, aspect, and elevation gradient of the target area are obtained and standardized preprocessed (uniformed to the [0,1] interval to eliminate scale differences) to obtain topographic input data. The preliminary prediction results and topographic input data are substituted into the topographic-soil moisture response relationship model to quantitatively calculate the soil moisture distribution correction value. A reasonable soil moisture threshold is preset (based on the region's historical observation extreme values ​​and soil physical properties), and the correction value is checked to see if it exceeds the threshold. If it does, the preliminary prediction results are dynamically corrected according to the degree of deviation, and finally, the future soil moisture prediction results of the target area are obtained.

[0075] Finally, the future forecast sequences of core meteorological elements and the future soil moisture forecast results are used as joint driving inputs to import the improved Shuttleworth-Wallace model (a two-layer evapotranspiration model, including an upper vegetation canopy evapotranspiration module and a lower soil surface evaporation module). The model calculates the transpiration process of the upper vegetation and the evaporation process of the lower soil, and outputs a preliminary evapotranspiration component sequence including vegetation transpiration, soil evaporation, and total evapotranspiration. The measured historical evapotranspiration component data of the target area are selected for the same period, and the deviation statistics of the preliminary sequence and the measured data (such as the coefficient of determination and root mean square error) are calculated. Based on the index, the correction factor is determined, and the preliminary sequence is corrected time-by-time to finally obtain the adjusted evapotranspiration component sequence.

[0076] This embodiment specifically addresses the shortcomings of existing technologies, such as high uncertainty in meteorological element prediction, neglect of surface and topographic influences in soil moisture prediction, and low accuracy due to insufficient input quality of evapotranspiration models. Through end-to-end optimization, it significantly improves the reliability of evapotranspiration component prediction. Firstly, the core meteorological element screening process ensures the relevance of input data. By combining CMIP6 multi-climate model weighted fusion, emergence constraints, and iterative verification of physical consistency, it fully leverages the simulation advantages of different models, effectively reducing the uncertainty in meteorological predictions caused by single models or simple averaging. The output core meteorological element future prediction sequences have higher accuracy and stronger physical rationality. Secondly, the hydrological cycle model couples with underlying surface characteristic parameters of the target area. Combined with topographic data standardization and response relationship model correction, it accurately characterizes the impact of surface heterogeneity and topography on soil moisture distribution, avoiding the generalization bias of traditional soil moisture prediction and providing high-quality moisture constraint factor input for evapotranspiration component prediction. Thirdly, the improved Shuttleworth-Wallace model, driven by the combined results of core meteorological and soil moisture predictions, compensates for the shortcomings of traditional models that neglect interlayer energy interaction and dynamic feedback. Combined with bias correction of the preliminary sequence, it fully couples the comprehensive influence of meteorology and soil moisture on evapotranspiration components.

[0077] In one embodiment, the soil moisture distribution correction value can be calculated using the following formula:

[0078]

[0079] in, This indicates the correction value for soil moisture distribution. This indicates the preliminary prediction results of soil moisture. Indicates the terrain-underlying surface adaptation coefficient. , , , This represents the weighting coefficient, which is set based on the dominant type of the underlying surface in the region. , Indicates the average vegetation cover of the region. , Indicates soil porosity. , This indicates the percentage of impermeable area. The topographic sensitivity coefficient is calculated based on nearly 10 years of topographic-soil moisture observation data for the target area. Indicates the standardized terrain composite index, , , , This represents the weighting coefficient, which is set based on the dominant factors of regional topography. , Indicates slope, Indicates the maximum slope of the area. , Indicates slope direction. , Indicates altitude, This indicates the maximum and minimum elevation values ​​of a region.

[0080] This embodiment constructs a quantitative formula for soil moisture distribution correction values ​​that integrates underlying surface features and topographic conditions. It uses a weighted coupling of underlying surface parameters related to land cover, soil physicochemical properties, and hydrology to form a topographic-underlying surface fit coefficient k. Slope, aspect, and elevation are standardized and combined with the weights of regional topographic dominant factors to obtain a standardized topographic comprehensive index T. Then, a topographic sensitivity coefficient α based on nearly 10 years of measured data from the region is used to establish the response relationship between these two factors and the preliminary soil moisture prediction results. This achieves accurate quantification of the impact of topography and underlying surface on soil moisture distribution, effectively compensating for the shortcomings of traditional soil moisture prediction that ignores surface heterogeneity and topographic coupling effects. The corrected future soil moisture prediction results better match the actual spatial distribution characteristics of the target area, providing a high-precision, highly adaptable water constraint factor input for subsequent improvements to the Shuttleworth-Wallace model, further solidifying the core foundation for improving the accuracy and reliability of evapotranspiration component prediction.

[0081] In one embodiment, inputting the future forecast sequence of core meteorological elements and the future soil moisture forecast result into a two-layer evapotranspiration model outputs a preliminary evapotranspiration component sequence, and correcting the preliminary evapotranspiration component sequence to obtain an adjusted evapotranspiration component sequence may include the following steps:

[0082] Step S401: Input the future prediction sequence of core meteorological elements and the future soil moisture prediction results into the two-layer evapotranspiration model, and output the initial energy balance parameters of the upper vegetation canopy and the lower soil surface layer; the energy balance parameters include the radiation balance coefficient, the sensible heat flux coefficient, and the initial threshold of the latent heat flux; the two-layer evapotranspiration model is an improved Shuttleworth-Wallace model.

[0083] Preferably, the future forecast sequence of core meteorological elements (including time-series data of net radiation, temperature, wind speed, and relative humidity) and the future soil moisture forecast results (including quantitative data of spatiotemporal distribution of the target area) are used as joint driving inputs and imported into an improved Shuttleworth-Wallace model (two-layer evapotranspiration model). The model simulates the energy exchange process between the upper vegetation canopy and the lower soil surface layer, and outputs initial energy balance parameters between the upper vegetation canopy and the lower soil surface layer by combining the basic characteristics of the target area such as vegetation type and soil texture. The energy balance parameters include radiation balance coefficient (characterizing the absorption and conversion efficiency of solar radiation by each layer), sensible heat flux coefficient (quantifying the heat transfer intensity between each layer and the atmosphere), and initial latent heat flux threshold (setting the benchmark value of evapotranspiration energy consumption for each layer).

[0084] The improved Shuttleworth-Wallace model is based on the traditional two-layer evapotranspiration model. Its core improvement lies in breaking through the limitations of the traditional model, which relies on a single downscaled data or a simple multi-model average as a coarse input. It uses the low-uncertainty future forecast sequence of core meteorological elements generated by weighted fusion of multiple climate models of CMIP6, emergence constraints, and physical consistency verification, along with the high-precision future soil moisture forecast results of surface feature parameters and topographic correction of the target area, as joint driving inputs. At the same time, it retains the original two-layer energy balance framework of the model and optimizes the dynamic response logic of interlayer interaction to input data.

[0085] Step S402: Based on the energy balance parameters, the Newton-Raphson iterative solution method is used to iteratively calculate the energy balance equations of the upper and lower layers to obtain a preliminary evapotranspiration component sequence including time series data of vegetation transpiration, soil evaporation and total evapotranspiration; the energy balance equations include the balance relationships of radiation balance, sensible heat flux and latent heat flux.

[0086] Preferably, based on the output energy balance parameters, the Newton-Raphson iterative solution method is used to iteratively calculate the upper and lower layer energy balance equations, which include the balance relationships of radiation balance, sensible heat flux, and latent heat flux. The upper layer energy balance equation focuses on the balance of radiation absorption, sensible heat transfer, and evaporative latent heat consumption of the vegetation canopy, while the lower layer energy balance equation focuses on the balance of radiation absorption, sensible heat transfer, and evaporative latent heat consumption of the soil surface. During the iteration process, the calculated values ​​of sensible heat flux and latent heat flux are gradually corrected until the equations converge (the convergence condition is that the difference between two adjacent iterations is less than a preset threshold). Finally, time-series data containing vegetation transpiration, soil evaporation, and total evapotranspiration are obtained, forming a preliminary evapotranspiration component sequence. The time step of the sequence is consistent with the future prediction sequence of the core meteorological elements.

[0087] Step S403: Select the measured historical evapotranspiration composition data of the target area for the same period, compare the preliminary evapotranspiration composition sequence with the measured data time by time, and calculate the prediction accuracy index including the coefficient of determination and root mean square error.

[0088] Step S404: If the prediction accuracy index is lower than the preset accuracy threshold, an uncertainty adjustment factor is obtained based on the error characteristics of the future prediction sequence of core meteorological elements and the future soil moisture prediction results; the accuracy threshold is preset based on the regional evapotranspiration prediction application requirements.

[0089] Furthermore, based on the practical application needs of regional evapotranspiration forecasting (such as the requirements for forecast accuracy in scenarios like agricultural irrigation optimization and water resource allocation), a pre-set accuracy threshold (including the minimum threshold for the coefficient of determination and the maximum threshold for the root mean square error) is established. The calculated forecast accuracy index is compared with this threshold. If the index is lower than the accuracy threshold (i.e., the coefficient of determination is less than the minimum threshold or the root mean square error is greater than the maximum threshold), the error characteristics of the future forecast sequences of core meteorological elements and the future soil moisture forecast results are further analyzed, including the spatiotemporal deviation distribution of meteorological element forecasts and the spatial heterogeneity error of soil moisture forecasts. By quantifying the degree of influence of these errors on the forecast of evapotranspiration components, a targeted uncertainty adjustment factor is obtained. The magnitude of the adjustment factor is positively correlated with the degree of error influence.

[0090] Step S405: The uncertainty adjustment factor is incorporated into the time-by-time data of the preliminary evapotranspiration component sequence using a weighted correction method to dynamically correct the evapotranspiration component sequence, thereby obtaining the adjusted evapotranspiration component sequence.

[0091] Specifically, the future forecast sequences of core meteorological elements and the future soil moisture forecast results are used as joint driving inputs and imported into an improved Shuttleworth-Wallace model (two-layer evapotranspiration model). The model outputs initial energy balance parameters for the upper vegetation canopy and the lower soil surface layer. These energy balance parameters include the radiation balance coefficient, sensible heat flux coefficient, and initial threshold for latent heat flux. Based on these energy balance parameters, the Newton-Raphson iterative solution method is used to iteratively calculate the energy balance equations for the upper and lower layers, which include the balance relationships of radiation balance, sensible heat flux, and latent heat flux, to obtain time-series data on vegetation transpiration, soil evaporation, and total evapotranspiration. Preliminary evapotranspiration component sequence; Select measured evapotranspiration component data of the target area during the same period, and compare the preliminary evapotranspiration component sequence with the measured data time by time to calculate the prediction accuracy index, including the coefficient of determination and root mean square error; Based on the application requirements of regional evapotranspiration prediction, a precision threshold is pre-set. If the prediction accuracy index is lower than the threshold, an uncertainty adjustment factor is quantified based on the error characteristics of the future prediction sequence of core meteorological elements and the future soil moisture prediction results; Using a weighted correction method, the uncertainty adjustment factor is incorporated into the time-by-time data of the preliminary evapotranspiration component sequence for dynamic correction, and finally the adjusted evapotranspiration component sequence is obtained.

[0092] This embodiment achieves high accuracy and reliability in evapotranspiration component prediction through model optimization, iterative solving, and accuracy feedback correction. Firstly, it uses an improved Shuttleworth-Wallace model as its core, combined with core meteorological and soil moisture prediction data to output energy balance parameters that closely reflect the actual regional conditions, providing a high-quality foundation for subsequent iterative calculations. Secondly, the application of the Newton-Raphson iterative solution method ensures the convergence of the upper and lower layer energy balance equations, making the initial evapotranspiration component sequence conform to the physical laws of energy balance. Thirdly, by comparing the calculation accuracy indicators with historical measured data, prediction biases are accurately identified, and an uncertainty adjustment factor determined based on the error characteristics of the input data provides a quantitative basis for correction. Fourthly, a weighted dynamic correction mechanism specifically compensates for the biases in the initial sequence, effectively improving the accuracy and stability of evapotranspiration component prediction. The final adjusted evapotranspiration component sequence accurately reflects the temporal changes in vegetation transpiration and soil evaporation, providing reliable data support for scenarios such as forest-water relationship evaluation, vegetation restoration measure formulation, and water resource optimization scheduling.

[0093] In one embodiment, the preliminary evaporation component sequence can be represented by the following formula:

[0094]

[0095] in, Indicates the first Upper-layer transpiration in the next iteration Indicates the first Evaporation rate in the next iteration of the lower layer Indicates the first Total evapotranspiration in each iteration , Indicates the radiation balance coefficient between the upper and lower layers. , , , Indicates the sensible heat flux coefficients of the upper and lower layers. , , , This indicates the net radiation between the upper and lower levels. , Indicates soil moisture limiting factor, , Indicates field holding capacity. Indicates the moisture content during wilting. This represents the predicted future soil moisture value for the target area. This represents the interlayer energy coupling coefficient. , This represents the reference evapotranspiration, calculated based on the core meteorological forecast sequence. Represents the iterative correction coefficient. , Indicates the number of iterations. The iteration termination condition is .

[0096] This embodiment constructs a quantitative iterative formula that integrates the energy balance laws between upper and lower layers, dynamic constraints on soil moisture, and interlayer interaction effects. This formula directly correlates the radiation balance coefficients and sensible heat flux coefficients of the upper and lower layers with the net radiation of each layer, thus affecting energy distribution. It also utilizes soil moisture limiting factors. (Based on future soil moisture forecasts and quantification of field capacity and wilting water content) it dynamically reflects the constraint of water on transpiration, uses the interlayer energy coupling coefficient η to represent the energy interaction between upper and lower layers, and combines iterative correction coefficients. The termination condition ensures computational convergence, while also referencing evaporation. By associating core meteorological forecast sequences, the system quantifies the evolution of upper-layer transpiration, lower-layer evaporation, and total evapotranspiration during the iteration process. This effectively compensates for the shortcomings of traditional models, such as simplified interlayer interactions and static water constraints. The resulting preliminary evapotranspiration component sequence not only conforms to the physical laws of energy balance but also accurately responds to the combined effects of meteorology, soil moisture, and vegetation cover. This significantly improves the physical consistency and prediction accuracy of the preliminary sequence, providing a high-quality foundation for subsequent dynamic correction based on historical measured data.

[0097] In one embodiment, such as Figure 2As shown, this application also provides an evapotranspiration component prediction system, which may include the following steps:

[0098] The meteorological element screening module 501 is used to obtain four types of meteorological elements that affect the evapotranspiration components of the target area: net radiation, air temperature, wind speed, and relative humidity. Core meteorological elements are selected from the four types of meteorological elements based on the regional vegetation coverage and soil moisture retention characteristics.

[0099] The future weather forecast module 502 is used to calculate the Pearson correlation coefficient between the historical simulation sequence of core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion forecast value, and to apply a preset emergence constraint to the initial fusion forecast value to generate the future forecast sequence of core meteorological elements.

[0100] The soil moisture prediction module 503 is used to input a preset hydrological cycle model based on the future prediction sequence of core meteorological elements, and obtain the future soil moisture prediction results of the target area through simulation and correction of the hydrological cycle model.

[0101] The evapotranspiration adjustment module 504 is used to input the future prediction sequence of core meteorological elements and the future soil moisture prediction results into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence, and to correct the preliminary evapotranspiration component sequence to obtain the adjusted evapotranspiration component sequence.

[0102] The aforementioned evapotranspiration component prediction system employs a meteorological element screening module to acquire four meteorological elements affecting evapotranspiration components in the target area: net radiation, air temperature, wind speed, and relative humidity. Combining this with vegetation cover data and soil moisture retention characteristics data of the target area, the system analyzes the correlation strength between these four meteorological elements and vegetation transpiration and soil evaporation to screen out core meteorological elements. The future weather prediction module then constructs a historical simulation sequence based on these core meteorological elements. This sequence is compared with the historical observation sequences of the corresponding core meteorological elements in the target area, and the Pearson correlation coefficient is calculated. Based on this correlation coefficient, an initial fusion prediction value for the core meteorological elements is determined, and a preset surge is then applied to the initial fusion prediction value. Under the given constraints, a future forecast sequence of core meteorological elements is generated. The soil moisture prediction module inputs the future forecast sequence of core meteorological elements into a preset hydrological cycle model. By simulating the water cycle process in the target area through the model, a preliminary soil moisture prediction result is obtained. After correcting and adjusting the preliminary prediction result, the future soil moisture prediction result of the target area is obtained. The evapotranspiration adjustment module takes the future forecast sequence of core meteorological elements and the future soil moisture prediction result as joint driving inputs and imports them into a two-layer evapotranspiration model. The model calculates and outputs a preliminary evapotranspiration component sequence containing vegetation transpiration, soil evaporation, and total evapotranspiration. The preliminary evapotranspiration component sequence is corrected and processed to finally obtain the adjusted evapotranspiration component sequence.

[0103] This embodiment achieves accurate prediction of evapotranspiration components through multi-module collaborative linkage and end-to-end data optimization. First, the meteorological element screening module eliminates redundant data, ensuring the correlation between core meteorological elements and evapotranspiration components, providing targeted input for subsequent prediction stages. Second, the future meteorological prediction module effectively reduces the uncertainty of future predictions of core meteorological elements and improves sequence reliability by leveraging Pearson correlation coefficient fusion and emergence constraints. Third, the soil moisture prediction module simulates and adjusts using a hydrological cycle model, ensuring that future soil moisture prediction results align with regional realities, providing accurate moisture limiting factors for evapotranspiration component prediction. Fourth, the evapotranspiration adjustment module, through joint driving of a two-layer evapotranspiration model and preliminary sequence correction, fully couples the comprehensive influence of meteorology and soil moisture, compensating for the shortcomings of traditional models, ultimately improving the accuracy and stability of evapotranspiration component prediction, and providing reliable data support for scenarios such as forest-water relationship evaluation, vegetation restoration measure formulation, and water resource allocation.

[0104] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0105] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the evapotranspiration component prediction method as described above.

[0106] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0107] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0108] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for predicting evapotranspiration components, characterized in that, The method includes: Four meteorological elements affecting evapotranspiration components in the target area were obtained: net radiation, air temperature, wind speed and relative humidity. Core meteorological elements were selected from these four meteorological elements based on the regional vegetation cover and soil moisture retention characteristics. The Pearson correlation coefficient between the historical simulation sequence of the core meteorological elements and the corresponding historical observation sequence of the target area is calculated to determine the initial fusion prediction value. A preset emergence constraint is applied to the initial fusion prediction value to generate the future prediction sequence of the core meteorological elements. Based on the future prediction sequence of the core meteorological elements, a preset hydrological cycle model is input, and the future soil moisture prediction result of the target area is obtained through simulation and correction of the hydrological cycle model. The future prediction sequence of the core meteorological elements and the future soil moisture prediction results are input into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence. The preliminary evapotranspiration component sequence is then corrected to obtain the adjusted evapotranspiration component sequence.

2. The method according to claim 1, characterized in that, The process of calculating the Pearson correlation coefficient between the historical simulation sequence of the core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion prediction value, and applying a preset emergence constraint to the initial fusion prediction value to generate the future prediction sequence of the core meteorological elements includes: Multiple climate models were selected from the CMIP6 climate model set, and the historical simulation sequences output by each selected climate model and the historical observation sequences of the core meteorological elements in the target area were obtained. Calculate the Pearson correlation coefficient between the historical simulation sequence and the historical observation sequence for each of the selected climate models; Based on the Pearson correlation coefficient, the weight coefficients corresponding to each of the selected climate models are determined. After normalizing the weight coefficients, they are weighted and summed with each of the core meteorological elements to obtain the initial fusion prediction values ​​of the core meteorological elements. Pre-defined emergence constraints are applied to the initial fusion prediction values ​​to obtain the constrained fusion prediction values ​​of the core meteorological elements; the emergence constraints are pre-set based on the physical laws of the climate system in the target area and the relevant constraints for evapotranspiration component prediction. Generate a future forecast sequence of core meteorological elements based on the fused forecast values ​​of the core meteorological elements; The future prediction sequence is verified to meet the preset physical consistency threshold. If it does not meet the threshold, the weight coefficients of each selected climate model are readjusted based on the verification result to generate a future prediction sequence of core meteorological elements that meets the physical consistency threshold. The physical consistency threshold is preset based on the physical attributes of the meteorological elements themselves and the regional climate characteristics.

3. The method according to claim 1, characterized in that, The method of inputting the future prediction sequence of the core meteorological elements into a preset hydrological cycle model, and obtaining the future soil moisture prediction result of the target area through simulation and correction by the hydrological cycle model, includes: The future prediction sequence of the core meteorological elements is input into a hydrological cycle model coupled with the surface feature parameters of the target area for simulation to obtain preliminary prediction results of soil moisture in the target area; the surface feature parameters include surface cover parameters, soil physicochemical parameters, and hydrological correlation parameters; Acquire key terrain feature data, including slope, aspect, and elevation gradient of the target area, and perform standardized preprocessing on the key terrain feature data to obtain terrain input data; Based on the preliminary prediction results and the terrain input data, the soil moisture distribution correction value is obtained by quantitative calculation using the terrain-soil moisture response relationship model. The soil moisture distribution correction value is checked to see if it exceeds a preset reasonable threshold. If it does, the preliminary prediction result is dynamically corrected and adjusted according to the degree of deviation between the soil moisture distribution correction value and the reasonable threshold to obtain the future soil moisture prediction result.

4. The method according to claim 3, characterized in that, The soil moisture distribution correction value is calculated using the following formula: in, This indicates the correction value for soil moisture distribution. This indicates the preliminary prediction results of soil moisture. Indicates the terrain-underlying surface fit coefficient. , , , This represents the weighting coefficient, which is set based on the dominant type of the underlying surface in the region. , Indicates the average vegetation cover of the region. , Indicates soil porosity. , This indicates the percentage of impermeable area. The topographic sensitivity coefficient is calculated based on nearly 10 years of topographic-soil moisture observation data for the target area. Indicates the standardized terrain composite index, , , , This represents the weighting coefficient, which is set based on the dominant factors of regional topography. , Indicates slope, Indicates the maximum slope of the area. , Indicates slope direction. , Indicates altitude, This indicates the maximum and minimum elevation values ​​of a region.

5. The method according to claim 1, characterized in that, The process involves inputting the future prediction sequence of the core meteorological elements and the future soil moisture prediction results into a two-layer evapotranspiration model to output a preliminary evapotranspiration component sequence, and then correcting the preliminary evapotranspiration component sequence to obtain an adjusted evapotranspiration component sequence, including: The predicted future sequences of the core meteorological elements and the predicted future soil moisture are input into the two-layer evapotranspiration model, which outputs the initial energy balance parameters of the upper vegetation canopy and the lower soil surface layer. The energy balance parameters include the radiation balance coefficient, the sensible heat flux coefficient, and the initial threshold of the latent heat flux. The two-layer evapotranspiration model is an improved Shuttleworth-Wallace model. Based on the energy balance parameters, the Newton-Raphson iterative solution method is used to iteratively calculate the energy balance equations of the upper and lower layers, resulting in a preliminary evapotranspiration component sequence including time-series data of vegetation transpiration, soil evaporation, and total evapotranspiration; the energy balance equations include the balance relationships of radiation balance, sensible heat flux, and latent heat flux. Select the measured historical evapotranspiration composition data of the target area during the same period, compare the preliminary evapotranspiration composition sequence with the measured data time by time, and calculate the prediction accuracy index including the coefficient of determination and root mean square error. If the prediction accuracy index is lower than the preset accuracy threshold, an uncertainty adjustment factor is obtained based on the error characteristics of the future prediction sequence of the core meteorological elements and the future soil moisture prediction results; the accuracy threshold is preset based on the regional evapotranspiration prediction application requirements. The uncertainty adjustment factor is incorporated into the time-by-time data of the preliminary evapotranspiration component sequence using a weighted correction method to dynamically correct the sequence, resulting in the adjusted evapotranspiration component sequence.

6. The method according to claim 5, characterized in that, The preliminary evaporation component sequence is represented by the following formula: in, Indicates the first Upper-layer transpiration in the next iteration Indicates the first Evaporation rate in the next iteration of the lower layer Indicates the first Total evapotranspiration in each iteration , Indicates the radiation balance coefficient between the upper and lower layers. , , , Indicates the sensible heat flux coefficients of the upper and lower layers. , , , This indicates the net radiation between the upper and lower levels. , Indicates soil moisture limiting factor, , Indicates field holding capacity. Indicates the moisture content during wilting. This represents the predicted future soil moisture value for the target area. Indicates the interlayer energy coupling coefficient. , This represents the reference evapotranspiration, calculated based on the core meteorological forecast sequence. Represents the iterative correction coefficient. , Indicates the number of iterations. The iteration termination condition is .

7. A system for predicting evapotranspiration components, characterized in that, The system includes: The meteorological element screening module is used to obtain four types of meteorological elements that affect the evapotranspiration components of the target area: net radiation, air temperature, wind speed and relative humidity. Core meteorological elements are screened from the four types of meteorological elements based on the regional vegetation coverage and soil moisture retention characteristics. The future weather forecast module is used to calculate the Pearson correlation coefficient between the historical simulation sequence of the core meteorological elements and the corresponding historical observation sequence of the target area to determine the initial fusion forecast value, and to apply a preset emergence constraint to the initial fusion forecast value to generate the future forecast sequence of the core meteorological elements. The soil moisture prediction module is used to input a preset hydrological cycle model based on the future prediction sequence of the core meteorological elements, and obtain the future soil moisture prediction result of the target area through simulation and correction of the hydrological cycle model. The evapotranspiration adjustment module is used to input the future prediction sequence of the core meteorological elements and the future soil moisture prediction results into the two-layer evapotranspiration model to output the preliminary evapotranspiration component sequence, and to correct the preliminary evapotranspiration component sequence to obtain the adjusted evapotranspiration component sequence.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images